Closed-circuit television (CCTV) cameras, sometimes referred to as security cameras or surveillance cameras, have become a necessary component of contemporary life and have a big influence on many facets of our existence. There is an enormous amount of digital video footage as a result of the daily increase in CCTV surveillance camera numbers. Many CCTV cameras run around the clock, but the data they record is in raw form, with little to no motion footage typically being captured, wasting a lot of storage space. Condensing a longer film into a more manageable, shorter version while maintaining its most significant and educational material is known as video summary. The enormous volume of recorded CCTV footage can be condensed into shorter, more insightful summaries with the use of summarization. Four processes make up the suggested system: object tracking, background extraction, background subtraction, and, lastly, copying and superimposing moving objects. We highlight the importance of the KNN (k-nearest neighbors algorithm)-based background subtraction method over the conventional MOG (mixture of Gaussians) algorithm in our study by utilizing its superior capabilities. KNN provides improved precision for detecting motion or disruptions in the video stream. Our method smoothly incorporates an object tracking algorithm to track the trajectory after detecting an object’s initial presence and movement. Events are carefully given timestamps, which offer important temporal data for in-depth event analysis. All of the moving items in the movie are expertly identified and tracked by the computer, which then superimposes these occurrences into a single clip. This improves object tracking effectiveness and strengthens the surveillance system, which is essential for accurately tracking numerous occurrences.