In India, the number of surveillance cameras has increased rapidly, making it challenging for law enforcement to go through the vast amount of recorded footage. We suggest a technique for quickly reviewing large amounts of video feed in order to address this issue. By utilizing sophisticated algorithms such as facial recognition, lightweight YuNet model face detection, and motion detection through background subtraction, our system accelerates the analysis of videos from various camera systems. Our model is a great tool for any agency looking to improve its operational efficiency when handling large amounts of monitoring data because it can quickly identify important events and produce concise summaries with timestamps. This piece of research contributes towards development within this field since such new methods are combined into one single solution thereby advancing surveillance technology while at the same time providing concrete benefits for everyday challenges faced by law enforcement agencies.