Abstract:

Traditional block matching video stabilization algorithms cannot remove the influence of moving objects completely and needs a preset threshold value to decide similarity in
processing procedure. It generates calculation errors easily and has lots of searching positions, as a result, increasing computation. First the algorithm searches the matching blocks at adjacent frames in X and Y independent direction and determines the optimal location adaptively. Second according to the optimal location the algorithm removes the moving object areas and adjusts step length to next searching until the searching step length is zero. Finally the algorithm calculates the motion vector of adjacent frames with motion filtering and compensates the dithering of video frames. The processing result shows that this algorithm can stabilize the video with large size dithering, global motion and moving objects.