Greeting & Message from

Principal NUST-SEECS
Dr. Muhammad Ajmal Khan, SI(M)
PhD (Michigan State University, USA)

It is a pleasure to introduce to you the graduating classes of The programs offered at SEECS include partnerships in Pakistan. Through Open House, the
2022 in the disciplines of Electrical Engineering, Software Electronics, Digital Systems, RF and Microwave, industry can witness the outstanding research of
Engineering, and Computer Science. Telecommunication and Networks, Artificial various disciplines that are being offered at NUST-
Intelligence, Machine Learning, Big Data, Cyber SEECS. At the same time, the industry feedback
NUST-School of Electrical Engineering and Computer Security, and Cloud Computing. Hands-on training helps us update our curriculum according to the
Science (SEECS) is committed to providing first-rate higher in these domains augments the basic knowledge contemporary market trends. Henceforth, I take
education in Pakistan. We emphasize making SEECS a of our students, giving insights into its practical immense delight in presenting the batch of 2022
center of excellence for imparting high-quality education application, an essential prerequisite for potential as the proud product of SEECS and wish them
in the areas of Electrical Engineering and Computer technical leaders of the 21stcentury. success as they go forward in their respective fields,
Science that would lead to the promotion of research with all the best for their journey ahead.
and scholarly achievements at National & International The projects showcased in the Open House
levels. We foster a passion for creativity and productivity demonstrate the skill set of our graduating
in our students through an enabling environment of state- students, and the highly interactive sessions with
of-the-art labs, arranging industry visits, seminars, and the industrial professionals provide them a platform
international conferences, etc. Besides imparting thorough for networking. Another aim of holding this event
professional knowledge, we also believe in instilling sound is to address the dire need for industry-academia
entrepreneurial, social and humanitarian values.

2 Abstract Booklet BESE 2022 3 Abstract Booklet BESE 2022

 Faculty Heads

Dr. Abdul Wahid

Tenured Associate Professor
PhD. Computer Science and Engineering
Kyungpook National University, Daegu, Rep of Korea
Head of Computer Science

Dr. Safdar Abbas Khan

Assistant Professor
PhD. Wireless Sensor Networks
University of Paris, France
Head of Innovative Technologies in Learning & Information Technology
Message from Associate Dean - Department of Computing

Dr. Faisal Shafait Engr. Dr. Rafia Mumtaz

Professor
PhD (Computer Engineering) Tenured Associate Professor
PhD, University of Surrey, UK.
Dear Graduating Students, Head of Software Engineering

I am very pleased to see you achieving a momentous millstone of your life. At Department of Computing, we aim
to produce graduates who not only have excellent professional and technical competence, but are also inculcated
with the norms and etiquettes, which are essentially required to successfully excel in the professional career. Our Dr. Muhammad Moazam Fraz
Tenured Associate Professor
responsibility is to well equip our students so that they contribute positively and resolve the issues in our society PhD
through their technical skills. We also polish the intellectual and entrepreneurial skills of our students and hence we Kingston University London, UK
believe that you will be able to make a difference in the market, as you embark upon the new journey. Head of Artificial Intelligence & Data Science

You are the flag-bearers now and we are very confident that you have the right knowledge-base and technical skill set
required to flourish in the computing industry. My advice would be to always welcome challenges in your professional Engr. Dr. Hasan Tahir
Assistant Professor
life with dignity, honesty and never forget that there is no replacement for hard work. So step ahead and make your PhD. Computer Security
parents, teachers and your alma mater proud! Wish you all the best for your future endeavors. University of Essex, UK
Head of Information Security
4 Abstract Booklet BESE 2022 5 Abstract Booklet BESE 2022
 Overview of

OPEN HOUSE

NUST SEECS organizes its annual open house to show case the skills of its graduating students. The idea
is to provide a platform where our students and industry representatives can mingle and have informal
or formal discussions. The students showcase their final year projects which represent their skill set categories. You will find a large diversity of projects including mobile applications, computer vision
and enable potential employers to identify any matching requirements. The projects are presented by based intelligent driving assistant, cloud based security resource sharing, an Alzheimer’s prediction
students from two main departments: application and many more. You will not only find potential employees but excellent ideas as well that
can be turned into products.
Computer Science (CS)
Software Engineering (SE) At NUST SEECS, we take pride in molding our bright entrants into well trained and appropriately
groomed professionals in Computer Science, Software Engineering & Electrical Engineering disciplines.
This year Department of Computing has divided the projects into different knowledge areas, including Our graduates are actively sought by the industry and our Alumni are occupying promising positions
Algorithms and Complexity, Architecture and Organization, Computational Science, Graphics and in some of the most prestigious industrial and business houses, both in public and private sectors. We
Visualization, Human-Computer Interaction, Information Management, Intelligent Systems, Platform- hope that you will enjoy the hard work of our students and find the right candidate or the next big idea
based Development, Programming Languages, Social Issues and Professional Practice, Software for your company.
Development Fundamentals, and Software Engineering. There are around 80 different projects in these

6 Abstract Booklet BESE 2022 7 Abstract Booklet BESE 2022

 INDEX Computer Science Abstracts

Sr. No PROJECTS Page. No

Algorithms and Complexity

1. Extensible CMS for Financial Modeling 10 21. Deep Learning for Self-Driving Cars for Collision Avoidance and Lane Finding 18
22. Mobile App for emergency services 18
Architecture and Organization 23. Recipebot 18
2. Tagging multimedia events for Content Advertisement 10 24. Automated Voice Dubbing System 19
25. Real time anomaly detection for security 19
26. Sahulat App - A Pakistan-focused Integrated System Connecting Citizens and Governing Bodies 19
Computational Science 27. SOP Violation Detection in Fenceless areas through Live Videos 20
3. Modeling, Implementation and Verification of Smart Doors 10 28. U-Assistant - Urdu speech based virtual assistant for mobile phones 21
29. UniD-Marketing 21
Graphics and Visualization
4. Analysis of reforestation potential in Pakistan using Machine Learning and Satellite images 11 Platform-based Development
5. Horror Game 12 30. Bridge 22
6. Virtual Shirt Try-On 12 31. Goal Quest 22
32. Personalized Education Platform 22
Human-Computer Interaction
7. Hotel Booking Website for Tourism 13 Programming Languages
8. Imecha 13 33. Integrated Development Environment (IDE) for Financial Modelling 23
9. On-Device Fruit Disease Classification 13
10. Real-Time Facial Recognition and Motion Detection 13
Social Issues and Professional Practice
34. Nigehbaan - Child Lost and Found 13
Information Management
11. Analytics based Mess Resource Management 14
12. COVID-19 Contact Tracing App 14
Software Development Fundamentals
35. Crime Alert: A Realtime Incident Reporting Web App 24
13. Location based Lost and Found System 14
14. Online Auction Service for Properties 15
15. Online marketplace for Vehicles with Data Analytics 15 Software Engineering
16. Rapid à la Mode Craze Scrutinizer (Trends Analyzer) 15 36. Alfabolt Analytics Platform 24
17. Structured Information Extraction from Text Documents 16 37. Perho Pakistan 24

Intelligent Systems
18. Chatbot based emergency hotline 16
19. Crop Disease Scouting using AIIOT Technology 16
20. CryptoCash (C2): Enabling instant transfers with digital payments and making investments
accessible through cryptocurrency 17

8 Abstract Booklet BESE 2022 9 Abstract Booklet BESE 2022

Knowledge Area: Algorithms and Complexity Knowledge Area: Computational Science
issues with the traditional doors and thus improves the reliability, security, efficiency and
Extensible CMS for Financial Modeling safety. Smart doors are one of the instances that human instinct wishes for, to acquire
solace and facilitate its daily life. They do so by the automated sliding entryway, which
Advisor: Dr. Hasan Ali Khattak has temperature sensors and a well-established security system tested thoroughly against
Group Members:
Muhammad Saad Jamil reachability, fairness, liveness and safety in the plan interaction. As an individual steps on
Ever wondered how businesses make key decisions for growth and sustainability?
Muhammad Talha Javed carpet, a signal is sent to the doors which notify the source to start as a result of which, the
Unless you have a finance background, this is impossible. But it shouldn’t be that
doors expand and lights are lit. The doors automatically close when the person has passed.
hard. This project is about enabling non-finance teams in a company to model
If an individual is standing between the doors, a safety sensor installed keeps the doors
assumptions about the business’ growth and participate in critical decisions.
opened. The lights are turned off when the person has passed. In addition to this, sensors
A financial model is not very different from a program. The process of writing and debugging
check the temperature of every person passing through the door for preventive measures
a financial model has a lot of similarities to writing and debugging programs in any high
from COVID. In case of a high body temperature suggesting a fever, an alert is displayed.
level language. This project will involve building an IDE for financial modeling with similar
Security system for theft protection is installed to assure that an alarm rings if anyone tries
capabilities that we have come to expect from programming IDEs. The exact design will
to break in. Monitored alarm system followed by a smart capture camera ensures safety in
determine the extent to which lexical analysis, parsing and ML based predictive insights are
cases if any unusual event is sensed by the other devices. In this project, we aim to model,
needed. This project will gain interoperability (between Financial Platform components),
implement and verify smart doors.
ensure an easy integration and will provide the environment for knowledge exchange.

Knowledge Area: Graphics and Visualization

Knowledge Area: Architecture and Organization
Analysis of reforestation potential in Pakistan using
Tagging multimedia events for Content Advertisement Group Members: Machine Learning and Satellite images
Advisor: Dr. Muhammad Shahzad Younis Amna Faisal Advisor: Dr. Muhammad Ali Tahir
Group Members:
Muhammad Deedahwar
Sarmad The main idea behind this project is to identify under forested areas (due to various
Idea of the project is to tag the keywords with audio/video streaming/stored content and Mazhar Qureshi
user be sent the alerts when the given keywords are mentioned in the program/content. This natural or human causes) in Pakistan and measure their reforestation potential.
App could be used for targeted advertisement of products/places/packages etc. While forests play an important role in mitigating the effects of climate change, the
process of reforestation is expensive, difficult to plan and even harder to execute and
maintain. Therefore, it is important to understand the efficiency of reforestation efforts
and develop solid estimates of areas where such efforts will actually yield results.
Knowledge Area: Computational Science Using satellite imagery, terrain, and weather data as an input to a machine learning
algorithm, this system aims to determine the potential vs actual vegetation cover. The
proposed algorithm is supposed to automatically identify geographical regions with high
Modeling, Implementation and Verification of Smart potential for future reforestation or afforestation efforts.
Group Members: Doors
Laiba Advisor: Dr. Sidra Sultana
Wadiya Gillani
Due to the advancements in the innovative local area and modernization, it is fundamental
for a human instinct to desire for more comfort and solace to his/her life. Smart doors
are the digital upgradation of the traditional doors having an ability to overcome the

10 Abstract Booklet BESE 2022 11 Abstract Booklet BESE 2022

Knowledge Area: Graphics and Visualization Knowledge Area: Human-Computer Interaction

Horror Game Hotel Booking Website for Tourism

Advisor: Dr. Muhammad Ali Tahir
Group Members: Advisor: Engr. Taufique ur Rahman
Group Members:
Natiq Ali Syed It is a horror game with FPS shooting type controls. This game is specifically for pc. The Muhammad Shahzil
Ramish Amir game can be made in Unity Engine. The player will have to get out of the hotel by solving There is a lack of availability of hotels on hotel booking websites in Pakistan, which does
Syeda Amna Jah Bukhari
Zaid Wajid riddles and surviving from the hotel creatures by killing them or by hiding himself. There not fully advertise the numerous, economical hotels and inns to tourists and travellers. Our
will be different quests for progressing the game. The main mission for this game is to website will help travellers to browse through multiple hotels and inns in various cities across
enhance the player analyzing abilities for different scenarios. Mostly horror games are just Pakistan, and find an economical place to stay. The website will provide a user-friendly
about solving the riddles or just surviving in the gameplay, but we will combine these two interface to both users and hotel managers.
parts to make it more interactive. Maximum games have simple 2d game menu, but we will
use 3D game menu as it will give user more interaction and the interest of the player will
increase. The players will be able to change the graphics according to their requirements. In
game, the player can find the tools for surviving like health kits or guns. Imecha
Group Members: Advisor: Dr. Seemab Latif
Virtual Shirt Try-On Adeen Jawad Qazi
Advisor: Dr. Zuhair Zafar Ali Tariq Our e-commerce system is built for a running business of hand painted khussas with the
Group Members: brand name ‘imecha’ where the customers can design their own khussa and place orders
Hassan Javaid Ayna Shairi
Augmented reality is on the rise today making its way into all of life from health care online. It is being built for the convenience of customers as an option of choosing the
Hassan Khalid Butt and education to entertainment. The provision of more data and recent advancements in design themselves will be available. The main goal of this system is to expand the customer
Hissaan Ali Shah Computer Vision and Deep Learning algorithms has made it even more easier to develop audience and help them find better quality products.
cutting edge technologies. With Loreal Paris and Nike already releasing their technology for
facial makeup and shoes try-on respectively, we intend to develop a try-on system for men’s
clothing. Given an image of a target person and a target clothing, our try-on technology On-Device Fruit Disease Classification
would seamlessly transfer the desired piece of garment style onto the target person. This
technology would also include geometric transformations to adapt to the body figure of the Group Members: Advisor: Dr. Hasan Ali Khattak
target person and generate more realistic results. Anber Arif
Safi Ur Rehman Automated system to getting information from the video stream to classify fruit diseases
using image processing

Real-Time Facial Recognition and Motion Detection

Group Members: Advisor: Dr. Mehvish Rashid
Abdul Hannan
Fahad Imran Facial recognition is one of the fastest growing AI technologies with vast applications in the
Faizan Riasat field of security, marketing, and even medicine. This project focuses on the security-related
aspects of facial recognition to create a robust solution for monitoring and surveillance
aimed at organizations. It relies on computer vision and deep learning to detect and identify
suspicious personnel using a centralized database. A web-based dashboard and mobile
application will be developed for real-time monitoring, generating live alerts, and interfacing
with the algorithm.
12 Abstract Booklet BESE 2022 13 Abstract Booklet BESE 2022
Knowledge Area: Information Management Knowledge Area: Information Management

without it. They user can choose if they found an object or are looking for a lost object.
Analytics based Mess Resource Management If they found an object, they can provide the necessary details along with images. They
Advisor: Dr. Muhammad Ali Tahir can also share their location. They must provide their contact information as well. The
Group Members:
Abdullah Sheikh integrated Firebase database will store and handle all the data and Google Maps will handle
A dining hall management may charge hostel students regardless of whether the mess is
Bilawal Ahmed the location functionality. Similarly, if someone lost a valuable object, they can check up
actually being used or not. This causes hostelites to over expend their already limited budget
Danish Khan on the app by searching or by looking up the map to view all nearby entries. They can also
and also wastage of extra amount of food being prepared. For this purpose, the proposed
easily contact the person to retrieve their property. Finally, after verifying the owner and
Mess Management System would greatly help the students in terms of managing their
retrieving the object, the entry inside the database can be marked as delivered and will be
budget effectively and efficiently. On top of that, the service would yield benefits for the
removed.
management team as well as it would provide feedback from the hostelites and help improve
the service even more based on the collected data. Also, the students will be able to manage
their meals from their cell phones which will minimize the effort of visiting the warden to Online Auction Service for Properties
get the mess off. Advisor: Dr. Syed Taha Ali
Group Members:
Umair Imran Abbasi An online portal to run auctions for properties
COVID-19 Contact Tracing App
Advisor: Dr. Farzana Jabeen
Group Members:
Fatima Khan
Online marketplace for Vehicles with Data Analytics
The recent outbreak of COVID19 has taken the world by storm, forcing lockdowns and
Advisor: Dr. Mehvish Rashid
placing a strain on public healthcare systems. With significant amount of asymptomatic Group Members:
carriers the disease has been hard to detect. Testing of population at large scale have Ali Hasnat Khan Cars marketplace + data analytics is an online marketplace for selling and
proved to be challenging due to limited testing kits and other issues. To tackle this Muhammad Ahmad Bilal purchasing cars that shall use advanced data analytics tools and techniques to
problem government and healthcare workers use contact tracing of infected individuals Usama Ejaz make the user experience better. Our project shall also use advanced Machine
to identify those who may have come in contact with them known as primary contacts. Learning techniques to predict the price of automobiles based on certain features.
These contacts are then quarantined or tested depending upon their symptoms. Our project will consist of two parts. A marketplace where sellers will be allowed to place
Traditional methods of contact tracing are erroneous and difficult. A large fraction of the ads of used vehicles and buyers will be able to contact the sellers. Such marketplaces already
population owns smart phones. The idea is to develop an app that uses Bluetooth technology exist but we will introduce the predictive analytics part as well to our project. We will use
to record close interactions between the users. Thus, the contacts of covid victims can be machine learning and predictive analytics techniques to guess the price of vehicles based on
traced using this app. This can reduce the time required for contact tracing dramatically and various key factors. This will prevent sellers who set unrealistic prices.
mitigate the spread of virus.
Rapid à la Mode Craze Scrutinizer (Trends Analyzer)
Location based Lost and Found System Group Members:
Advisor: Dr. Farzana Jabeen
Advisor: Mr. Jaudat Mamoon Muhammad Uzair
Group Members: In today’s world, those who get a hold of trends and latest developments are the quickest are
Ahmed Hassan Ismail The project is a mobile application where the users can list any found object by providing Saud Kamran the winners. Facebook, YouTube, and other social media websites are hotspots for finding
Hamza Amjad details, images and location. We can also view all the listed objects nearby or search for any out what everyone’s talking about. The idea is to create a tool which will mine these websites
Zain Javed item and see if their details match anything we might have lost and contact the person directly. and get a hold of key developments taking place. Finding the right information quickly is
Often times we find something valuable but are unable to decide how to locate the owner. crucial for understanding the latest developments. Through web scraping and data mining
Similarly we might lose something. Almost anyone can encounter such problem and it can be very techniques the latest trends can be compiled into an easy to read format for a beautiful
troublesome. Based on our current location, the system effectively lists all the finds nearby. With analysis. Natural Language processing techniques will be used to identify and extract
the Lost and Found system we can list any object and make sure it’s rightful owner can receive it. relevant information, including insights which may have remained unknown.
The project will be a Java application for mobile devices. The user could sign-in or continue
14 Abstract Booklet BESE 2022 15 Abstract Booklet BESE 2022
Knowledge Area: Information Management Knowledge Area: Intelligent Systems

(iii) rust susceptible. The trained models will be uploaded on the cloud and will also be
Structured Information Extraction from Text installed on the edge device (Agri-DMaP). For this purpose, two Amazon Web Services
(AWS) will be used including (i) AWS IoT Core which will act as a bridge between the
Group Members: Documents cloud and the edge devices, (ii) AWS IoT Greengrass which allow the deep learning models
Abdul Rafey Advisor: Dr. Mehvish Rashid to retrain on the cloud after a specific time and run these trained models on the end devices.
Hammad Ahmed Additionally, a mobile application along with a web portal will be developed to facilitate the
Laraib Arjamand The project is based on information extraction. Information extraction is the process of agricultural user to see the area under rust attack, its severity levels, and recommendations
extracting the structured information from the unstructured textual data. In information to apply treatments (Pesticide) accordingly to the disease level. To increase the outreach
extraction system we can build a system that extract data in tabular form, from unstructured of the solution a mobile application will be developed to assist agricultural community to
text. For example, given a receipt, information extraction task would be to identify a Date, timely detect, diagnose the severity level of the disease. The affected area of the wheat crop
Name, Amount, Company Name etc. And store it in a structure, like table, file, database will be annotated by the severity levels (such as level 1, level 2, and so on). This information
etc. This task can be done by applying deep learning algorithms to text data and choosing will be available both in audio/visual format supporting Urdu and English languages.
the optimal algorithm. This solution will offer many advantages such as containing the spread of rust disease by
detecting it in early stages, which leads to increased wheat yield, and spending money on
the right pesticide will help the farmer to protect his crop yield.
Knowledge Area: Intelligent Systems
CryptoCash (C2): Enabling instant transfers with
Chatbot Based Emergency Hotline Group Members: digital payments and making investments accessible
Group Members:
Advisor: Dr. Muhammad Ali Tahir Eesha Arif through cryptocurrency
Maaz Bin Tahir Saeed Advisor: Dr. Seemab Latif
Bushra Muqaddas Chatbot applications are the latest inventions of digital design. These applications are well
Maryam Zafar known for automatic conversational agents that run on computer programming or a kind of CryptoCash (C2) is the first truly digital wallet which enables instant transfers
artificial intelligence (AI) interaction between the users and machines with the intervention with digital payments and makes investments accessible through cryptocurrency. C2
of Natural Language Processing (NLP). In this project we aim to develop a Voice Based allows its users to go cashless by scanning QR codes available at partner merchants
Telephonic Chatbot that responds to emergency calls and generates corresponding alerts and for the first time ever, make payments through cryptocurrencies. Fiat payments are
and actions using Machine Learning and Natural Language Processing techniques. supported through interoperability between fiat currency PKR and stable coin vPKR
(virtual PKR). Since the application is built on blockchain and smart contracts, it ensures
Crop Disease Scouting using AIIOT Technology unbreakable security and greater transparency than traditional financing platforms. In
phase 1, C2 wallet users will be able to send money to anyone, anywhere in Pakistan
Advisor: Dr. Rafia Mumtaz
Group Members: and make digital payments at supported merchants. In phase 2, C2 wallet will make
Hassan Raza Bukhari Rust is considered the most harmful fungal disease for wheat which can cause tremendous investments accessible, allowing anyone to invest with it while keeping a low barrier for
Salman Inayat damage to the crop. With a continuous rise in population, it is essential to increase the crop entry and removing any middleman like banks, brokerage houses etc. The investments
yield to combat imminent food security. Towards such an end, we propose an IoT and AI- will be pooled from its users and use a number of mechanisms such as staking,
based solution for rust disease detection and prediction in its early stages. The system will stacking, trading algorithms etc. to earn a competing annualised return on investments.
consist of a camera connected with a Single Board Computer (SBC) that will capture the C2 is built as a human centered finance service. At the end of the final year
high-resolution images of wheat and send them to cloud. To detect the rust disease and project duration, C2 will aim to have completed a case study with one of the partnered
predict its severity levels, deep learning models such as Region-based Convolutional Neural merchants where C2 will be deployed in production and users will be able to make QR
Network (R-CNN) will be trained on the wheat images to perform image segmentation, payments through cryptocurrencies.
background extraction, and detection of rust disease. The model will further be trained to With its success, C2 will act as a cornerstone in laying down the foundations for
classify the detected rust disease into different classes including (i) healthy, (ii) rust-resistant, digital payments in Pakistan and solve the problem of lack of a good payments service
for online businesses in Pakistan.
16 Abstract Booklet BESE 2022 17 Abstract Booklet BESE 2022
Knowledge Area: Intelligent Systems Knowledge Area: Intelligent Systems

Deep Learning for Self-Driving Cars for Collision Automated Voice Dubbing System
Group Members: Avoidance and Lane Finding Group Members:
Advisor: Dr. Rafia Mumtaz
Ali Asad Advisor: Dr. Muneer Ahmad Amna Abid Video Dubbing has become an essential part of our culture now. However, it requires a lot
Awais Qamar Zainab-Binte-Iftikhar of time, effort, and skill to dub a video in your native language. Automated Voice Dubbing
Faiz-Ul-Hassan Self-driving cars are one of the hottest areas of research and business for tech giants. System is a platform to translate videos in the English language to our native language
The idea of autonomous cars originates by witnessing accidents due to careless driving i.e., Urdu. And this process will be carried out through a web application. The subtitles
of people which could become extremely harmful. They also impose serious challenges, of an English video are extracted, translated into Urdu, and then synthesized to speech
not only on technological but social and ethical grounds as the global average of road car automatically.
fatalities increases to 18.2 deaths per 100,000 people. This project will focus on two things
i.e., driverless cars and safety of self-driving car users. This will be achieved by using deep
learning and computer vision techniques to design an algorithm for detecting lanes on the Real time anomaly detection for security
road that allow vehicles to drive and change lanes automatically while avoiding collision
with other vehicles and obstacles on the road in order to prevent road accidents and improve Advisor: Dr. Muhammad Ali Tahir
Group Members:
passenger safety. Abdul-Basit Chishti Nowadays, there has been a rise in the amount of disruptive and offensive activities that have
Atisam Ul Haq Ijaz been happening. Due to this, security has been given principal significance. Public places
Muhammad Ahrar Ahmed
Mobile App for emergency services like shopping centres, avenues, banks, etc are increasingly being equipped with CCTVs
to guarantee the security of individuals. Since constant observation of these surveillance
Advisor: Dr. Muhammad Ali Tahir cameras by humans is a near-impossible task. It requires workforces and their constant
Group Members:
Faizan Ahmed An android app for a centralized helpline like 911. The user shall be able to write his/her attention to judge if the captured activities are anomalous or suspicious. Hence, this drawback
Talia Hamid emergency or enter it through voice. The audiobot will get the necessary information and is creating a need to automate this process with high accuracy. Moreover, there is a need
will inform the concerned department. The user will be shown emergency tips to apply until to display which frame and which parts of the recording contain the uncommon activity
the help arrives. The app will be user friendly. which helps the quicker judgment of that unordinary action being unusual or suspicious.
Therefore, to reduce the wastage of time and labour, we are building anomaly detection
system. Its goal is to automatically identify signs of criminal acts in real-time. From there,
Recipebot we can raise a detection alert for the situation of a threat, indicating the suspicious activities
Advisor: Dr. Rabia Irfan at an instance of time.
Group Members:
Afia Abdul Rehman Technologies like Internet of Things (IoT) and Artificial Intelligence (AI) has completely
Mariam Batool changed the ways of looking at things at home and with these technological advancement Sahulat App - A Pakistan-focused Integrated System
the concept of smart homes and automation is now a realization. Cooking is an activity
Group Members: Connecting Citizens and Governing Bodies
that has been a part and parcel of living in every home. Kitchen appliances and gadgets are
Ammar Junaid Abbasi Advisor: Dr. Yasir Faheem
getting smarter day by day and provide support to modernize cooking in different ways. One
Salman Nazir
thing which is less being thought is of recipe management. Recipes books, collections, recipe
apps are nowadays freely available. With which the user can try new and different recipes. Reporting domestic and municipal issues has never been easy for the citizens of Pakistan.
These recipes can be watched, printed and can be displayed on the devices’ screen while They have to undergo a long procedure to report problems such as waste disposal, sewage
cooking. Realistically it happens that a person cooking keeps on looking at the recipe while leaks, short water supply, etc. Even though some bodies accept complaints via call, there is no
cooking again and again, which could even be dangerous. In this project we thought to take guarantee that reported issues are taken seriously. Most of the complaints go unanswered and
the concept of AI based chatbot to a newer level and with voice/speech recognition support unresolved because of lack of governance and progress-checks in the local governing bodies.
can provide the user a recipebot that can communicate and provide relevant content through To facilitate this complaining procedure, the aim of this project is to develop an integrated
speech/text/video support to act like a assistance during cooking. system that lets citizens lodge complaints against daily-life disturbances. A mobile application

18 Abstract Booklet BESE 2022 19 Abstract Booklet BESE 2022

Knowledge Area: Intelligent Systems Knowledge Area: Intelligent Systems

is proposed which provides an easier way for citizens to report problems, with the option
of remaining anonymous. The problematic zones can be geo-tagged from smartphones and
U-Assistant - Urdu speech based virtual assistant for
the cameras within them can provide visual proof of the problem. The complaint is stored Group Members: mobile phones
in the central server, and is sent to the concerned authority depending upon the nature of Abdul Rafay Ahmad
the problem. A web-based application is also proposed for the governing bodies. This web- Advisor: Engr. Taufique ur Rahman
Muhammad Hamza Amir
application intends for higher authorities to keep progress checks over lower authorities.
We plan to make a Google Assistant/Siri-like voice-based assistant application that
Infrastructural problems in specific regions are better highlighted to authorities in that same
understands Urdu. The target audience will be Pakistani common users who may not be
region via the web portal. The system overall ensures that a lodged complaint is assuredly
well-versed with technology and the English language. Our application will provide video/
considered by the authorities, all while the progress on said issue is presented to the citizens.
image tutorials to help the users to understand the usage of our application. Our assistant
will assist the users to perform common tasks on the phone as well as get information easily
SOP Violation Detection in Fenceless areas through through Urdu voice-based question answering. The general idea is that users can perform
tasks by issuing a voice command in Urdu and the AI assistant will respond in Urdu as
Group Members: Live Videos well. For example, a user can set the alarm, ask the weather, set the caller ringtone, set the
Ayesha Siddiqua Advisor: Dr. Muhammad Ali Tahir wallpaper, ask to save a contact on the smartphone, make a call, send a message, and much
Mariam Ali more. These are the features we thought of but we will do more requirement gathering by
In the current wake of a global pandemic and community spread of the virus, government asking the common people since they are the focus of this project. We will also train the
officials and administrators have found restricting population movement and social currently available deep learning model and our focus will be on developing a user-friendly
interaction, a mammoth task. The high-priced schemes and lockdown containment measures application.
are still questioning the authorities regarding the relentless and ill-informed actions of the
public such as the omission of social distancing, overcrowded public places, non-masked faces
and unnecessary gatherings. With such contagious disease and regular SOP violations, the
UniD-Marketing
government and local authorities desire a non-pharmaceutical alternative since herd immunity Advisor: Engr. Taufique ur Rahman
Group Members:
can only be achieved if about three-quarters of the population contract the virus vaccination. Abdul Rafay Ahmad With the recent boom in digital marketing, it is increasingly difficult for
To curb the effect of amid breach of SOPs, we propose an intelligent system that will ensure Muhammad Hamza Amir business owners to keep track of the heuristics of advertising on the
the detection of SOP violations through various imagery sources such as CCTV footages.
different social media platforms available. To facilitate managing
The technology behind the proposed system is connected to machine learning and computer
different accounts catering to contrasting audiences, our platform will
vision for the human-level understanding of digital graphics. The captured and recorded
be a one-stop solution for business owners who need to manage their
CCTV images will be at first categorized in discrete classes of the same features. But before the
online marketing campaigns. With the APIs exposed by social media
classification, we must have a granular understanding of the image objects such as suspects,
giants such as Facebook and Google, we aim to streamline the process
the in-contact device, masks and much more. Thus, at this point image segmentation will be
of marketing a business owner’s product efficiently.
performed for pixel-masking of image objects using TensorFlow and OpenCV. TensorFlow
will detect the prominent humanoid violators through their extracted features such as faces,
physical distance, number of appearances per frame and other constraints. Whereas, for
live motion detection OpenCV will be used for tracking down the compliance measures.
The project will be advantageous in the sense that it will ensure the identification of illiterate
and violating community, digital assistance to the authorities and even medical departments,
business workplaces and public places. We expect that the project will be focused more on
developing a mitigating and extensible system that will assist the government officials and
businesses shortly for COVID-19 precautions.

20 Abstract Booklet BESE 2022 21 Abstract Booklet BESE 2022

Knowledge Area: Platform-based Development Knowledge Area: Platform-based Development

due to mentoring and more than 4 in 10 workers who don’t have a mentor say they’ve
Bridge considered quitting their job in the past three months. Moreover, people with autism and
Advisor: Engr. Taufique ur Rahman dyslexia can never be treated without one on one mentoring sessions. Seeing this great
Group Members:
Anum Mujahid importance of one on one mentoring yet one cannot find mentors easily is something that
This project is aimed to develop a cost-effective web application that would assist a novice
Fatima Khalid we are lacking. We aim to create a platform that simplifies the process of connecting with a
user to easily share and schedule posts for multiple social media platforms like Instagram,
Ghazala Bibi domain expert. In short, the client can easily search his/her domain and can find the related
Facebook, Twitter etc. It is primarily aimed to assist business enterprises of any size/
domain experts that he can connect with and get one-on-one training/advice. In this case,
magnitude and content creators to handle their content(s) effectively and conveniently on
the expert will be paid for his services and the client will have to pay for availing that service.
multiple social media platforms from a single platform, named as “Bridge.

Goal Quest Knowledge Area: Programming Languages

Advisor: Dr. Arsalan Ahmad
Group Members:
Bilal Waheed People often struggle when it comes to finding a path or a roadmap that can help Integrated Development Environment (IDE) for
Malik Muhammad Idrees them in achieving their goals. Most of the time they don’t precisely know what
Muhammad Nafees Tariq they can do to achieve their objectives. Primarily, this is because there is no proper Group Members: Financial Modelling
counseling system to help them. It would be helpful if there was a system or a person Hamza Ijaz Advisor: Dr. Seemab Latif
that could provide students and young people with a path to reach their goals. Syed Muhammad Ubadah
Our system aims at providing said path. Data would be gathered from different resources Jafry A financial model is not very different from a program. The process of writing and debugging
and that share job posts and their required skills. From that, our system will generate a a financial model has a lot of similarities to writing and debugging programs in any high-
roadmap that could be used to acquire those skills. In addition, the idea incorporates a level language. This project will involve building an IDE for financial modelling with similar
system to gather content from the web, that can be provided to users to gain skills, as a part capabilities that we have come to expect from programming IDEs. The exact design will
of the roadmap. The content would be filtered and shaped according to the knowledge determine the extent to which lexical analysis, parsing and ML-based predictive insights are
level of users. Consequently, a wide range of people can benefit from the system. needed.
The system would provide a rough timeline in which any skills can be gained. Timelines
could be made specific to users based on their status and routines. All in all, the system Knowledge Area: Social Issues and Professional Practice
would aim to provide a road that leads to the users to success.

Personalized Education Platform Nigehbaan - Child Lost and Found

Advisor: Dr. Arsalan Ahmad Advisor: Dr. Mehvish Rashid
Group Members: Group Members:
Alina Shoaib Qureshi The literacy rate of Pakistan is around 60% with around 64% of youth being under the age Hassaan Ali Mehmood
The main objective of “Nigehbaan” is to provide a valuable software application for the
Muhammad Asim Khaskheli of 30. It means we are missing out on a potential part of our population and with the Jalees Mukarram
society. A great cause has been attached with “Nigehbaan” which is to find the lost children in
pandemic, we have seen how it has impacted education and opportunities among the Mamnoon Ul Hussnain
Pakistan. Each year thousands of children are lost in Pakistan and no trace is found because
youth. Almost all developed nations promote one on one mentorship for youth so that these kids are either unable to give their home details or they are kidnapped. “Nigehbaan” is
they can be accelerated in the right direction. According to statistics, Mentees are an idea to recover such children. “Nigehbaan” will connect lost children with their parents.
promoted 5 times more often than those without mentors and the mentors themselves Parents will be able to upload details of lost children and the system will try to find them
are 6 times more likely to be promoted. Furthermore, 87% of mentors and mentees feel out of those who were missing and found by someone in the country. The system will try to
empowered by their mentoring relationships and have developed greater confidence. If found the lost faces across the country and public points using AI technology.
we talk about businesses, then 67% of the businesses reported an increase in productivity

22 Abstract Booklet BESE 2022 23 Abstract Booklet BESE 2022

Knowledge Area: Software Development Fundamentals

Crime Alert: A Realtime Incident Reporting Web App

Advisor: Dr. Mehdi Hussain
Group Members:
Hamid Ayub A web application that ensures the safety of general public. A crime alert based web
Muhammad Umer Farooq application works similar to a social networking app such as WhatsApp, Facebook or
Instagram but only crime alerts. People who see crime in their area can post about the
incident in real time at this platform, and everyone registered in the app in that area will be
notified of the crime. This app is capable to add features such as live sharing space, photos,
and videos to the web application platform.

Knowledge Area: Software Engineering

Alfabolt Analytics Platform

Advisor: Dr. Seema Jehan
Group Members:
Haashir Bin Shakeel The goal is to build a SaaS platform to enable developers to track all users in their mobile
Muhammad Ahmad Ejaz / web apps. Along with users, we want developers to be able to track all events within
Muhammad Mobeen their application. Each event could be linked to another user and we want to be able to
track links as well. We also want to be able to allow users to trigger secondary events
automatically based on an event e.g. if a user triggers a “message sent” event, they might
also want to auto trigger a “message received” trigger incase they are looking for users that
got the maximum number of events over a given time period.
Using this data, we want to offer our users:
Funnels:Tracking a series of steps and examine percentage of people dropping off at each
step.
Flows:Tangential analysis.
A-HA moment finder: Find the action or actions that drives customer value.

Perho Pakistan
Advisor: Dr. Khawar Khurshid
Group Members:
Abdul Mughees Awan Solving one of the most critical challenges faced by Pakistan: 25 Million out-of-school
Babar Hussain children.
Hammad Ali A social platform to provide education opportunities to all the children through
personalized citizen sponsorship. Each sponsor can keep track of the progress of his
sponsored student.

24 Abstract Booklet BESE 2022