Summarization Extraction Method for Surveillance Video Based on Color Spatial Distribution Characteristic

This paper proposes a novel key frame extraction method by using color spatial distribution histograms for surveillance video representation and summarization. Pixels in the same color are contiguous or non-contiguous to each other in an image, which is a palpable spatial characteristic. With this attribute, the pixels are classified as contiguous and noncontiguous ones, and then, refined histograms distinguish the small changes of video content better than the original histograms between frames, and it allows fine distinctions that are similar to original histograms. With the evaluation of our experiments, this approach extracts summarization more efficiently and accurately.