Efficient Video Storage Technique for CCTV Surveillance Using OpenCV

Surveillance cameras, dashboard cameras, and police body cameras all record thousands of hours of digital footage each day, which are frequently utilized whenever there is any investigation by the officers. These video recordings may take a lot of time for the officers while investigation and check suspicious activity; they only need the footage of any movement in that area. And it will take a huge amount of space in the database. So for recording only motion-based footage and store that footage in the database and consume less space in the database, we have proposed this system. In this paper, we provide a machine learning (ML) approach motion-based surveillance CCTV video, i.e., we developed a system that only stores any change or movement part of a video in the database, and our system will help to quickly review thousands of hours of video recordings, i.e., it reduces the surveillance time for investigation and produce only the motion-based video for the surveillance in CCTV.