Extracting Key Frames for Surveillance Video Based on Color Spatial Distribution Histograms

Authors:
Jun Chang;Ruimin Hu;Zhongyuan Wang;Bo Hang
Affiliations:
Computer School, Wuhan University, Wuhan, China 430072 and National Engineering Research Center for Multimedia Software, Wuhan University, Wuhan, China 430072;National Engineering Research Center for Multimedia Software, Wuhan University, Wuhan, China 430072;National Engineering Research Center for Multimedia Software, Wuhan University, Wuhan, China 430072;National Engineering Research Center for Multimedia Software, Wuhan University, Wuhan, China 430072 and Mathematics and Computer Science College, Xiangfan University, Xiangfan, China 441053
Venue:
PCM '09 Proceedings of the 10th Pacific Rim Conference on Multimedia: Advances in Multimedia Information Processing
Year:
2009

Citing 7
Cited 0

Image processing on compressed data for large video databases

MULTIMEDIA '93 Proceedings of the first ACM international conference on Multimedia
A Shortest Path Representation for Video Summarisation

ICIAP '03 Proceedings of the 12th International Conference on Image Analysis and Processing
A new general framework for shot boundary detection and key-frame extraction

Proceedings of the 7th ACM SIGMM international workshop on Multimedia information retrieval
Interactive key frame selection model

Journal of Visual Communication and Image Representation
Two-stage hierarchical video summary extraction to match low-level user browsing preferences

IEEE Transactions on Multimedia
An integrated scheme for automated video abstraction based on unsupervised cluster-validity analysis

IEEE Transactions on Circuits and Systems for Video Technology
A novel video key-frame-extraction algorithm based on perceived motion energy model

IEEE Transactions on Circuits and Systems for Video Technology

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper proposes a new key frame extraction method by using color spatial distribution histograms. Pixels in the same color are contiguous or non-contiguous to each other in an image, which is a noticeable spatial character. With this property, the pixels are classified as contiguous and non-contiguous ones. Therefore, refined histograms distinguish the small changes 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 extract key frames more efficiently and accurately.