Final Year Project Title
Project Proposal
Supervisor
Umar Draz
Submitted by
Tooba Zoraiz
{BS-IT-M1-01-20}
Department of Information Technology,
University of Sahiwal.
[Date of Submission]
1. Project Title
AI Virtual Mouse using Python and Computer Vision
2. Introduction '
The AI Virtual Mouse project aims to develop a system that enables
users to control their computer cursor using hand gestures captured through a webcam.
This project utilizes computer vision techniques and machine learning algorithms to
interpret hand movements and translate them into mouse actions.
4. Problem Description
The transition from manual mouse control to gesture-based control
offers several benefits, including enhanced accessibility, convenience, and potential for
reducing repetitive strain injuries. However, challenges arise in accurately detecting and
interpreting hand gestures, ensuring robustness against varying lighting conditions and
backgrounds, and maintaining real-time responsiveness. Additionally, fault tolerance is
crucial to mitigate potential errors in gesture recognition and ensure a seamless user
experience.
5. Methodology
The project will employ computer vision libraries such as OpenCV for
hand detection and tracking. Machine learning algorithms, possibly including deep
learning models like Convolutional Neural Networks (CNNs), will be trained to
recognize and classify hand gestures. Real-time processing techniques will be
implemented to ensure low latency between gesture detection and cursor movement.
6. Project Scope
The project will focus solely on the development of the AI virtual mouse
system using Python and computer vision techniques. It will not address hardware-related
aspects or specific applications beyond basic cursor control. Assumptions include a
standard webcam setup and a relatively static background environment.
7. Feasibility Study
i. Risks Involved: Risks include inaccurate gesture recognition, high
computational requirements for real-time processing, and potential compatibility issues
with different hardware setups. These risks will be mitigated through thorough testing,
algorithm optimization, and user feedback. ii. Resource Requirement: The project will
require a computer with a webcam, sufficient processing power for real-time video
processing, and Python libraries such as OpenCV and NumPy.
8. Solution Application Areas
The AI virtual mouse system has applications in various industries, including
accessibility technology, gaming, and interactive presentations. It offers an intuitive and
hands-free input method, particularly beneficial for individuals with disabilities or those
seeking alternative interaction methods.
9. Tools/Technology
Python, OpenCV, NumPy, machine learning frameworks (e.g., TensorFlow,
PyTorch).
10. Expertise of the Team Members
The team possesses knowledge in Python programming, computer vision,
and machine learning, with relevant coursework and a shared interest in the project's
objectives.
11. Milestones
1. Research and familiarization with relevant libraries and technologies
2. Data collection and preprocessing for gesture recognition
3. Development of hand detection and tracking algorithms
4. Training machine learning models for gesture classification
5. Integration of gesture recognition with mouse control
6. Testing, optimization, and user feedback iteration
12. References
Book: Rosebrock, Adrian. "Practical Python and OpenCV: An Introductory, Example
Driven Guide to Image Processing and Computer Vision." 2nd ed., PyImageSearch,
2019.
Online Article: Brownlee, Jason. "A Gentle Introduction to Convolutional Neural
Networks." Machine Learning Mastery, 2019.
[https://machinelearningmastery.com/convolutional-neural-networks-gentle-
introduction/]
Internet Web Page: OpenCV Documentation. "OpenCV: Open Source Computer
Vision Library." Accessed Jan 28, 2024. [https://opencv.org/]
