Title :
Multivideo mosaic based on SIFT algorithm
Author :
Yang, Jinkun ; Pei, Yijian ; Li, Bo ; Tao, Qingsong
Author_Institution :
Comput. Sci. Dept., Yunnan Univ., Kunming, China
Abstract :
Video mosaic is an extend technology which based on image mosaic. It has a wide range of applications such as video surveillance, television, military inspection and other fields. SIFT algorithm is widely used in computer vision fields, this algorithm can be used in multivideo mosaic. but this algorithm is of high complexity. So we present an improved SIFT algorithm based on SIFT feature extraction, and we use this improved algorithm in the fast video sticking area. This algorithm is achieved with multicore system. Experimental results show that our improved SIFT algorithm can complete muliti-video mosaics, and can work better than original SIFT algorithm in combining a greater view. at the same time, this improved algorithm has a good performance than other similar algorithm.
Keywords :
computer vision; feature extraction; image segmentation; multiprocessing systems; transforms; video signal processing; video surveillance; SIFT feature extraction; computer vision; image mosaic; improved SIFT algorithm; multicore system; multivideo mosaic; video sticking area; Marketing and sales; Mercury (metals); SIFT algorithmic; image matching; parallel computing; video splicing;
Conference_Titel :
Computer Science and Network Technology (ICCSNT), 2011 International Conference on
Conference_Location :
Harbin
Print_ISBN :
978-1-4577-1586-0
DOI :
10.1109/ICCSNT.2011.6182249
