CCTV based Face Recognition System Using Machine Learning

In this busy life Parents are very careful ??about their children's security. They don?t take any risk for their children. The reported annual incidence of violence against children ranged from 1.57 to 3.18 per 100,000 for Pakistan. Reported cases of child abuse increased by 11 percent in the year 2

Project Title

CCTV based Face Recognition System Using Machine Learning

Project Area of Specialization

Artificial Intelligence

Project Summary

In this busy life Parents are very careful ­­about their children's security. They don’t take any risk for their children. The reported annual incidence of violence against children ranged from 1.57 to 3.18 per 100,000 for Pakistan. Reported cases of child abuse increased by 11 percent in the year 2018 compared to 2017, with more than 10 children suffering some form of abuse every day in Pakistan last year, a report released by the NGO Sahil has revealed. Moreover, according to NGO Sahil girls were found to be more vulnerable between ages 0-5 and 16-18, and boys were found to be more vulnerable between ages 6-10 and 11-15. Nowadays students bunk their classes and leave colleges, some students don’t go their college or school after leaving their home and spend their time in different places and their parents are free of the stress that their children are in school or colleges.

In this model, the primary objective is to develop a Machine learning-based architecture using Haar Cascade Algorithm (HCA) for the face recognition of students to protect students for their parents. Standard benchmark datasets will help in fine-tuning the system. Moreover, the coordination with local institutes and industries in Pakistan will be an integral part of the research which shall facilitate the new issues of organization related to student security in the early diagnosis of security purpose. State-of-the-art Machine learning methods will be employed to achieve the high accuracy of this HCA system. The secondary objective of the proposed implementation model is to create awareness among people about this issue. The development in this area will improve the student security facilities of all Pakistani institutions.

Project Objectives

The purpose of this project is to develop and implement a face recognition system is to help the students’ Parents to upgrade their school/college time activities.

• To develop a student’s college security system based on face detection.
• Parents whom children haven’t reached college or entered or left the college whenever their child enters, leaves, or didn’t get to college so they will receive an alert.
• To generate reports regarding student attendance to assist the lecturer/staff in analyzing and tracking student attendance.
• A secondary purpose to carry out this system is to overcome the stress of Parents and School/college management because they are so afraid about children's security.

Project Implementation Method

The primary goal of this research was the practical employment of these state-of-the-art developed using today's most advanced techniques: Haar cascade. Haar is the most commonly used Machine Learning architecture. Haar Cascade have the unique capability of feature learning, that is, of automatically learning features based on the given dataset. Our system is then able to use computer vision and a deep neural network to find a prospective face within its stream. We used standard datasets for benchmark purposes which help out to tune this system. I used my pictures and those of 2 of my group members’ pictures to make a dataset of 10 subjects with about 10 pictures each. Our purpose is to make a facial recognition system that needs as little training data as possible. The main reason behind this constraint is the fact that it is more useful for a supervisor to have trained the model with one or few pictures for each student rather than having to make a large Dataset with many images for the same person. The secondary objective of the proposed research-based implementation model to overcome the stress of Parents and School/college management because they are so afraid about children's security. The research and development in this area will improve the computer vision of Pakistan.

Benefits of the Project

• This system will also helpful for the usage of college student security can use their existing infrastructure.
• This will reduce the workload of our employee management system.
• Comparison to the picture provided in admission time to the real-time face about more than 80% to 85% match gets marked.
• It is used to verify that when the student entered college is to help to recognize the student of that college.
• This will also help to overcome the stress of Parents and School/college management because they are so afraid about children's security.

Technical Details of Final Deliverable

Figure: Block Diagram of proposed Model

In our proposed model, we will use a Machine learning architecture, we take students to video and detect faces from this video and using anti-spoofing techniques to select high-resolution frame then compare with the dataset and recognize frame using Machine learning classify Haar Cascade algorithm (HCA). After recognize find accuracy if accuracy reaches 70 or above then it will generate an alert and send an alert automatically to the parents for security and attendance purposes.

Final Deliverable of the Project

HW/SW integrated system

Core Industry

Education

Other Industries

Medical , Others , Security

Core Technology

Artificial Intelligence(AI)

Other Technologies

Others

Sustainable Development Goals

Industry, Innovation and Infrastructure, Responsible Consumption and Production

Required Resources

If you need this project, please contact me on contact@adikhanofficial.com