Video Compression By Eliminating Irrelevant Frames

Regarding the security of ATM systems situated at various locations throughout world, constant monitoring via surveillance machine is required. Even though ATMs are equipped with manual protection, several theft events have made headlines over past few decades. Since continuous provides an alternative to security, it not feasible for banks do so every one these ATMs. With much data generated by system inside and outside ATM, becomes expensive manage in terms memory. As a result, goal this job reduce number frames video that was collected removing all pointless frames. The technique employs transfer learning on Mask RCNN Network identify containing people before any such network, which typically trained COCO dataset, will be used detect humans so-called relevant subsequently filter out no humans. pertinent after turned into kept backup. plan decreases amount storage needed monitoring. Keywords- Region-Based Convolutional Neural (RCNN), Networks (CNN), Closed Circuit Television(CCTV), Discrete cosine transform (DCT), GPU (Graphics Processing Unit), Deep Learning techniques, area classification,