Application of An Improved Motion Estimated Three Step Search Algorithm in infrared Video

摘要

Infrared video image compression is an important video processing technology infrastructure. This thesis proposes a efficient and fast block matching motion estimation algorithm. One strategy of this algorithm is to begin searching from two positions the predictive search centre and the(0,0)position, which earl still hold direction when the relatively predictive error is big. The other strategy is to adopt the Big Small Square search pattern, which chooses search step according to the moving state of objects. Experimental results show that the algorithm galls highly reduce computational complexity with little video quality dropping.