Title :
Key Frame Extraction based on frame blocks differential accumulation
Author :
Cao, Changqing ; Chen, Zehua ; Xie, Gang ; Lei, Shaoshuai
Author_Institution :
Taiyuan Univ. of Technol., Taiyuan, China
Abstract :
Key Frame Extraction is a key step for video content analysis and video retrieval. According to the motion of object and camera, a method of key frame extraction based on frame blocks differential accumulation with double thresholds is proposed. Firstly, image frames are partitioned into equal-sized sub-images. Then the blocks in current frames are compared with the corresponding blocks in last extracted key frame to detect weather their contents changed. At last, the key frames of the video can be determined by the number of blocks whose contents changed a lot. The experiment results show that this method has a satisfactory efficiency.
Keywords :
cameras; content-based retrieval; feature extraction; image motion analysis; image segmentation; image sequences; video cameras; video retrieval; video signal processing; camera motion; double-thresholds; frame block differential accumulation; image frame partitioning; key frame extraction; object motion; subimages; video content analysis; video retrieval; Cameras; Color; Data mining; Feature extraction; Histograms; Image color analysis; Redundancy; Block; Double Thresholds; Frame Differential Accumulation; Key Frame;
Conference_Titel :
Control and Decision Conference (CCDC), 2012 24th Chinese
Conference_Location :
Taiyuan
Print_ISBN :
978-1-4577-2073-4
DOI :
10.1109/CCDC.2012.6243092