A Keyframe Extraction Using HDBSCAN With Particle Swarm Optimization

Currently video summarization is an emerging research field. Video summarization which is also termed as Keyframe extraction provides an abstract view of the original video. It aims in finding the important events and objects in a video to give an overview of the original video. This research work focuses on fimding the keyframes of the video through the proposed Keyframe Extraction using Particle Swarm optimization. Two features namely color histogram and histogram of Local Binary Pattern are extracted and clustered using Hierarchical Density-Based Spatial Clustering of Applications with Noise. The parameters of clustering are optimized using Particle Swarm optimization. The experimental results show that the proposed method gives better results when compared to the existing methods.