STKVS: secure technique for keyframes-based video summarization model

Video security has emerged as an essential study inultimedia security in recent years because videos are the most effective and widely used multimedia format and immediately establish a connection with users. It is necessary to prevent sensitive information from being stolen or destroyed in various domains, including the military, finance, and education. The proposed STKVS framework ensures the summarized video’s security in secret keyframes (SK) rather than securing the whole video to reduce the computation, complexity, and processing needs and recover lossless SK. STKVS framework consists of a secure keyframes-based generic Video Summarization (VS) model to generate a secure video summary. It extracts SK from the video using the proposed Probability-Based VS (PBVS). Then, it utilizes the proposed BEMSS (Blockwise Encryption-based Multi Secret Sharing) scheme to provide multi-secret image sharing to the SK. The proposed model PBVS achieved an average F-score of 0.73 and 0.76 on two benchmark (OV and YT) datasets, respectively, showing its effectiveness in producing informative video summaries. Additionally, BEMSS outperforms the other related security models.