​​Project Title ​Android application “Therapist for students”
​Group Members ​Zain Saleem​
​​Project Supervisor(s) ​​Asst. Prof. Wasi Haider Butt
​Project Abstract ​Do you struggle with depression and have you considered using any apps to manage your feelings and mood but couldnotfind any, satisfying your needs well Try Therapist. Although a lot of work has been done in this domain and u will find many android/apple applications that are especially designed for people struggling with everyday depression but therapist not only provides you to different chat rooms where u can discuss your problems with the people who are suffering from the same syndrome but can also encounter the problems by following 7 different steps that lead to immediate relief for the person these steps enhance your mood by asking you to perform and watch different things and asking you to state the things u think you are thankful for in your life .in chat rooms you can upload status to tell people how you feel just like with self-help books, you may gain greater insight into your patterns of thinking and behavior which could boost awareness of how your mental scripts impact on your relationships, as well as increasing your levels of personal motivation throughout each day. asking greater responsibility for the management of your mood could help with your feelings of sadness or lack of enjoyment with pursuits which had previously given you pleasure. Additionally, if used in a healthcare setting, clinicians may be able to correlate data from app to track and monitor wider health issues and could potentially modify your medication to better treat your depression. A user first has to register himself and can also give the feedback of the app. Online data base will be used .android app will be connected to the online database to establish connection between user and server.
​​Project Title Traffic Monitoring System
​Group Members ​Muhammad Ali Khan   Moeed Sultan Kundi  Faizan Ijaz
​​Project Supervisor(s) ​ Dr Shoaib Ahmed Khan
​Project Abstract ​Today, the revolutionary computer-tech industry is doing wonders for mankind that are unprecedented. Computer engineers need to play their role in alleviating the most fatal stigma faced by humans: Traffic Accidents and Rush Hours. To extract the images exclusively of fast moving vehicles at a highway and applying background removal processing techniques, the computer program could help predict the future course of the vehicle and stymie the possibility of an accident to occur. Motion vectors is a very useful resource in this regard. A lot of work has been accomplished on tracking man-driven vehicles but our project would have the upswing of foretelling traffic jams and auguring the probability of a vehicle to encounter an accident by predicting its future course using motion vectors. Our algorithms would be based on Python and Open CV for tracking objects for their motion vectors. We would use a camera contraption to obtain a holistic geographic view of the highway to cover maximum possible routes and thus increase the efficacy of our algorithm. To test the durability of our hardware, we test our device on a number of different camera contraptions which produce the most efficient results along with testing it on different operating system. We use three frames and their absolute difference give us the detected vehicle. Stability techniques are used to mitigate the effect of vehicular headlights to make our project versatile to function at any time of the day.​
​​Project Title Incorporation of Leap Motion in a Real Time Strategy Simulation
​Group Members ​Arham Abdullah Malik,   Basharat Ali,   Subhan Tariq
​​Project Supervisor(s) ​​Dr Arslan Shaukat
​Project Abstract For now over a decade, real-time strategy (RTS) games have been challenging intelligence, human and artificial (AI) alike, as one of the top genre in terms of overall complexity. RTS is a prime example problem featuring multiple interacting imperfect decision makers. Elaborate dynamics, partial observability, as well as a rapidly diverging action space render rational decision making somehow elusive. Humans deal with the complexity using several abstraction layers, taking decisions on different abstract levels. Current agents, on the other hand, remain largely scripted and exhibit static behavior, leaving them extremely vulnerable to flaw abuse and no match against human players. In this paper, we propose to mimic the abstraction mechanisms used by human players for designing AI for RTS games. Furthermore, to differentiate from generic RTS games, we propose the incorporation of the hand gesture recognition from Leap Motion to act as the controller. The game itself will be created using already created software and tutorials available on the internet, while the inter-operability of the Leap Motion will be researched upon and implemented from said research. In short this project’s goal is to create a better decision making AI and to have the user’s hand gestures control the game.
​​Project Title ​Neurofeedback Therapy System for ADHD Treatment​
​Group Members ​Maryam Binte Zafar
​​Project Supervisor(s) ​​Dr Usman Akram
​Project Abstract ​Attention Deficit Hyperactivity Disorder (ADHD) is a neurological/psychological disorder that is characterized by slow brain activity. About 7 percent of school-age children worldwide are diagnosed with ADHD, according to an Australian study (2015). People diagnosed with ADHD find it difficult to perform cognitive tasks, pay attention and focus. Recently, Neurofeedback therapy has emerged as a complimentary treatment for ADHD, alongside drugs. It involves training the brain and tuning it to function more effectively through real-time monitoring of physiological activity of the brain while the user engages in a task requiring attention, concentration and cognition. The method most commonly used for real-time monitoring is electroencephalography (EEG); different features of the EEG signals are monitored which include SMR (Sensorimotor Rhythm) activity, Theta/Beta ratio and SCP (Slow Cortical Potential). This being a relatively new area of research, other features of EEG may lead to better monitoring. This project aims at developing an interactive application for brain training, designing a real-time signal acquisition protocol, designing real-time signal processing and analysis protocols and developing an interface to visualise brain activity at real-time. The project will employ the Emotiv EPOC headset to acquire EEG signals according to the 10 – 20 system. The acquired signals will be processed and the features will be extracted using the Emotiv SDK that runs with MATLAB. The analysis of thesefeatures will revealthe progress of the subject. The utility of this project is using neurofeedback therapy to compliment other ADHD treatments and improving concentration and attentiveness in otherwise healthy subjects. ​
​​Project Title ​Smart Home Loads Automation System
​Group Members Mishaal Safeer , M. Haider Kamal  , Foha Khalid
​​Project Supervisor(s) ​​Dr Usman AkramAli Saeed

Sajid Gul Khawaja

​Project Abstract ​Energy saving and automation has become the hype of the new era. The introduction of this in homes is being practiced around the world. However, we detected a need to design such a system that is easy and cheap to deploy that offers complete automation and control of home appliances. The aim of the project is to build a working prototype which will compromise of a switch that will communicate with a load to operate its functionality. All communication will be wireless, over Wifi. The system, once deployed, will offer more than a remote control of appliances. A combination of machine learning algorithms will be employed to learn the schedule of the user in order to emulate and optimize the usage of electricity in a given home. The algorithms will take into view the preferences and routine of the user, and will utilize other relevant data taken from online sources. Using this data, a schedule is prepared, whose main goal is to minimize both the electricity usage of the home and the total bill cost. With the permission of the user, this new schedule may be applied, and would control the timing and duration of when each device is turned on or off.
​​Project Title A System for detecting and regulating anxiety through Haptics
​Group Members ​Mariam Jafri

Maryam Ibrar

Zohaib Khalid

​​Project Supervisor(s) ​​Dr Usman Akram

Dr Sajid Gul Khawaja

​Project Abstract ​A person’s normal reaction to any stressful condition is basically anxiety where a person feels butterflies in his/her stomach or feels apprehensive or fearful. Anxiety symptoms include symptoms as common as headache, stomachache, vomiting, insomnia, restlessness, diarrhea, and as serious and rare as brain tumor. The normal breathing rate of a person is 16-20 breaths per minute, but an anxious person’s breathing rate reaches up to 35-40 breaths per minute. Given the prevalence and adverse effects of anxiety and anxiety disorders, our product helps people regulate their anxiety through slow-paced breathing. Slow-paced breathing reduces the perceived stress which results in regulating person’s emotions. A chest-band consisting of sensors and actuators paired with a mobile-app will collect the real-time data of a person. The user can personalize their experience by first training the device and the app according to his breathing pattern. If it detects an increase in breathing rate, it will automatically generate vibrotactile patterns that are going to be similar to a person’s breathing pattern already configured by the user. Since the goal of the product is to pace breathing inconspicuously, the person will start breathing at the pace of the vibrotactile patterns being generated by the device. Hence reducing the increased rate in breathing and indirectly calming the person and reducing stress.
​​Project Title Machine Learning Based Currency Exchange Rate prediction
​Group Members ​GC Qaisar khan
GC Rawal Riaz
GC Saddam Zafar
GC Asfandyar
​​Project Supervisor(s) Dr Arslan Shaukat
​Project Abstract Most of the currency prediction systems built were based on price history of currency. We combined price history with Fundamental data to get maximum accuracy in prediction. For Fundamental analysis we applied Sequential Ridge regression on variables like interest rates, inflation rates, gross domestic product, GDP gap, money supply of country with varying window size (past monthly values). For Technical analysis we perform more than twenty different analysis on variables open price, close price, high price, low price, volume of candle of specific timeframe. Developed system can provide more than 60% accurate signals. Developed system can be used to get signals for investment ​
​​Project Title Intelligent Adaptive Grasping with Prosthetic Hand using Visual Techniques
​Group Members Syeda Maryam


Syeda Maryam

​​Project Supervisor(s) ​​Dr Usman Akram

Dr. Sajid Gul Khawaja

Dr. Yasar Ayaz

​Project Abstract ​Any human limb is irreplaceable such that if any limb is lost unfortunately, it has a lot of psychological and social impacts on the amputee. With the world excelling rapidly in the field of robotics and prosthetics, it is developing new ways to help amputees. It is being done by making artificial limbs that mimic the same function. So our system will be different from the conventional prosthesis techniques and would be trained for data acquisition and data processing in which the prosthesis grasping will be applicable to the objects lying under the common categories of  Power grasp, Precision grasp, Card grasp, Cylindrical and Tripod grasp. The proposed control system will be based on development of a software mechanism for grasping of the prosthetic hand through the extraction of 3D information from digital images by installing a camera for visual feedback from the patient who wants to move his hand for grasping some object. It will move the prosthetic arm in the specified direction, minimizing patient’s efforts for better grip and improved movement.
​​Project Title ​DSMAC (Digital Scene Matching and Area Correlator)
​Group Members Zainab Ahmad

Nabeeha Zia

​​Project Supervisor(s) ​​Dr Shoab Ahmed Khan

Dr. Farhan Riaz

​Project Abstract ​Unmanned Aerial Vehicles (UAVs), as the name suggests, are aircrafts without a pilot on board. They can be remote controlled or can fly on preprogrammed flight plans. UAVs generally navigate using the Inertial Measurement Unit, which consists of on-board sensors like GPS, Accelerometer, Gyroscope, which provide different functionalities to the UAV. However, rapid growth of error in GPS coordinates can cause bias and drift in the original path. These errors can occur due to low voltage signals, intentional jamming or unintentional interferences. To minimize the dependency on GPS, alternate methods can be explored for trajectory correction in target critical missions. Our project focuses on exploring the potential for Synthetic Aperture Radar imagery in trajectory correction.
‘DSMAC’ or Digital Scene Matching and Area Correlator, is a system that provides a solution to overcome the deviations in the path occurring due to the absence of GPS coordinates. This includes correlation of images with neighboring data based on initial location and detecting the current position from the matched frame of the input video while updating them on the map.
​​Project Title ​Cyber Heart
​Group Members ​Muhammad Saad Salahuddin

Ahmad Moiz

Hassaan Ahmad Siddiqi

​​Project Supervisor(s) ​​Dr Qaiser Chaudry
​Project Abstract ​Heart is a very difficult organ to study in action, considering the person needs to be on the table with their chest open, which is not readily available learning tool for students. ECG is one of the important parameters in studying the Heart. In Cyber Heart we have taken data of ECG signals of healthy and disease patients and link it with 3D heart animated model for the study of ECG on the motion of heart and their co-relation. Medical Students can interact with the beating heart through user interface and visualize it in Virtual Reality through VR headset.
​​Project Title ​MedOpt (Medical Benefit Optimizer)
​Group Members ​Moiez Mahmood

Muhammad Hamza Sajjad Abbasi

Muhammad Ehsan Ullah

​​Project Supervisor(s) ​​Dr Shoab Ahmad Khan

Dr Wasi Haider Butt

​Project Abstract ​Knowledge extraction from data has always proven significant to organizations, also, detection of fraudulent cases has always remained a challenge for health insurance companies/governments. Healthcare programs have to bear significant financial losses due to fraud cases. In this project, an artificial intelligence (AI) enabled medical benefit optimizer system developed that provides, real-time advance visual and predictive analytics of providers and patient financial data. It will automatically generate alarm on digital suspicious transactions. MedOpt provides real-time business intelligence to manage resources efficiently resulting in significantly easier decision-making process. It provides patients a platform to approach better healthcare services. The software also facilitates healthcare support systems in detection and prevention of fraud.
​​Project Title Moboscope
​Group Members Palwisha Gul

Umair Masood

Isbah Shahnaz

​​Project Supervisor(s) ​​Dr Usman Akram

Dr Sajid Gul Khawaja

​Project Abstract ​The problem of high technological scientific facilities not being available to people in affordable prices creates a gap between people and medical diagnosis. Hence to bridge the gap we tend to implement the idea of a low-cost microscope (3D) with image processing techniques. The major aim of this project is to build a low-cost, portable microscope, and to facilitate people by providing them a cost-effective means to connect with frugal science. Moboscope works by capturing high resolution microscopic images using a self-made low-cost microscope connected to a mobile application for real time analysis of the image. The objectives are to provide access to low cost scientific tools that must be available to a huge number of people, to diminish the distance between doctors and patients by providing an easy to carry device and to connect science and society through frugal science.
​​Project Title ​Cyber Heart
​Group Members ​Muhammad Saad Salahuddin

Ahmad Moiz

Hassaan Ahmad Siddiqi

​​Project Supervisor(s) ​​Dr Qaiser Chaudry
​Project Abstract ​Heart is a very difficult organ to study in action, considering the person needs to be on the table with their chest open, which is not readily available learning tool for students. ECG is one of the important parameters in studying the Heart. In Cyber Heart we have taken data of ECG signals of healthy and disease patients and link it with 3D heart animated model for the study of ECG on the motion of heart and their co-relation. Medical Students can interact with the beating heart through user interface and visualize it in Virtual Reality through VR headset.
​​Project Title ​Cyber Heart
​Group Members ​Muhammad Saad Salahuddin

Ahmad Moiz

Hassaan Ahmad Siddiqi

​​Project Supervisor(s) ​​Dr Qaiser Chaudry
​Project Abstract ​Heart is a very difficult organ to study in action, considering the person needs to be on the table with their chest open, which is not readily available learning tool for students. ECG is one of the important parameters in studying the Heart. In Cyber Heart we have taken data of ECG signals of healthy and disease patients and link it with 3D heart animated model for the study of ECG on the motion of heart and their co-relation. Medical Students can interact with the beating heart through user interface and visualize it in Virtual Reality through VR headset.
​​Project Title AI Based Crime Analyzer and Crime Detector
​Group Members PC Sher Islam

NC Qaiser Ali

NC Had Ur Rehman

GC Irfan Haider

​​Project Supervisor(s) ​​Dr Wasi Haider Butt

Dr Ali Hassan

​Project Abstract ​Crime in Pakistan is present in various forms. Organized crimes Includes drug trafficking, money laundering, extortion, black marketeering, political violence, terrorism, abduction etc. Therefore, it is evident that work needs to be done in order to stop these organization that are involved in criminal activities. To predict the future criminal attacks, data analysis needs to be done. To achieve the task of data analysis, the information of previous crime incidents is required. The aim of our project is not only to deliver this data to the users rather provide them real time crime condition  and display the important information regarding crime incidents to the data collector in a way so that he/she can classify and extract and process important aspects of data.
​​Project Title End to end encryption of cellular communication
​Group Members ​Asfa Ali Masood
​​Project Supervisor(s) Dr Mohammad Abbas

Dr. Wasi Haider

​Project Abstract ​Standardized communication protocols enable mobile devices to exchange short text messages. With approximately 8 trillion text messages sent everyday increasing amount of communication apps, governments and non-state actors are also finding new means to compromise and retrieve communication’s data. Sophisticated software and hardware tools are being used to intercept and decrypt traffic e.g by IMSI catchers.In order to overcome these issues certain applications like WhatsApp, Signal etc. provide end to end encryption. However, they require internet to function and are server based. This means the server acts as a mediatory and can be used in order to intercept and decrypt traffic. This will provide user end to end encrypted with no third party or internet using signal protocol ,AES-256 Encryption and hashing
​​Project Title VIZCON (Visual Construction)
​Group Members ​Hafiza Neha Shahid
​​Project Supervisor(s) Dr. Shoab Ahmad Khan
​Project Abstract ​Alone in Pakistan housing market is worth more than 150 million rupees. 40% of home owners are not entirely happy with their home design. A house is a lifetime investment ,when dealing in the realm of architecture ,for a common person the most difficult task can be visualizing a space in order to build it more efficiently .Properly visualizing not only help bring the idea to life but also helps architects and clients to organize and plan every move. Therefore, we need to develop a service that will help clients and architects to visualize their yet-to-be-built property, and get a much better understanding of the architecture they are investing in. VIZCON will help them to better convey their vision to clients when pitching a new idea and inspire the clients to see what their final product will look like. For visualization VR (Virtual reality) technology is being used. With collaboration clients and architects would be able to make new architecture, deactivate different flaws at different stages in time and also furnish it. This allows clients to get a realistic view of the project outcomes and make any desired changes before construction begins. Getting this level of client involvement early in the process helps prevent costly changes later and keeps clients engaged in the project.
​​Project Title Virtual Reality for Amblyopia (VRAM)
​Group Members ​Hadia Zafar
​​Project Supervisor(s) Muhammad Usman Akram
​Project Abstract Amblyopia is a vision disorder in which the brain fails to process the signals sent to it from one eye, as a result the vision of that eye becomes blurred. The World Health Organization (WHO) estimates 19 million children less than 15 years of age are visually impaired; of those, 12 million are impaired due to uncorrected refractive errors and amblyopia. However, in Pakistan and other developing countries with limited screening programs and many children not going to schools, amblyopia remains undiagnosed. If left untreated, it can lead to turning of eyes and eventually blindness as well. VRAM serves as a low cost solution for the treatment of amblyopia. By adjusting the inter ocular contrast in virtual reality, different stimuli are sent to each eye. Dichoptic training enhances the patients ability to use both eyes in conjunction. Patients can perform therapy at home. Data stored on cloud will enable the doctor to track the patient’s progress remotely.
​​Project Title Thermal Peek
​Group Members ​Mohammad Roshaan Butt
​​Project Supervisor(s) Dr. Usman Akram
​Project Abstract ​Safety and security has become a concern on every level whether it is on individual level or group of individuals. To visualize and remain conscious about the surroundings in order to know what’s happening has become crucial and essential especially where visual sight is unable to detect the objects like in darker areas. Therefore, we need to develop the mobile app which is capable to detect the surrounding environment in a thermal mode and let the user know in real time about the entities present in his/her surroundings. The hardware to be used is the thermal mobile camera attached with the android phone which will take the input feed in real time and after processing the data, it will display the required results accordingly.
​​Project Title IIOT based Smart Garment Factory (OBVMS)
​Group Members ​Aimun Tariq

Esha Sadia

Zeb Un Nisa

Uzair Shehzad

​​Project Supervisor(s) Dr Shoab Ahmed Khan

Dr. Ali Hassan

​Project Abstract ​The world is currently witnessing the fourth industrial revolution that is normally well-known as industry 4.0. This analysis has been centered on chance of developing a custom sensible attire production plant which can be a unique approach within the fashion business supported the known problems with the present stitching method and the suggestions to boost them. Recent technological innovations have made-up the method for realizing the sensible industrial plant vision with succeeding historic period known as Industry 4.0. Industry 4.0 has important potential to vary whole producing processes and business models for labor intensive wear and attire factories. Therefore, the businesses within the wear and attire business must be compelled to invest in Industry 4.0 and its enabler technologies to begin the transition from normal plants to the sensible industrial plant to achieve competitive advantage. However, connected studies within the literature area unit restricted and not sufficiently comprehensive to state however these comparatively new technologies might influence the wear and fashion business. The sensible garment ought to provide comfort to user by flexibility, handling and style.

An essential facet once considering e-textiles is that the integration and interconnection of textiles and physics. as a result of textile and physics area unit from materials with completely different

characteristics, embedding electronic elements is restricted. The solutions incorporate attaching electronic devices to textile substrates.

​​Project Title Autonomous Industry Resource Optimization Solution (AIROS)
​Group Members ​Aitazaz Qaiser Khan

Muhammad Usman Khalid

Yaqoob Jamal

Muhammad Siddiq

​​Project Supervisor(s) Shoab Ahmed Khan
​Project Abstract ​As the advent of the technological era streamlines industry and manufacturing a new set of problems arises, one of which is the optimization and efficient usage of resources in manufacturing. The solution that AIROS aims to provide is optimization of machines regarding their usage in industrial manufacturing shop floors. The solution will be presented in the form of a fully functional dashboard containing live information updates to the floor manager alongside a visualized progress of the day’s schedule. The visual display will allow the floor manager using AIROS to observe what task each machine is being utilized for and their progress on that task against the daily agenda. AIROS will provide options to cater for complications arising on the shop floor and allow the floor manager to delay the day’s agenda or reschedule it entirely depending on the state of the shop floor. The options will contain methods to dynamically reschedule the shop schedule keeping dependencies of fabrication orders in mind. Progress of machines will be monitored by an IoT Node that transmits live current profiles of working machines. Algorithms using models trained on a databank of accumulated profiles will then classify the profile and display the manufacturing progress on the aforementioned display. AIROS uses the current profiles and its derived statistics to generate this dashboard for the floor manager and provide a resource optimization solution.
​​Project Title Automated SQL Query generation from Natural Language using NLP
​Group Members ​Muhammad Ali Uppal

Muhammad Shahroze Ali

Muhammad Zeeshan

Amir Hassan Goraya

​​Project Supervisor(s) Dr. Wasi Haider Butt

Ali Saeed

​Project Abstract ​This project aims at developing a system which will accept English query from user and convert it into SQL. This helps novice user who can easily get required contents without knowing any complex details of SQL languages. We can store huge amount of data in databases but casual user who doesn’t have any technical background not able to access data. So we propose a system that will convert English statement given by user to all possible intermediate queries so that user can select appropriate intermediate query and then system will generate SQL query from intermediate one. Finally system will fire SQL query on database and gives output to user. When an interpretation error occurs, users often get stuck and cannot recover due to a lack of guidance from the system. To solve this problem, we present a natural language query recommendation framework
​Group Members ​Waleed Ahmed

Hamza Ehsan Khan

Haania Aamir

Rameen Jafar

​​Project Supervisor(s) Dr. Sajid Gul Khawaja

Saad Zia

​Project Abstract ​A smart camera is a computer vision system which, in addition to the ability of capturing images, is capable of extracting application-specific information from the captured images, along with generating event descriptions or making decisions that are used in an intelligent and automated system. The amount of processing done depends on the information to be extracted from the image. However, in most of the applications, object or data extraction must be done real-time, requiring heavy computations. Processors are being replaced by FPGAs which provide high parallelism , along with reconfigurable computing. This is an opportunity to revolutionise computational speeds in the field of computer vision and implement a stand-alone computer vision system powered by FPGAs.
​​Project Title MANZAREH – A Helping Tool for Visually Impaired People
​Group Members ​Usama Sarfaraz Khan

Rufaida Kashif

Taha Nadeem Dar

​​Project Supervisor(s) Dr. Ali Hassan
​Project Abstract ​Approximately 1.3 billion people in the world, suffer from some or the other form of vision impairment. Out of these, 2 million people reside in Pakistan, who are visually impaired. These people require some additional help in the form of a guide to tell them about the surroundings. In order to assist them, we have developed a system that will read the surroundings for them and later interpret it. For this, we will be using a cellular phone and a cloud service. As soon as the user takes the image of what is in front of them, the image will be sent as an input to the model. The model is deployed on the cloud as well as on the local system. After adequate processing, the model will generate the caption for that image in the form of a structured sentence. By Natural Language Generation tools, the generated sentence will be read out loud to the user. This will not only read the scene for the visually impaired person, but will also allow the user to recognize the people around them. It will operate only upon the desire and requirement of the user. Henceforth, causing no additional burden for them by constantly giving input. In this way, it will reduce the user’s dependency on other person as their guide, making them independent in many matters.
​​Project Title Data Driven Analysis for Project Management System
​Group Members ​Muhammad Mushahid

Ghulam Mujtaba

Azhar Hanif

Mujahid Shaheryar

​​Project Supervisor(s) Dr. Wasi Haider Butt

Ali Saeed

​Project Abstract ​Today most enterprises are providing services or taking services from other organizations. These services are commonly known as projects. Each project has different format of data and fields. This data is collected in excel files and processed manually for reporting. All collected data is stored at different locations. There is no central place where data is stored. Due to this data, redundancy and duplication problems occur and lead to consume more efforts and time on data. To address this issue, A “Management System” is developed which intends to provide all details and interfaces for maintaining projects. The application “HEC Project Management System” is a new, self-contained product which is being made to fulfill the requirement of all stakeholders and end user for official use to manage different projects. By implementing this system, the objective is to achieve efficient management of projects and effective decision making.
​​Project Title Patient Experience Monitoring (PEM)
​Group Members

Rushmah Saleem

Usman Fakhar

Fizza Akram

Myra Faisa

​​Project Supervisor(s) Dr Shoab Ahmed Khan

Dr. Arsalan Shaukat

​Project Abstract ​Patient Experience Monitoring focuses on designing and reacting to patient experience to determine the best possible ways to provide the optimal patient satisfaction, loyalty and advocacy. This is done mainly by using an RFID framework or it deals with the emotion-based survey filled by patient which caters different factor like environment, registration, doctor etc., as an overall experience of a patient in a hospital. Our project deals with the introduction of a framework which combines Radio Frequency Identification (RFID) technology, with the Android based automated outpatient feedback form. The output from RFID is the overall time spent at various stations in a hospital. While the output from the automated survey is the overall experience a patient has during his stay in a hospital which is an overall satisfaction index (number). We will then use this data and apply an algorithm that predicts the experience of patients using the waiting time and environmental factors captured by IoTs. Furthermore, we will apply analytics on the data gathered and provide the results to the hospital management using dashboards.
​​Project Title Real Time Safety Precaution System (RTSPS)
​Group Members ​Muhammad Saariya Faiz

Hamza Sohail

Muhammad Khizar Hayat

Umer Abdullah

​​Project Supervisor(s) Dr. Shoab Ahmed Khan

Ali Syed

​Project Abstract ​Mostly people in buildings or offices remain carelessly indulged in their routine activities while being unaware of their surroundings i.e. there may possibly be smoke, fire ignition or gas leak, etc. This minor negligence can lead to major tragedy which can adversely affect the office’s future. Therefore, we need to develop a product which can help monitoring the office’s environment. The aim of RTSPS is to predict about any calamity prior to it happening in order to avoid any unforeseen circumstances. The environmental statistics from sensors in real time are sent over to the cloud via physical node. IP camera is used for the environmental live feed being accessible through the web portal as well as android application. The cloud then saves statistics in database, apply decision rules on data and sends the data as well as decisions to the web portal and android application in the form of alerts or notifications respectively. Web portal and android application shows the graphical view of data from every sensor separately as well as the live feed of cameras. The designed system is limited to the specific firm where the user monitors the office’s environment and is responsible for precautionary steps
​​Project Title ​Vehicle monitoring & Management system​
​Group Members ​Syed Zulqarnain Mushtaq,   M. Faisal Rabbani,  Naqash Ali
​​Project Supervisor(s) ​​Asst. Prof Sajid Gul Khwaja Dr M. Usman Akram
​Project Abstract ​Vehicle management holds a firm ground in both manufacturing and service industry. Fleet management plays a substantial role in an organization’s growth. Maintenance of vehicles is critical in terms of cost, availability and customer satisfaction. The major cause for delays are inefficient and ineffective maintenance procedures. Most of the organizations are using manual record keeping. The paper-based system makes it difficult to maintain service history and failure trends for vehicle components. To avoid major market losses a maintenance model is generated. The maintenance system, integrates the flow of information by incorporating different maintenance strategies. The data collected includes both manufacturer and operator requirements. The model incorporates the inspection method by using checklists and maintenance schedules. The system is capable to provide the basis for an advance system development, but it is limited by infrastructure requirement, high personnel skills, and state-of-the-art technology.
​​Project Title Virtual Reality Exposure Therapy (VRET) using Google Cardboard
​Group Members ​Qaim Hassan Abbas (EME, NUST),  Muhammad Abeer (NUCES)
​​Project Supervisor(s) ​​Dr Shoab Ahmed Khan,  Dr Usman Akram
​Project Abstract International companies and top retailers worldwide are seeking ways to generate profits from advanced processes in their operations, promotions, merchandising, human resource, and security departments. Video analytics is one of the hot topics these days and video based solutions are not onlybeing utilized for surveillance, but also as a solution with intelligence and analysis ability, creating great value in retail environment. Although a lot of work has already been done in this area, the problem still lies in processing the videos in real time and making decisions. In our project, we are designing a solution for trolley counting and till management for stores in real time. It uses HOG (Histogram of Oriented Gradients) for locating objects of interest, which are trollies and people. HOG operates on local cells, therefore it is invariant to geometric and photometric transformations, except for object orientation. The algorithms are tested on actual data gathered from stores and are validated by experts. For real time analysis, we embed or solution on raspberry pi which is directly connected with IP cameras in the store. We have also designed a dash board in order to provide real time analytics for policy makers who can use this data to give different kind of promotions in order to improve their sales.
​Group Members ​Rana M junaid, AwaisShahbaz, Bilal Riaz, UmairJavaidManj
​​Project Supervisor(s) ​​Dr Shoab Ahmed Khan,  Asst. Prof. Sajid Gul Khwaja
​Project Abstract ​In this age of wireless communication and wireless networks the need to transmit high data rates without the expense of high transmit powers and high bandwidth utilization has provencontinuous phase modulation (CPM) one of most attractive modulation schemes in band limited channels. It is for this reason that NATO uses binary CPM modulation for wireless communication at physical layer for its narrow band channel i.e. 25KHz. Increasing the data ratedirectly increases spectral Width of CPM which become a direct source of inter symbol interference (ISI) and hence depredating the overall efficiency of system in channels with limited capacity. We propose a solution to the reception of band limited CPM signal that is based on Soft inputsoft output (SISO) MLSE receiver/Equalizer using the model of Andres- Moctezumawith a serial concatenation of well known BCJR decoder. At the end performance of this receiver isevaluated using Stanford University Interim channel (SUI). Simulation results shows markeddifferences in bit error rate performance for conventional versus proposed receiver. NBWF is expected to operate in the same frequency range, employing the same channelization; furthermore, the NBWF should allow operation in the tactical UHF band as well. The current waveform proposal is based on Continuous Phase Modulation (CPM), which is widely used in mobile communications, due to the constant envelope property of the modulation scheme.
​​Project Title Ambience Characterization Using Machine Learning
​Group Members ​Asad Mansoor Khan
​​Project Supervisor(s) ​​Dr M. Usman Akram,  Asst. Prof. Sajid Gul Khwaja
​Project Abstract ​The amount of digital content has increased manifold over the past few years. The same can be said of the compute power that is now accessible thanks to a number of companies providing cloud computing at inexpensive tiers and plans. Most of this digital data is present in the form of photos and this compute power can be used to make “sense” of the emotions in a photo by tagging them automatically. While there are a few systems already in place that do the same (think of Facebook’s auto tagging of photos), none of them gives access to a portal where anyone can use their own set of data to get an analysis of ambience of a picture. In my project, the scope of detection of emotions is further extended by including the photos in which there aren’t any people present. This way, a machine learning system can not only detect the emotions of a picture in which there are people but can also detect the “emotions” or the ambience of scene like a playground, waterfall etc. In this project, I am using Google’s Cloud Platform and Google’s Prediction API. The Platform provides a portal to store data and can become a frontend for a web app while the Prediction API uses machine learning to create an accurate model using the data that has been provided to it. Leveraging Google’s Cloud Platform provides an inexpensive yet powerful solution for computing and data storage.
​​Project Title Sky Guard Radar Simulator​
​Group Members ​Irfanullah
​​Project Supervisor(s) ​​ FarooqeAzam
​Project Abstract ​Simulation is the imitation of the operation of a real-world process or system over time. The act of simulating something first requires that a model be developed; this model represents the key characteristics or behaviours of the selected physical or abstract system or process. The model represents the system itself, whereas the simulation represents the operation of the system over time. Sky Guard Radar Simulator is a simulator developed for helping technicians and trainees for training purposes. It has a set of 128 buttons, which have been interfaced with PC. The system is user friendly and will prove to be very effective in skilfully training the learners.
​​Project Title Human Recognition from crowd using video analytics​
​Group Members Aamir Saeed, Ijlal Ahmed
​​Project Supervisor(s) ​​Dr Ali Hassan
​Project Abstract ​It was required to build a smart program for the purpose of surveillance; that will provide information about every individual possible in a crowded place e.g a train station. In this regard, the video analytics tools are applied using live feed of camera. The operator in the main control panel is able to detect each person from a crowd remotely , where ever the surveillance camera is installed. The program will certainly with a small size processor keeping in view the mobility issue. So the use of Raspberry Pi is applicable under such computational and mobility requirements. It’s almost equal to a credit card in size but with a great computational power and plug and play ports.Carrying the attributes of video analytics, the program is capable tobreak the live video input into frames. Then these frames are passed through filters and face detecting algorithms. The acquired images are matched with the project’s database already stored in the control room of the organization.This program is can easily detect the crowd behaviours and wanted persons also using face recognition algorithm. It’s a program for the sole purpose of security survillence in sensitive areas like military setups and housing schemes like bahria town etc.
​​Project Title ​ Disease Detection of Gastroenterology Image using C++
​Group Members ​Aamir Saeed, Ijlal Ahmed
​​Project Supervisor(s) ​​Dr Ali Hassan
​Project Abstract ​It was required to build a smart program for the purpose of surveillance; that will provide information about every individual possible in a crowded place e.g a train station. In this regard, the video analytics tools are applied using live feed of camera. The operator in the main control panel is able to detect each person from a crowd remotely , where ever the surveillance camera is installed. The program will certainly with a small size processor keeping in view the mobility issue. So the use of Raspberry Pi is applicable under such computational and mobility requirements. It’s almost equal to a credit card in size but with a great computational power and plug and play ports.Carrying the attributes of video analytics, the program is capable to break the live video input into frames. Then these frames are passed through filters and face detecting algorithms. The acquired images are matched with the project’s database already stored in the control room of the organization.This program is can easily detect the crowd behaviour and wanted persons also using face recognition algorithm. It’s a program for the sole purpose of security survillence in sensitive areas like military setups and housing schemes like bahria town etc.
​​Project Title Vaccination Surveillance and Planning System
​Group Members ​Abdul AleemKhusro,  Behzad Ali Haider,  Ali Waqqas
​​Project Supervisor(s) ​​Prof. Shoab Ahmed
​Project Abstract ​”The vaccination preventable diseases can be eradicated with help of the proper monitoring and surveillance. The process for monitoring has to be handy and quick, so in our project we are developing an information management system which will store the data of all the children who are being vaccinated, and what vaccines they have received, and what future vaccinations they should receive, as prescribed by WHO. The system will also store fingerprint data. The android client on the field, with a fingerprint scanner attached, will communicate with the server, and make CRUD operations on the server. The system will also send SMS alerts about future vaccinations through a GSM service. The server will be using the .Net framework with web API, and Entity Framework”.
​​Project Title Biometeric Voting Machine
​Group Members ​Awais Ali Zia,   Sohail Ahmad Shahzad,  Muhammad Vacceph Khan,  IhtishamIkram
​​Project Supervisor(s) ​​ Dr Shoab Ahmed Khan
​Project Abstract ​The fate of any country is decided by the elected members of Parliament and National Assembly of that country. Therefore it is very important that election process of that country must be fair enough so that only the most deserving persons are elected. All of us today are very well aware of all the kinds of pre-poll rigging like irregularities, mismanagement, corrupt individuals and other human errors which can effect this very important election process  of any country. In order to overcome these problems we are replacing this old and manual system by an “ELECTRONIC VOTING MACHINE”. Our final product will be a machine for a Polling Station having two types of database, one for all the people eligible to cast vote and other for the candidates appearing in that area for election. There will be a thumb scanner interfaced to our machine and a touch screen. All the machines on different polling stations are connected to each other and all the results will be displaced on a website in the form of graphs in real time. The process will be carried out in such a way that a person who is willing to cast his vote will appear to respective polling station and he will first enter his ID card number and then scan his thumb. If that thumb impression matched with already stored impression against that id number than he will be allowed to cast his vote. This will ensure the fairness and dignity of this very important process as well as time efficiency increases also as compared to manual system.
​​Project Title Real-time Water Purity Surveillance through implementation of IoT​
​Group Members ​Arsalan Soomro, Farhad Satti, Saad Umar, Talha Javed
​​Project Supervisor(s) ​​Dr Shoab Ahmed Khan
​Project Abstract ​Water impurity and the diseases that come with it present a very real and serious threat in current times. Every day, 2 million tons of sewage and industrial and agricultural waste are discharged into the world’s water, the equivalent of the weight of the entire human population of 6.8 billion people. The UN estimates that the amount of wastewater produced annually is about 1,500 km3, six times more water than exists in all the rivers of the world. We wish to bring these alarming numbers down to some extent by interfacing and interconnecting different sensors into one single unified node. Each sensor in the node would monitor a different impurity factor of water. These nodes will then be installed at sufficient distances along the irrigation path of water of a certain locale. Through the implementation of IoT (Internet Of Things), these nodes will constitute a surveillance network of sorts which would be providing feedback in real time if and when it detects an abnormality in impurity level at some node. The Internet of Things (IoT) is the network of physical objects—devices, vehicles, buildings and other items embedded with electronics, software, sensors, and network connectivity—that enables these objects to collect and exchange data. In our case, nodes of the network we wish to implement.
​​Project Title Low cost automated monitoring system for MPLS based ISP network​
​Group Members ​Hafiz Tahir Idrees,   Raja Ahsan Hayat,  Haroon Iqbal
​​Project Supervisor(s) ​​Dr Moazzam A. Khan
​Project Abstract ​The computer networks play their role in almost every place now-a-days. With the evolution of e-commerce, electronic banking and many concepts like this has increased the use of computer networks a lot. Moreover, the internet use has also become immense. The companies, enterprises open offices at various locations and allow connectivity to its employees and customers. With growing traffic on the network with each passing day, different techniques are adopted to ensure proper routing and efficient use of network resources. And in similar way the troubleshooting process has become very important. The project team vows to work on the relatively new technology called “MPLS” (multiprotocol label switching). A network shall be designed/ simulated which would replicate just as the company networks are and would be using MPLS as routing technique which would help for efficient routing. The team would develop a monitoring software that would monitor the complete network and would tell the administrator as soon as something goes wrong in the network. In this way he would not have to look into the detail of each interface, rather this would directly tell which interface or connection has gone low. In this way, a company network will be fully and effectively monitored by the administrator and would help in swift trouble hooting.
​​Project Title Traffic Automation and Routing System
​Group Members ​Sheryar Ahmed,   Ali Abbas,  Yasser Zubair, Rubba Rehman
​​Project Supervisor(s) ​​Asst Prof. Sajid Gul Khwaja
​Project Abstract ​The traffic congestion problem is a common issue for the residents of metropolises. Although expanding the capacity of transportation systems and stimulating the public transportation may decrease the traffic congestion, they cannot completely solve the traffic congestion problem. The proposed solution is to design an artificially intelligent system that is equipped with Algorithms enabling it to run on its own Routing algorithms, shortest paths algorithms and Image processing algorithms. It will also be capable of automating the traffic system of a city by calculating shortest routes to given destinations, by apprehending the traffic flow and accordingly managing traffic signals. It can analyze short time & long time trends by running data mining algorithms on daily data stored in the cloud and make smart route recommendations accordingly.In its implementation, Linux will be used for hardware interfacing and python as the primary programming language. The hardware we are using are the cameras and Raspberry Pi. Having implemented this system, we will come up with a monitoring app for the traffic control room which will be used to keep the human administrators up-to-date on the traffic situation and a Traffic guidance app for end user which will help them in finding the current congestion free routes to their destination.
​​Project Title Smart Garden
​Group Members ​Fatima Ehsan,   Hafsa Anjum,  Sammon Azam
​​Project Supervisor(s) ​​Dr M. Usman Akram
​Project Abstract ​Smart gardening technology refers to the ease of gardening by interacting with the plants without any prior knowledge or know how about handling different types of plants. The key goal of the project is to provide user with the user friendly application over the android cell phone with the help of which he is able to monitor the health of his plants. Data would be taken from the sensor nodes deployed over the garden, communicating with the android application through cloud and displayed in a readable form for the user. User would also be notified if the nodes detect any fluctuating changes in the environment of the plants that might not be suitable for them. In addition, automated sprinkling system would also be provided through which the user can water the plants with the ease of a button.
​​Project Title Blue Force Tracking
​Group Members ​Arooj Waqar,   Fatima Hassan,   Sabaina Haroon,   Sherin Batool
​​Project Supervisor(s) ​​Dr Shoab Ahmed Khan
​Project Abstract ​The basic idea behind the project is to create a wireless sensor network between the soldiers at Warfield. Sensing the vital signs data, their location through GPS, etc. and transmit them over the network to the base station where the commander will have an app which would track the location of each soldier through GPS and display it on map GUI. He would also have the facility of having a mission editor to assign missions and send commands to the soldiers accordingly. At Soldier end two functionalities will be implemented. Firstly, each Soldier will have a smartphone with him; the smartphone will be connected through Bluetooth to vital signs detection watch and an IOT for weapon detection .Secondly, the smartphone will collect all the data from the sensors (currently we are collecting the data from GPS) and send the sensor data back to the commander at base station via Ad Hoc network
​​Project Title Blind’s Eye
​Group Members ​Hamza Ali Khan,  Adil Hanif,  Haris Ashfaq,  Sohail Rashid
​​Project Supervisor(s) ​​Dr M.Usman Akram
​Project Abstract ​Over 285 million people in the world are visually impaired, of whom 39 million are blind and 246 million have moderate to severe visual impairment (WHO, 2011). Smart Phones being an effective part of our Life can be used in a good positive manner to help the Blind Person to be more contributing towards the society. By using Computer Vision technology we can detect people/objects in runtime and by applying post and pre-processing techniques valuable information can be extracted. In our project, we use Viola Jones algorithm for Recognition of Person. For Recognition of Objects we are using SURF and SVM algorithms. The android app will also give a social interaction to Blind Person by direct video chat.  The algorithms mentioned are not only responsive but also provide accurate results. Using a smart phone camera visually impaired person can easily interact with their surroundings using this application. By implementing the application using Cloud Computing technique it will be more reliable and approachable. By using Computer Vision techniques we are trying to make the life more easy for the Blind Persons. Our work is not only unique but very much contributing towards the life of visually impaired person.
​​Project Title Speaker Bulb controlled via android application​
​Group Members ​Naqash Ahmed, Tahir Ayub,  Jabir Ali, Hasnain Maqbool
​​Project Supervisor(s) ​​Dr M.Usman Akram
​Project Abstract ​The purpose of this project is to control a speaker bulb via bluetooth So that we can use it for our desirous purpose e.g to set Alarm, Entertainment, etc. Android Application will be an interface between user and Project. This project will enable a user to play all audio files present in his phone or SD card. This project is highly feasible i.e Technically and Financially. During making Smart work will be done. Easy to understand and implement.
​​Project Title Low cost automated monitoring system for MPLS based ISP network​
​Group Members ​Hafiz Tahir Idrees,   Raja Ahsan Hayat,  Haroon Iqbal
​​Project Supervisor(s) ​​Dr Moazzam A. Khan
​Project Abstract ​The computer networks play their role in almost every place now-a-days. With the evolution of e-commerce, electronic banking and many concepts like this has increased the use of computer networks a lot. Moreover, the internet use has also become immense. The companies, enterprises open offices at various locations and allow connectivity to its employees and customers. With growing traffic on the network with each passing day, different techniques are adopted to ensure proper routing and efficient use of network resources. And in similar way the troubleshooting process has become very important. The project team vows to work on the relatively new technology called “MPLS” (multiprotocol label switching). A network shall be designed/ simulated which would replicate just as the company networks are and would be using MPLS as routing technique which would help for efficient routing. The team would develop a monitoring software that would monitor the complete network and would tell the administrator as soon as something goes wrong in the network. In this way he would not have to look into the detail of each interface, rather this would directly tell which interface or connection has gone low. In this way, a company network will be fully and effectively monitored by the administrator and would help in swift trouble hooting.
​​Project Title Hospital Management and Automation System​
​Group Members ​Haras Ahmad Zeeshan, Abdul Hannan, Nokhez Naeem
​​Project Supervisor(s) ​​ Dr Shoab Ahmed Khan
​Project Abstract ​Many hospitals currently use a manual system for the management and maintenance of critical information. The current system requires numerous paper forms, with data stores spread throughout the hospital management infrastructure. Often information is either incomplete or does not follow management standards. Paper work done is often lost in transit between departments requiring a comprehensive auditing process to ensure that no vital information is lost. Multiple copies of same data exist that may lead to inconsistencies in data in various data stores. Basically the information typically involves; patients personal information and medical history, staff information, room and ward scheduling, staff scheduling, operation theater scheduling and various facilities waiting lists. All of the information must be managed in an efficient and cost wise fashion so that an institutions resources may be utilized effectively. This system will automate the management of the hospital making it more efficient and error free. It aims at standardizing data, consolidating data, ensuring data integrity and reducing inconsistencies. Furthermore adding features like access through RFID card and usage of Bluetooth enabled smart medical devices will help to raise standards and effectiveness of system in various manners. Applying data mining on data collected through this system will allow authorities to fix problems in the environment as they (problems) have a direct connection to people living in that environment. Business intelligence might play its part in data mining and led to business growth certainly.
​​Project Title Detecting the leakage of gas using Microcontroller
​Group Members ​Umair Arif
​​Project Supervisor(s) ​​Dr Shoab Ahmed Khan
​Project Abstract ​Home security has been a major issue where crime is increasing and everybody wants to take proper measures to prevent intrusion. In addition there was a need to automate home so that user can take advantage of the technology. The project is aimed at developing the security of Home against Intruders, Gas Leak and Fire. In any of the above three cases any one met while you are out of your home than the device sends Notifications on your phone to the emergency no provided to it. The report consists of a background into the area of Arduino microcontroller and mobile App communication, how they are interfaced to each other and AT commands set used in communication.
​​Project Title Real Time EEG Signal Acquisition & Analysis System
​Group Members ​Aaima Mansoor,   Qamar Ali,  UzairAthar
​​Project Supervisor(s) ​​Dr M. Usman Akram  Asst.  Prof. Sajid Gul Khwaja DrImran Khan niazi
​Project Abstract ​Neurological Disorders are prevalent nowadays. Neuro centers are few and not easily accessible, especially to those living in remote areas. EEG machines are fairly expensive for a third world country, like Pakistan. Digital Signal Processing can be proved helpful in the medical field, and especially for heart and brain related diseases. In this respect, this project aims to create an application that allows digitized data from a digital electroencephalogram (EEG) device to be graphed as a waveform in a personal computer and may also be printed. Analog brain signals from a patient are acquired using the digital grass electrodes and electroencephalogram hardware devices. These signals are then converted to digital form using an analog-to-digital (ADC) circuit. Data acquisition is also at regular intervals, which is accomplished by using a random access memory (RAM) chip. Moreover, an analysis of the acquired digitized signals is done to detect brain abnormalities, if present, and this is done by comparing the signal values with the values acquired from the available medical databases.
This project is a C++ console-based implementation primarily intended for a signal composer. This will include the basics in composing a signal and acquiring its attributes and features. Moreover, RaspberryPi and Raspbian (Debian-based OS) are being used for the said purpose.
​​Project Title Real Time Gesture Recognition for Sign Language
​Group Members ​Zeeshan Khalid,   Arif Sultan,  Umar Riaz, Nouman Mushtaq
​​Project Supervisor(s) ​​Dr Arslan Shaukat
​Project Abstract ​Computer vision needs to play its role to overcome the personal constraints. Human computer interaction is exploring all new avenues. People being able to command computers using just their voice. People with disabilities cannot use this facility as much. There is a need to facilitate people with speaking and hearing disabilities. In our project, we use OpenCV algorithms for hand gesture recognition. These algorithms not only allow speedy detection and tracking; but also provide very accurate results. We will be using videos initially to train the system to recognize the gestures. Finally we will use live stream from webcam to capture gestures. Generally, machine learning will be used to teach the system different signs and corresponding words. Finally the system will be able to interpret what a person is saying using sign language.
​​Project Title Application Aware OSI L7 Router using NetFGPA 10G
​Group Members ​NS Nauman Shahid  ,NS Noman Sharif  ,NS Asadullah
​​Project Supervisor(s) ​​ Dr Shoab Ahmed Khan (EME) , Dr. Zaheer Ahmed (C@RE)
​Project Abstract ​The exponentially increasing amounts of data over time and the need for even faster means of transferring that data and/or carrying out communication over computer networks calls for major changes in the digital communications world. We are already witnessing a switchover from IPv4 to IPv6 to cater for expanding networks of communication devices. With all these faster means and modes of communication, we also need to reduce redundancy and increase bandwidth efficiency and usage. Our Senior Project is something from the future and for the future of Computer Networking. Using NetFPGA 10G (A joint effort of the Leland Stanford Junior University and The University of Cambridge) Networking Development/Prototyping Platform, we have developed an Application Aware Router which sits as an intermediate device between the end user and the servers. Our device aims at filtering our network packets using their URLs and Protocols to direct them to specific servers depending on the required bandwidth by those URLs. We aim to achieve Higher Server Cache Hit Rates and Higher Bandwidth Efficiency.
​​Project Title Analysis of OCT images for noise removal and layer Segmentation
​Group Members ​NS Sumsam Ullah ,  NS Shujjat Ali  , PC Usman Tariq  , PC Syed Fraz Ali
​​Project Supervisor(s) ​​Dr  Usman Akram (EME)
​Project Abstract ​Many important eye diseases as well as systemic diseases manifest themselves in the retina. While a number of other anatomical structures contribute to the process of vision, this project focuses on retinal imaging and image analysis. Retinal layer thickness, evaluated as a function of spatial position from optical coherence tomography (OCT) images is an important diagnostics marker for many retinal diseases. However, due to factors such as speckle noise, low image contrast, irregularly shaped morphological features such as retinal detachments, macular holes, and drusen, accurate segmentation of individual retinal layers is difficult. So the very first step before we can go through to layer segmentation is noise removal. We are doing research on how to efficiently remove noise without losing any of the valuable data.  Once the noise is removed layer segmentation would be less difficult to do. And the individual segmented layers will help diagnose the ocular diseases.
​​Project Title Multi-Agent Distributed Autonomous System
​Group Members ​Khawaja Waleed Iqbal  , Alamgir Nasir                                                                                            Dania Murad  , Amal Rauf
​​Project Supervisor(s) ​​Dr Yasar Ayaz (SMME)  , Asst. Prof. Sajid Gul Khawaja(EME)
​Project Abstract ​The primary objective is to address the field of multi agent cooperative environment, to develop such an algorithm to enable machine-machine interaction with no human involvement. The secondary is to participate in the ROBOCUP 2014, to demonstrate our work in this field. Our aim is to   compete with different teams from all over the world, who are working in the same area of research. The target is the Standard Platform League in ROBOCUP 2014.
This project involves cooperative multi agent robotic systems in dynamic adversarial environments. We shall be working on different modules of the Robot i.e. vision, behavior, motion, path planning, navigation, kick, obstacle avoidance etc, developing different strategies and algorithms to make the robot play soccer by  cooperating with the team members.
Another aspect surveyed is the vision system trend that is shifting from global vision, to local Omni-directional vision, to front-facing local vision, which shows the evolution is towards biologically inspired robot soccer agent, the human soccer player. Each team member has a different role, goal keeper, defender and attacker. We shall implement different strategies for each role.
​​Project Title Jarvis (Artificial Intelligence) Interface
​Group Members ​Muhammad Ali  , Muhammad Ajmal                                                                                    Muhammad Asim   , Muhammad Sami
​​Project Supervisor(s) ​​Dr Usman Akram (EME)
​Project Abstract ​This project provides an interface through which we can control computer and mobile phones using other mobiles and computers through voice recognition or commands. There is an application in IPhone5 named “Siri” if we install it on an IPhone5, we can control it with our voice usually used while driving. In “Siri” if we ask the mobile to call a specific person, the app converts it to the text this text goes to server where it is converted to the command and then is executed on the phone as result the call is made. Using this concept we are trying to make the app in which we can control our other mobiles and also our personal computers via one phone or computer. It will enable us to send texts or make calls from a phone that is out of our reach also we can fully control our computers but they all must be connected with the Internet.
​​Project Title Public sentiment analysis of news
​Group Members ​Burhan ud din abbasi  , Taha habib ,  Qalab-e-abbas                                                                        Sameed hayat
​​Project Supervisor(s) ​​Dr Shoab Ahmed Khan (EME)
Dr Usman Qamar (EME)
​Project Abstract ​Recent years have brought a significant growth in the volume of research in sentiment analysis. The intent of this project is to identify the trends in role of media in forming public opinion. The news articles makes use of different text that either yields positive , negative or neutral impact on sentiment .This project will make use of news report publicly available online to analyze their impact on sentiments of public. The novelty of this project is adopting news from social media website ‘twitter’. We identified two subtasks that need to be addressed: identification of target (Pakistani public), and categorization of news into political, economical, financial, crime etc separation of separate positive or negative opinion from good or bad news. These sentiments could be used to predict the role media is playing in shaping public’s response and sentiments towards certain news. The outcome of the process is a set of news organized by their degree of positivity and negativity.
​​Project Title Smart Real Time Traffic Monitoring System
​Group Members Umar Majeed
Qasim Bajwa
Furqan Masood
Areeba Ajaz
​​Project Supervisor(s) ​​Sajid Gul Khawaja, Dr. Shoab Ahmad Khan
​Project Abstract ​Time is everything in this busy world; everyone is in hurry and has to get to some place within a blink of the eye. Sadly in Pakistan, with the number of cars increasing by the day, being stuck in traffic for hours has become in evident specially during the office opening and closing hours and the increasing congestion levels in the traffic are a mere distress for the mind. Imagine a system through which you can automatically monitor traffic congestion at busy intersections and plan your journey accordingly.A similar system is proposed in this project which monitors traffic flow at intersections and tells users the congestion levels at their respective routes. This system intends to find the congestion in traffic flow using live video stream which is taken from the camera, deployed at the road side.The video frames are pre processed and features of moving objects are defined and detected, during the tracking process a filter is devised to detect the velocity of the traffic flow. To give the results more meaning for layman the results are sent to a centralized traffic database management center via a GSM module. And an application is developed, which uses these congestion levels from a centralized database to mark the congestion levels on a map.
​​Project Title Remote/Epileptic Seizure Detection using EEG
​Group Members ​Ammama Furrukh Gill   ,  Syeda Alishbah Fatima  , Aafreen Nawaz  , Ayesha Nasir
​​Project Supervisor(s) ​​Dr. Shoab Ahmed Khan (EME)  , Dr Usman Akram (EME)  , Asst. Prof. Sajid Gul Khawaja (EME)
​Project Abstract ​Generally, seizures are known to be a sudden surge of electrical activity in the brain which enables the effected person to behave abnormally for a short time period. These seizures are not diseases in themselves but in fact, are the symptoms of various different disorders that might be affecting some areas of the brain. Electroencephalography (EEG) is a device used to measure the fluctuations in the brain activity through the electrodes, pasted to different areas of the brain. This machine helps record the spontaneous brain activity for a short time span which makes it a handy tool for detecting brain abnormality that could help diagnose the disease causing it. The ultimate aim of this project is to come up with a remote EEG machine which is able to automatically detect a seizure, without the help of an expert. CHB-MIT Scalp EEG Database is being used as the main data set. The data found has been pre-processed already. So, it was changed to the MATLAB readable format and then, was tested by applying a variety of features to it. After that, the GMM and the SVM classifiers were applied to understand the classification accuracy of the features. Once the feature set with the best classifier results have been identified, an improved version of the classifiers was worked upon and then tested on the available data. Real-time processing of the data is being done on a device known as the ‘Raspberry pi’ and the preferred language of programming is C.
​​Project Title Centralized Control for Multi Agent Autonomous Systems
​Group Members ​Usman Shahid , Umer Javed  , Musaub Shaikh
​​Project Supervisor(s) ​​Dr Shoab Ahmed Khan (EME)
​Project Abstract ​The aim of the project is to design and implement centralized intelligent control for multi-agent distributed autonomous systems which is responsible for the decision support and collaborative interaction of the individual agents. The aforementioned scheme is implemented for the development of a team of autonomous robots which could effectively play a game of soccer in real-time dynamic environments.
The basic software architecture follows a synchronous modular approach which basic modules responsible for data acquisition, strategy planning, path planning, control generation, and transmission of command signals. The system first acquires the visual data for the environment and detects the respective physical objects. The acquired positional data is then transferred to a strategy planner, which uses statistical models and algorithms to predict the most optimal position for each agent. The data containing final positions is passed to the path planner whose basic task is to plan the path from initial to final positions. Next, the control generates the necessary control signals to follow the predicted path. Finally the transmitter transmits these control signals to the individual agents. The language domain for the system is C# which implements a highly parallel system through Task Parallel Library.
​​Project Title Automated Air Defense Deployment
​Group Members ​Musa Ali  , Asad Amin Khatak , M.Sajjad Khan
​​Project Supervisor(s) ​​Dr Shoab Ahmed Khan (EME)
​Project Abstract ​Automated air defense deployment (A3D) is a process to deploy radars and weapons using 3D DEM data. A 3D digital elevation model (DEM) is to represent the continuously varying topographic surface of the earth, and it is a common data source for terrain analysis and other spatial applications. Automated air defense deployment, which saves time and resources wasted on manual search of deployment areas, will maximizes coverage and security under available resources. The project aims to develop a system which would find deployment points using DEM data of interested land, based on user specified valuable assets/points (VA/VPs), attack direction of enemy, weapons and radars properties, using mathematical models and training data through artificial intelligence algorithms. This system is artificially trained to find the best deployment points keeping   constraints of the scenario and all important aspects of deployment in consideration. The software is linked to available Radar and Weapon database. Systems also provide users the details of a DEM file in terms of common metrics used in Air Defense.
​​Project Title Dental Biometrics: Human Identification Based On Dental Work Information
​Group Members ​Taimur Mehmood ,  M.Zeeshan Zafar   , M.Bilal Bhatti
​​Project Supervisor(s) ​​Dr Shoab Ahmed Khan (EME)  , D Usman Akram (EME)
​Project Abstract ​Dental biometrics is used in the forensic odontology to identify individuals based on their dental characteristics. Unlabeled post-mortem dental radiographs are compared with labeled ante-mortem dental radiographs stored in a database to find the highest similarity. This work presents a method for human identification based on dental work information extracted out of panoramic dental radiographs . The proposed method consists of three main processing stages: (i) feature extraction, (ii) creation of a dental code and (iii) matching. In the first stage, seed points of the dental works are detected by thresholding. A refined segmentation is obtained by using a snake (active contour) algorithm. The dental code incorporates information about the position (upper or lower jaw) and size of the dental works, and the distance between neighboring dental works. The matching stage is performed with the Levenshtein distance (Edit distance) which is used to infer the subject’s identity. The costs of the insertion, deletion and substitution operations, were tuned to make the matching algorithm more sensitive. The method was tested on a database including 68 dental radiographs and the results are encouraging. An equal error rate (EER) of 9.8% using ROC analysis and an accuracy of 82% for top-1 retrieval were achieved.
​​Project Title Crime Classification of GTD (Text Mining)
​Group Members ​Jaffar Hameed  , Raza Ullah
​​Project Supervisor(s) ​​Sir Wasi Haider Butt
​Project Abstract ​In this project we implement a software which is capable of analyzing the data held in GTD, the global terrorism database. This software will be taught about the data already present in GTD and by analyzing this data it will generate a completely new data set or information not known before. Our project is a text mining project which will use naive Bayes classification algorithm. The technologies we have use in our project are C#.Net, Sql Server Database and Microsoft Excel (for presenting the data).The efficiency for our results is about 99.3% .
Our actual task is to take the summary of each record of crime and identify that which type of crime it is.
We have about 46000 valid records in our data base so as a rule of thumb of naive Bayes algorithm, we will divide them into two sets of which one will be called training data set and the other one, testing data set. We will manage for the required information by ourselves in training data set and this information would be feed into the software from which it will learn and generate a structure which will help in getting the required information in testing dataset.
​​Project Title iFundus: Disease Diagnosis and Mining over Fundus Cloud
​Group Members ​Ahmad Farooq ,   Nida Ibrahim  ,  Amna Shahnawaz
​​Project Supervisor(s) ​​Dr Shoab Ahmed Khan (EME)
Dr Usman Akram (EME)
​Project Abstract ​285 million people worldwide are visually impaired! 39 million of these people are blind and 80% of blindness is avoidable! The only reason for them to be blind is the lack of professionals and lack of costly equipment for proper diagnosis in time. We present a cost effective solution to replace up to 3 million rupee Fundus Cameras by a 20K lens and a simple smartphone!
Our proposed system ‘i-Fundus’ is a mobile phone app along with a clip-on attachment which enables us to take fundus photographs of posterior segment of eye which can be sent using mobile internet from remote areas to Armed Forces Institute of Ophthalmology for diagnosis and prescription. A trained personal will take fundus photographs using a smartphone and 20D Lens attachment to be sent onto the cloud for diagnosis.
The second part deals in the algorithms for diagnosis of diabetic Maculopathy using the artificial intelligence algorithms and last part will deal in the trend analysis to help ophthalmological research and ease most effected remote areas of the country by utilizing data acquired from AFIO and Lahore General Hospital.
​​Project Title iAgri – ICT Research Based Online Agriculture Portal 
​Group Members ​Muhammad Haseeb Sohail
​​Project Supervisor(s) ​​Dr Farhan Riaz (EME)
​Project Abstract ​This project uses the data mining, social networking and the latest e-marketing techniques to keep updated the farmers with modern agriculture techniques and news. This project also provide an easy and accurate solution to their problems in an easy way. As we know that 70% of land of Pakistan is arable and 60% of people are linked with this profession. So there is a need to provide them knowledge about modern agriculture techniques and latest technological solution to their problems by which they can get maximum yield from their crops. This project uses data mining techniques for extracting the agriculture related data from the internet. Farmers are get notified on their mobile phones with weather forecasting and pest ratio in their region and the methods which are to be adopted to cater far bugs/worms problems.
An online database is developed which contains the data of the users which are registered with our service. Registration of farmers can be done manually by using online portal or automatically by sending a SMS. The registered farmers then will be able to get the latest agriculture/dairy farming updates on their mobile phones in their native language i.e. URDU. This project will facilitate the farmers and will be very useful in creating a small social network of farmers where they will share their success stories of modern agriculture.
​​Project Title Mobile App for Enterprise Resource Planning (ERP)
​Group Members ​Muhammad Usman Arif
​​Project Supervisor(s) ​​Dr Farooque Azam (EME)
​Project Abstract ​This application is specifically designed for the EME College ERP System and it contains three main modules Students, Faculty Members and Admin.
By using this App Students could register themselves in college database students can check grades, attendance, marks of the quiz’s, assignments and sessional, and get notifications for short attendance, exams, assignment dead line and college events. Students can also register for the subjects and download course related material. Students could explore the Faculty member’s public profile and could take appointment for meeting.
Teacher use this App to mark attendance, enter marks and grades, search a particular student and can check E-dossier of the students.
Admin can use this app for Announcement, to watch Executive Summaries and for student evaluation.
​​Project Title Plan Ahead: Smart Traffic Monitoring System
​Group Members ​Talal Shoukat   ,   Muhammad Ali  ,  Usama Arif   , Noor ul ain Shuja
Asst Prof. Sajid Gul Khawaja (EME)
​Project Abstract ​Time is everything in this busy world. Sadly in Pakistan, with the number of cars increasing by the day, being stuck in traffic for hours has become in-evident specially during the office opening and closing hours and the increasing congestion levels in the traffic are a mere distress for the mind. Imagine a system through which you can automatically monitor traffic congestion at busy intersections and plan your journey accordingly. A similar system is proposed in this project which monitors traffic flow at intersections and tells users the congestion levels at their respective routes. This system intends to find the congestion in traffic flow using live video stream, which is taken from the camera deployed at the road side, as well as loop induction sensors. The video frames are pre-processed and features of moving objects are defined and detected, during the tracking process a filter is devised to detect the velocity of the traffic flow. In case of loop sensors, detectors attached to them automatically return the frequency of cars passing over the sensors. For ease of access for users, the results are sent to a centralized traffic database management center via a GSM module, and an application is developed, which uses these congestion levels from the database to mark it on a map and suggest the shortest and quickest route based on distance and congestion level. It also uses data mining for prediction of traffic flow before time. Emergency vehicles and accidents can be handled by communication through centralized GSM module.
​​Project Title Application Aware OSI L7 Router using NetFGPA 10GPublic Health Mapping and Planning for Balochistan
​Group Members ​NS Shah Alam Khan  , PC Manahil Ahmed Khan  , GC Ghulam Abbas
​​Project Supervisor(s) ​​Dr Shoab Ahmed Khan (EME)
​Project Abstract ​Improving routine immunization is one of the key indicators for ensuring a reduction in childhood mortality. Unfortunately many children are being needlessly affected by many of the otherwise vaccine preventable diseases. Despite the availability of safe and effective vaccines which are free for all, improving routine immunization still remains a major challenge in Pakistan. The project aims at improving the immunization by ensuring that the children are being given the proper vaccination in a timely manner. Through a web based system and an auto-scheduler, which will timely be notifying the parents through Short Message Services and at the same time will be keeping a check with the Public Health centers for the record of the child being vaccinated.  The results will be then mapped using google maps api for reporting and trend analysis.
​​Project Title SmartSteth
​Group Members ​Ali Haider  , M Daniyal Ashraf   ,  M Usama Azhar    ,  Syed Osama Maruf
​​Project Supervisor(s) ​​Dr. Usman Akram (EME)  ,  Asst Prof.Sajid Gul Khawaja (EME)
​Project Abstract ​Pneumonia kills nearly  2 million children around the world out of a total of 150 million children who are affected by it and about 98% of these deaths occur in developing countries.  The reason for high mortality rate for people who acquire pneumonia is because of a lack of trained personnel as the ratio of doctors to pneumonia patients is very low and another reason is late detection and diagnosis of the disease due to a lack of awareness among the masses.
The proposed system ‘SmartSteth’ dubbed as a portable doctor will be able to help a trained health worker in early detection of pneumonia as well as in its treatment.  A health worker will quiz the patient according to the standard of CRB-65 and then record the patients respiratory sounds by a device called ‘Steth-Mic’ which is just a stethoscope connected to the Smart phone via the audio jack.  The recording will then be uploaded to a cloud server where complex digital signal processing and machine learning algorithms will help in diagnosis of pneumonia with the same sensitivity and specificity as a human doctor.
The main aim of the project is to provide a cost effective solution to above mentioned problem.
​​Project Title Social Media Sentiment Analysis
​Group Members ​Pc Azwar Habib  , Ns Nauman Waqar   , GC Ahad Farooq
​​Project Supervisor(s) ​​Asst Prof. Sajid Gul Khawaja (EME)
​Project Abstract ​Facebook® is a social media service usually used as an instant communication platform. The capacity to provide information in real time has stimulated many companies to use this service to understand their consumers. In this direction, TV stations have adopted Facebook for shortening the distance between them and their viewers, and use such information as a feedback mechanism for their shows. The sentiment analysis task can be used as one such feedback mechanism. This Project corresponds to classifying a text according to the sentiment that the writer intended to transmit. A classifier usually requires a pre-classified data sample to determine the class of new data. Typically, the sample is pre-classified manually, making the process time consuming and reducing its real time applicability for big data. This project uses an automatic sentiment classifier for Facebook status updates. The automatic sentiment analysis reduces human intervention and, thus, the complexity and cost of the whole process.
​​Project Title Simulation of Cricket in Unity3d
​Group Members ​Irfan Abbasi
​​Project Supervisor(s) ​​ Dr Shoab Ahmed Khan (EME)
​Project Abstract ​The aim of the project is to create a prototype for simulation of cricket in unity3D. The game consists of characters modeled in state of the art 3D modeling tools. By capturing motion with Microsoft Kinect and editing motion files in motion building software, animations are applied to the humanoid characters. A combination of self-calculated mathematical model and Unity3D physics’ engine is employed to bring real time physics into the game. An AI system is modeled for intelligent decision making by the fielders and the player’s opponent depending on the role they are in. Furthermore, cutting edge image effects and lighting for graphics enhancement uplift the experience.
There are few cricket games for multiple platforms however the style of interactivity and the game play is monotonous.  To bring innovation into this project as a competitive advantage, the player will have the option to bat in first person or third person. In the first person mode, the player will be able to control the position and swing of the bat at real time to play shots. This will be attained through virtue of a Wii Remote.
​​Project Supervisor(s) ​​Dr Shoab Ahmed Khan
​Project Abstract ​This project develops a portable device implementing wireless technology and taking full advantage of the wide-spreading Internet to provide a convenient real time solution to monitor heart diseases and stroke .As heart diseases and stroke count up to 25% of all death in the world and 90% of the death can be prevented if timely medical actions can be taken. The health information acquired on the portable side transmits to the server wirelessly and can be accessed all over the Internet in real-time. All this enables people to know their risk of cardiac arrest themselves and notify the concerned doctor timely.
The data acquired from the ECG is uploaded wirelessly to the server regularly through the Beagle board. This information is processed through diagnostic algorithms of various diseases to detect that either patient needs urgent treatment or not. The doctors who can view the data online can send notifications to bring patient’s attention in case of serious situation. The importance of this project is that we are going to implement Network Enabled Cardiac Monitor System which has a special meaning for cardiac patients as they will be able to monitor their cardiac activity all the time and can ask for help in emergency situation. This device renders an affordable solution for bridging the gaps between doctor and patients.

​Group Members ​Mehreen Tahir , Sadaf Sahar , Rida Batool
​​Project Supervisor(s) ​​Dr Shoab Ahmed Khan
​Project Abstract A lot of research is being done on autonomous and self-driving vehicles in the top universities of the world, but there is a very high cost both time and resource wise to develop such vehicles because of hardware requirements. Not every institute can afford these costs, especially in developing countries like Pakistan. Our goal is to create an alternate development and testing environment that reduces the costs by replacing real life hardware with software simulations so research can be done without the need for hardware. In this project we aim to design software to develop and simulate these autonomous vehicles. The communications module connects all the other modules with each other and the hardware sensors (if any) and as such acts as a message broker. It also provides data monitoring,unified queue tracking,establishes connection between modules while providing decoupling and data transfer at high rate without any network  blockage or congestion.
​​Project Title Website Categorization using Artificial Neural Networks
(ANN) and Trend analysis of web traffic on a Network
​Group Members ​Muhammad Numair Mansur , Hassan Saeed
​​Project Supervisor(s) ​​
​Project Abstract
After the registration of the first ever URL on 15th March 1985, the number of users using Internet has
been increasing exponentially by the day. A rough estimate shows that the numbers of domains
registered between 1995 and today have gone up from a mere 15,000 to an enormous 350,000,000.
Different users use these websites for research purposes, communication on social networking
websites, advertising, and other economic and non-profitable purposes. Users use the Internet facility in
different ways. A teenager sitting at home on his/her laptop might be using Facebook and listening to
songs on You tube at the same time where as an adult might use the same computer later to order
clothes on line.
The aim of this project is to retrieve the URL which are requested using browsers from a web server
​​Project Title Face Recognition based Attendance System
​Group Members ​PC Amena Shahid , NS Sundus Javaid
​​Project Supervisor(s) ​​Dr Shoab Ahmad Khan
​Project Abstract
This project is about making an efficient attendance system using face recognition. The project is aimed at reducing the time wasted in the class for marking attendance. A camera is to be placed in front of the class that will take the picture of the whole class after every 10 minutes (for bigger scope we can actually design the camera to take pictures according to a specific schedule and do it automatically).The pictures will be then fed to the haar cascade identifier which will detect the faces from the picture. The detected faces will then be recognized and matched with the already maintained database. For 85% matches the attendance of the students in the picture will be marked present. The attendance will be marked along with the date and time when the picture is taken into our attendance system.
​​Project Title Fusion of Thallium scan and CT Angiography
​Group Members ​Intisar pervaiz , Aeman Nawaz , Madeeha Zafar
​​Project Supervisor(s) ​​Dr. Shoab A Khan , Dr Usman Akram (EME College), Dr Mohsin Saeed Sheikh (AFIC)
​Project Abstract We present a fast and fully automatic approach enabling the fusion of the main anatomical structures of the heart in CT images and left ventricle in Thallium Scanned images .The proposed method is based on the adaptation of segmentation of the heart in CT images and typical color level differences in Thallium Scanned images. CT angiography gives the location of blockage in an artery whereas thallium scan tells what effect that blockage produces in any area of the heart. The fused images will clearly show the relationship of relevant coronary arteries and the abnormal perfusion territory in all patients and will be useful for the evaluation of coronary artery disease. Instead of separately diagnosing the problem, this project will provide a fused image of both thus helping in quick and efficient diagnosis. Visual inspection by experts reveals that the proposed method of hybrid tool will immensely benefit biomedical field..
​​Project Title Show Room Management System
​Group Members ​NS Shaharyar Babar
​​Project Supervisor(s) ​​Assist Prof Wasi Haider
​Project Abstract  Now, when technology is shifting from manual and man dependencies to autonomous systems SMS automates all showroom processes. The purpose is to monitor all showroom activities, products, money transactions by this system. It will also calculate the credibility of customer on the bases of his credit history. Then on the bases customers data it will decide the credibility of new customer, for credit term business, on the bases of his occupation, residence and references. And give the result whether new/old customer is suggested for credit or not? Reports can be viewed from the system so that sold, current products and money transaction can be match against. Defaulter customers will be highlighted to take action against. Profit, lose and expense of showroom will be calculated in reports. So this  all will benefit from business, data security, and usage point of view. In future it can be extend with business requirements as desire.
​​Project Title Home Automation System Using Microsoft Kinect
​Group Members ​Bilal Ahmed  , Saad Nasir , Hassaan Bin Ali , Muhammad Usman
​​Project Supervisor(s) ​​Lec Sajid Gul Khawaja
​Project Abstract
This project is designed to save electricity costs and minimize the hassle involved when going all around the house to switch off or switch on the lights, fan or any other electrically operated appliance. The Kinect sensor will be placed in one corner of the room in order to have a maximum possible wide view. By using a libfreenect based application, the Kinect sensor will be able to monitor the movements of the person moving about in the room, using a set of simple rules to determine what action to take based on the gesture implied by the person. For example if the person is sitting at the desk, the table lamp will turn on and when the person moves away from the desk, the lamp will turn off. Similarly, the lights in the home will be turned off if the person exits from the main door and turned on when the person enters through the main door.
​​Project Title Intelligent Unit for Organization and Predictive Analysis of Medical Data
​Group Members ​Ns Masood Hussain (317)   , Ns Musa Bilal (323)  , Ns Syed M Azeem (341)
​​Project Supervisor(s) ​​Dr Shoab A Khan
​Project Abstract Medical informatics plays a very important role in the use of clinical data. In such Discoveries, pattern recognition is important for the diagnosis of new diseases and the study of different patterns found when classification of data takes place.
There is no proper mechanism present in the sustaining medical information systems. Due to this doctors have to rely on human analysis and decision making which can be suitable when analyzing 5-10 records but extremely difficult if we increase the number of records to be analyzed. Moreover it also causes inaccuracies in results and stress for the doctor.
The system we intend to develop aims to form a proper data warehouse; it intelligently stores the existing records and patient histories along with the incoming data for future. This is done with the help and training of IBM (International Business Machine) and using their popular tools. The system also aims at making the job of analysis as accurate as possible using analysis from a large number of records and medical test reports that we will maintain in an organized and intelligent database.
​​Project Title IT Trend Analyzer
​Group Members ​Shah poor khan , Santosh kumar  , Hafiz Muhammed Nouman shabbir
​​Project Supervisor(s) ​​Assist Prof Wasi Haider
​Project Abstract The IT field has become so vast that someone cannot easily choose the field, demanded most in the industry .The IT Trend Analyzer will make it easy for everyone. It is based on the Web scraping. First it will do the job of collecting data from different IT websites. Than the data will be parsed.  As a result of parsing the cluster is made, Cluster is than examined through the Data Mining techniques, for the analysis of finding the most adapted , demanded IT field in the market or Industry.  At last the result of the analysis through the DATA Mining Techniques, is displayed in form of Graph. This graph will show different  IT fields and their respective demand in the industry.
​Group Members ​NS AMNA BASHIR  (352) , NS AATIQA ANWAR  (349)
​​Project Supervisor(s) ​​Dr. Shoab A Khan  ,  Lec Sajid Gul Khawaja
​Project Abstract
The project constitutes of implementation of NATO narrow band waveform on SDR.
SDR, is a radiocommunication system where components that have been typically implemented in hardware are implemented by means of software on embedded system. A basic SDR system is equipped with ananalog-to-digital converter, preceded by some form of RF front end. Significant amounts of signal processing are handed over to the general-purpose processor and produces a radio which can receive and transmit widely different radio
waveforms.In our project we implemented NATO narrowband waveform that uses binary CPM modulation for wireless communication to transmit high data rates without the expense of high transmit powers and high bandwidth utilization. It transmits 6 voice channels with increased throughput, minimum BER and increased bandwidth.
​​Project Title Development and simulation environment for self driving vehicles
​Group Members ​Moazzam Abdullah Khan , Danial Khan , Ghazanfar Ali
​​Project Supervisor(s) ​​Dr Shoab A Khan
​Project Abstract A lot of research is being done on autonomous and self driving vehicles in the top universities of the world, but there is a very high cost to develop such vehicles because of dependence on costly hardware. Not every institute can afford these, especially in developing countries like Pakistan. Our goal is to create an alternate development and testing environment that reduces the costs by replacing real life hardware with software simulations so research can be done without the need for hardware. This project aims to design software to develop and simulate autonomous vehicles as they would behave in the real world and produce quantifiable results to develop and improve algorithms that drive and control a fully autonomous vehicle.
​​Project Title Real time object tracking in panoramic images
​Group Members ​Usama Shafiq  , Tanzeel-ur-Rahman , Daniyal Noman , Aamir Shahzad
​​Project Supervisor(s) ​​Sajid Gul Khawaja
​Project Abstract
The problem that has to deal is when the surveillance is taking placeusing multiple cameras. And from single camera there is limitation of coverage. And if there is anything suspicious then track it.The solution of this problem is video stitching. The goal of video stitching is to construct a single videooutput by stitching together multiple input video streams. The goal of project is to do the video stitching in real-time, so that it can be used for security purposes. The input videos are coming from the multiple cameras therefore the stitched video has large field of view. This can be done by finding the common region in the 2 videos and then merge these 2 videos along that region. The tracking is done by using the kalman filter. As all the processing is have to be done in real-time, therefore general purpose processor has not good use. Therefore we used the specific processor i.e beagle board.
​​Project Title Secure, High Data Rate Communication Model for Tactical Network
​Group Members ​Sameem Aslam , Muhammad Sami , Shahid Mehmood , Amir Jamil
​​Project Supervisor(s) ​​Dr  Shoab A Khan
​Project Abstract
Secure voice (alternatively secure speech or ciphony) is a term in  cryptography for encryption of voice communication over a range of  communication types such as radio, telephone or IP.The project proposes a Secure voice Transmission for which we make a high
speed engine which compresses and encrypts data that comes to it.  Encryption refers to algorithmic schemes that encode plain text into  non-readable form or cipher text, providing privacy. The receiver of the  encrypted text uses a “key” to decrypt the message, returning it to its   original plain text form. The key is the trigger mechanism to the  algorithm.  AES (Advanced Encryption Standard) Encryption Technique  is very  important for the securing of data. Its extension is the Rijndael  algorithm that will be used in our project. We will perform compression   using MELP/CELP algorithms  The actual scope of the project is Secure Communication in the war  scenario among The soldiers and the command and control systems. Tactical  Network Communication is the main stake holder of this project.Outcome of  the project is an efficient and secure communication with a high data  rate.
​​Project Title The WiDoc Monitoring
​Group Members ​GC Imad Ali Shah   — 385 , NS Syed AizazHussain Shah — 338 , GC Hassan Akbar   — 383

GC Syed Amir Hayat Shah  — 388
​​Project Supervisor(s) ​​Lec  Sajid Gul Khawaja
​Project Abstract
In our project “The WiDoc Monitoring” we would develop a wireless health monitoring system .The system will consist of a HW (embedded device), SW (server end application for monitoring and maintaining records) and establishment of communication between device and server.
The device may be used by any sensitive organization like law enforcement agencies for its personnel or by any factory for their workers, who are working in hazardous environment.
This device, on the basis of different input health parameters (temperature, heart rate, oxygen saturation) will inform the health status of the person and will communicate the location and parameters to the base station. In case of any fluctuations in the input parameters, all the parameters will be recorded by the ARM processor and incase of any anomaly, these parameters will be communicated to the server/base station.
​​Project Title A Multi-Agent Intelligent Autonomous Software System for RoboCup-SSL
​Group Members ​NS Usama Tariq , PC Muhammad Shahzad , NS Asif Mehmood
​​Project Supervisor(s) ​​Dr Shoab A. Khan
​Project Abstract
This Project aims to target Multi-agent Intelligent Electro-mechanical Systems (MIEMS), a relatively new field which focuses on making intelligent autonomous agents, capable of acting autonomously, to behave in a collaborative manner. The basic problem proposed is to apply the concepts of multi-agent autonomous systems to develop a team of robots,
which could effectively play a game of soccer in a real-time dynamic environment. Robot Soccer based environment is best suitable for multi- agent artificial intelligence because of its adversarial, collaborative and dynamic nature. For development, software system is further divided in multiple modules. Major modules include detection module, prediction module for tracking of robots using algorithms like Kalman filter and neural networks, strategy module for Intelligent strategy planning using quadtrees and genetic algorithm for a cooperative play, path planning for individual robots and radio module for wireless communication. Parallel processing techniques has been employed to cater for high computational requirements due to real time complex decision making.
​Group Members ​Waqas Shahid 326 , Hassam Sohail 369 , Munaim Ashraf 327 , Numair Zulfiqar 599
​​Project Supervisor(s) ​​Dr Saad Rehman , Dr Shoab A. Khan
​Project Abstract
The project proposes a web information retrieval system just like a mentor that refers to past, guides present and carves bright professional future.  A web portal with the most efficient, relevant and informative data profiling system in which the user obtains all the news relevant to his/her education, career, profession, interests at one place using automatic and intelligent dispatch system. The most sophisticated and intellectual first-ever automated recruitment portal integrated with all big professional and job informing portals. The concept of searching a job is replaced by an interview or test call. Subscribing to all job
portals is not required, the recruiters will find the candidate themselves. It is a huge issue for the HR hunters to seek for the best talent available for the post, resolving this problem this project provides the uniquely featured portal where the recruiters can fill in the specifications and in result find the list of candidates short listed automatically. Designing a new web subscription plug-in that provides the best data profiling and customized data for every user replacing the conventional RSS feed plug-in (which provides more irrelevant data).
​​Project Title Facial based automatic attendance system
​Group Members ​NS Azizullah Memon , NS Ghulam Akbar , NS Muhammad Shakeel , NS Teerath Das
​​Project Supervisor(s) ​​Dr Shoab A. Khan
​Project Abstract Facial Recognition system is a field of identifying a person from a facial database or from a live camera source specially used in security systems. The aim of this project is to provide an automated system where attendance of any institute or an organization will be done by using face recognition techniques within less time and efficiently. This system is going to be designed for automated attendance system using face recognition. This system takes input from a live camera (stationary or video) and will detect the faces from that image using various computer vision libraries such as opencv, emgu etc. The detected faces will be recognized from data base using eigen faces technique. On the basis of recognized images the attendance of the class will be marked.
​​Project Title Cognitive Surveillance System using Proximity sensors & PTZ IP Cameras
​Group Members ​Mavera Mazhar Butt  , Maria Khanum Bhatti
​​Project Supervisor(s) ​​Dr Shoab A. Khan
​Project Abstract Security threats these days have reached an alarming level, due to which there is a need of installation of robust and the most efficient surveillance systems in the country. To help ensure the best possible security over a particularly larger tract, a surveillance system with embedded proximity sensors and PTZ (Pan, Tilt & Zoom) IP cameras is designed. The system will make use of specialized proximity sensors which consume low power and cover a wider area, and based on the data (signals) provided by them, the central computer system will activate the PTZ cameras to lock the target. This technology is proficient as the speed of this special camera can be varied, it can pan (move left & right), tilt (move up & down) and zoom in, and capture key information that can be used to help in loss prevention. Tour functions can be programmed into the PTZ camera, enabling it to focus on a different door, window, parking area etc. For a minute’s interval, and repeating this tour over and over again.
​​Project Title Orthogonal Frequency Division Multiple Access(OFDMA) on SDR
​Group Members ​M.Talha Aftab , M.Zain Haider , Zohair Zahid , Omer Zia
​​Project Supervisor(s) ​​
​Project Abstract The goal of this project is to study the workings of OFDMA, by first understanding the concept of OFDM; then using that understanding to implement the modulation for multiple users on a MATLAB module. This will be done using a suitable algorithm of carrier allocation. After the MATLAB module is perfected, it will be converted to code which can be run on SDRs. The scope of the project also allows for the study of further techniques in how OFDMA can be optimized.
​​Project Title Fast frequency hopping for Anti-jamming
​Group Members
Nauman ahmed  , M. Majid sharif , Shahid rasool , Tanesh Kumar
​​Project Supervisor(s) ​​
​Project Abstract Fast frequency hopping is the method of transmitting signals that has many advantages over fixed frequency transmission. One of its major advantages is that it can be used for the purpose of anti-jamming. As in fixed frequency transmission signals can be jammed by adding noise to the signals. In fast frequency hopping, transmitter generates pseudorandom sequence of frequencies and it sends the signal over the channel. Receiver is synchronized by this pseudorandom sequence. The narrow frequency bands are generated by this method. Small bits of data are sent in the narrow band. Transmitter turns to another frequency and so on.
​​Project Title NOC (network on chip) based system controller
​Group Members ​Umer Hayat Khan , Hikmatullah , Idrees Ahmad , Jawad Ali Khan
​​Project Supervisor(s) ​​Lec  Sajid Gul Khawaja
​Project Abstract Network-on-chip (NoC) is generally regarded as the most promising solution for the future on-chip communication scheme in giga-scale integrated circuits. The main objective of the project is to install and communicate peripheral interfaces with Network-On-Chip (NoC) architectures. The design of this NoC based controller is to pass data from source device to destination device. The main peripheral devices are USB (USB 1.0, 2.0),VGA and ETHERNET. Implementation of this framework is on the FPGA and developing a user interface is one of the main deliverables of this project.
​​Project Title Analyzing popular discussions on social media
​Group Members Ali Shahbaz , Mashab Ali Javed  ,  Jahanzeb Khan  , Ali Liaqat
​​Project Supervisor(s) ​​Dr Ali Mustafa Qamar
​Project Abstract During the recent years, social networking sites such as Twitter (a micro blogging site) have gained immense popularity as a means for communication. Twitter allows users to post short text messages, also known as tweets, up to a maximum of 140 characters. It resembles SMS (Short Message Service) in significant ways: short message lengths, use of abbreviations, and capability of communicating with numerous recipients easily. An important behavioral element of a short text messaging network is that of popular discussions. These are the discussion threads or topics that involve a large number of messages and users over a short time period. Twitter has also gained popularity among the research circles and people have tried to analyze the Twitter data to perform different tasks like sentiment analysis and opinion mining [Pak and Paroubek], future prediction (revenue prediction for a movie, etc) [Asur and Huberman], classification of tweets [Sriram et al.]. The aim of this project is to analyze Twitter data (can be retrieved using Twitter API) and develop models for predicting popular discussions. We intend to utilize the message content as well as the network information in our models.
​​Project Title
IPV-6 10 Gbps Router Implemented on NETFPGA
​Group Members ​NS-Muhammad Salman  , NS-Shahi Dost , NS-Fazli Rabi , NS-Imran Ali
​​Project Supervisor(s) ​​Lec  Sajid Gul Khawaja
​Project Abstract In computer networks, data is sent in the form of packets. Routers are used to forward the packets from one network to the other. Since router`s processing is limited, so unsurprisingly packets have to spend a lot of time in router`s queue with the possibility of packet being dropped as well. To increase the processing time router was enhanced to 20Gbps with ipv4. Now we are going to implement the already enhanced router (20Gbps) with ipv6. As the number of users of internet is growing day by day and ipv4 is going to obsolete so that’s why it is necessary to implement it on ipv4. Another issue of ipv4 is that it is less secure than ipv6.
​​Project Title Flash OFDM
​Group Members

Subhan Bashir
Saqib Ejaz Awan
Shoaib Hashmi
​​Project Supervisor(s) ​​Lec Sajid Gul Khawaja
​​Project Title PEERCLOUD: A scalable multi-tenant cloud based framework for ERP systems
​Group Members ​Kush Kella , Shaneeb Kamran  , Khushnood Qadir
​​Project Supervisor(s) ​​Lec Wasi Haider Butt
​Project Abstract
Recent adoption of ERP systems by small to medium scale organizations where traditional solutions such as those offered by SAP and other companies fail due to budget, deployment and maintenance constraints, have given rise to various concerns. In situations such as these low maintenance software are required to serve for user needs. Moreover, shift to the cloud computing paradigm has enabled users to procure software as a service (SaaS) deployed on cloud servers that maintain both computation and data storage required by the application.
The two trends provide strong foundation for our project “PEERCLOUD: A scalable multi-tenant cloud based framework for ERP systems”. In this project we intend to develop a flexible ERP framework featuring multi-tenancy, customizable data models and application development. The target framework once deployed on cloud will cater for users needs by providing services in form of integrated applications based on data models and business logic designs.
We intend to build this application using JAVA and GROOVY, keeping in mind the efficiency requirements of the framework. Scalability of data storage will be an important aspect of the framework design because of the need to serve a large number of clients i.e. Massive scalability. In order to meet this requirement we intend to use scalable database solution such as mongodb or Apache couchdb.
On completion of the project we intend to achieve a scalable and customizable ERP solution hosted on cloud requiring zero maintenance on client side. It will cater for client needs by providing flexible services making use of basic features of the framework.
​​Project Title Hardware implementation of LDPC codes
​Group Members Sikander  , Rafeh  , Nadeem , Qasim
​​Project Supervisor(s) ​​Lec  Sajid Gul Khawaja