• DASU DASARI
• VIJJESWARAPU ANUHYA EVANGELINE

The escalation of anti-social activities has prompted a heightened emphasis on security measures. In response, numerous organizations have deployed Closed-Circuit Television (CCTV) systems to monitor individuals and their interactions continuously. In developed nations with sizable populations, each individual can be subjected to camera surveillance up to 30 times daily, resulting in a substantial accumulation of video data within defined time spans. For instance, video recording at 704x576 resolution and 25 frames per second yields an approximate daily data output of 20GB. However, manually scrutinizing this voluminous data for an abnormal event is a nearly unfeasible task due to the considerable human resources and sustained attention it necessitates. This challenge underscores the imperative to automate the monitoring process. Additionally, there is a crucial requirement to swiftly pinpoint the specific frame and segment containing unusual activities to expedite the identification of potential abnormalities. This study presents a solution by utilizing Convolutional Neural Network (CNN) algorithms. The approach involves converting video streams into individual frames and subjecting them to comprehensive analysis. By leveraging CNN, the system adeptly detects and categorizes individuals' activities, facilitating the identification of potentially suspicious behavior. The integration of CNN algorithms significantly enhances the accuracy of abnormal activity detection, making it an invaluable tool for ensuring public safety.

"WEB INTERFACE FOR SUSPICIOUS ACTIVITY DETECTION FROM CCTV FOOTAGE", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.10, Issue 8, page no.h63-h67, August-2023, Available :http://www.jetir.org/papers/JETIR2308709.pdf

"WEB INTERFACE FOR SUSPICIOUS ACTIVITY DETECTION FROM CCTV FOOTAGE", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.10, Issue 8, page no. pph63-h67, August-2023, Available at : http://www.jetir.org/papers/JETIR2308709.pdf