Towards Enhancing Keyframe Extraction Strategy for Summarizing Surveillance Video: An Implementation Study

The large amounts of surveillance video data are recorded, containing many redundant frames, which makes browsing and retrieval difficult, thus increasing bandwidth utilization, storage capacity, time consumed. To ensure the reduction in utilization capacity to barest minimum, keyframe extraction strategies have been developed. These implemented extract unique keyframes whilst removing redundancies. Despite achieved improvement processes, there still exist a significant number frames summarized videos. With view addressing this issue, current paper proposes an enhanced strategy using k-means clustering statistical approach. Surveillance footage, movie clips, advertisements, sports videos from benchmark database as well Compeng IP were used evaluate performance proposed method. In terms compression ratio, results showed that scheme outperformed existing schemes by 2.82%. This implies further removed whiles retaining quality. playtime, was average 27.32%, making content less cumbersome when compared with schemes. Implementation done MATLAB R2020b.