The main aim of Video Summarization (VS) attempts is to provide a condensed view of the video by eliminating redundancies and extracting key frames from the video. This paper proposes a novel image -block based technique for video summarization by dividing the frames of the video into blocks and calculating the mean, variance, skew, kurtosis histogram of every block and comparing the same with the corresponding blocks of the next frame. The proposed technique successfully detects the shot boundary and keyframes by considering the distribution of colors in the image which are captured by higher order color moments. From every shot detected, the frame with highest mean and variance is selected as keyframe. The experimental results of the algorithm are compared with other key frame extraction techniques. The proposed technique gives comparable and even improved result in many cases.

No abstract