Motion vector detection based on local autocorrelation coefficient

Download PDF
Open Webpage

Honghui Fan, Hongjin Zhu

Published: 01 Jan 2019, Last Modified: 05 Nov 2025Clust. Comput. 2019EveryoneRevisionsBibTeXCC BY-SA 4.0
Abstract: This paper introduced local autocorrelation (LAC) to the preprocessing of motion vector detection in order to increase the detection accuracy of the moving vector. We applied LAC coefficient, Sobel operator and moving average calculation to motion vector detection, and Peak Signal to Noise Ratio is used to quantitative evaluation and analysis of motion vector detection accuracy. LAC optimum parameters corresponding to motion vector detection are obtained by a comparative experiment. LAC with optimum parameters achieve higher detection accuracy for the two kinds of video image with illumination changes and without illumination changes. The effectiveness of the LAC coefficient as the preprocessing algorithm of motion vector detection is verified, and the accuracy of the motion vector detection is improved.