IT Ministry is developing an automatic security and surveillance system for the video stream to help law enforcement agencies to curb terrorism and identify culprits.
The system is being developed with the collaboration of the Al-Khwarizmi Institute of Computer Science of University of Engineering and Technology (UET) with a cost of around 26.62 million. In this regard, surveillance cameras would be installed in sensitive areas that will monitor and store the videos of daily life activities happening in that area.
According to officials, the automatic understanding of the videos will be analyzed using image processing methods which would help in developing multiple useful applications. Moreover, a textual description of the video sequence will be created using natural language processing which does not require huge storage requirement. The system is also meant to provide a web-based utility for videos to search most relevant videos in accordance with the security needs.
The officials said that technically this project mainly involves automatic understanding of the visual scenes collected from individual high level features (HLFs) present in videos and added by extracting high-level features, such as human’s face, age, gender, emotions, actions and objects, human-human interaction, human-object interaction, scene settings etc. In this way, the interpretation of a video stream can be easily obtained.
Other important purposes of this project are to monitor crowd in restricted access points and counting the occurrence of specific persons in the specific area. Furthermore, the project will also provide automatic controls for surveillance cameras other than human monitoring. Human monitoring can be compromised with due to various reasons like drowsiness or overwork circumstances, therefore, an automated solution is highly desirable since it would run all days and nights and on any number of cameras without human intervention.