RIMSA

Executive Summary:

The terrorism has been one of the major concern of the world for the last two decades.  After the incident of 9/11 in United states of America, it became severe, and now it is a major threat for almost all countries of the world. Our homeland Pakistan has also been facing the terrorism and insecurity for the last two decades, which has halted the process of peace and development in the country. Thousands of the Pakistanis have lost their lives in war against terrorism and according to media reports Pakistan has suffered more than $107 billion economic loss since then. This issue has become so complicated and requires thorough treatment with political, justice, economic, social, educational, and technological initiatives. However, from technological perspective, automated intelligent video surveillance has proven to be quite helpful in protection against terrorist activities. For this reason, there is a drastic rise in video surveillance cameras all over the world, and this has becoming common in Pakistan as well. In recent years, educational institutions, specifically the universities have been easy target of the terrorist activities due to the presence of thousands of people at one place and ineffective security measures.

This project is aimed at developing an automated video surveillance system for the university campus. It consists of four different modules that would be deployed at important locations of the campus and coordinated at the central location. One of its module would assist the security personnel in automatic screening of the vehicles for any arms and ammunitions that might have been attached at the bottom of the vehicle. The second module of the proposed system would be used for boundary crossing detection and automatically inform the authorities for immediate action with the type of intrusion and its location. The third and fourth modules will be installed on the campus gates and inside the buildings respectively for monitoring suspicious activities such as fighting, leaving bag unattended, arms, theft cases, or any other activity that would be defined as suspicious by the security department. All these modules would be interfaced with web and mobile applications for exchange of information and management by the security department.

The proposed system will use the surveillance cameras installed at designated areas of the campus for capturing the videos of daily activities in real time. Technically, Automatic understanding and interpretation of these videos will be performed in real time using image processing, computer vision and machine learning techniques. The major focus would be on deep learning-based techniques, which have been successfully deployed in many visual categorization tasks such as image analysis, speech recognition, object recognition, and activity recognition. This project when completed will provide a great support to the security department of the campus and law enforcement agencies in combating events of terrorism and capturing culprits. It will provide 24/7 monitoring without any human intervention. Due to its generic nature, this project will be equally useful for other organizations and commercialization.