Motion estimation in video compression based on Artificial Bee Colony
Nimmy P. George, J. Anitha · 2015 2nd International Conference on Electronics and Communication Systems (ICECS) · 2015
Block matching motion estimation is an unavoidable method in video coding applications. Motion estimation will remove temporal redundancy from the adjacent frames since it can achieve compression. A popular matching criterion used is sum of absolute difference, but it is computationally more expensive. Full search algorithm is the simplest BM method in which it provides most accurate motion vector. Here Artificial Bee Colony optimization is used in order to reduce the number of search locations in the video frame. This method does not contain any fixed pattern for search. Algorithm calculates the SAD values for the current position from the SAD values of the previously calculated neighboring positions. Simulation results show the improvement of motion vector from other fast block matching algorithm.