Performance analysis on weighting factor (a) on spatial temporal gradient technique and High Confidence Reliability with sub-pixel displacement

Traditionally, the weighting factor (α) is one of the most importance parameter of the optical flow based on the temporal gradient technique and directly impacts to the optical flow performance. This paper presents a performance analysis of sub-pixel optical flow on Horn-Schunk algorithm (HS) under the kernel model of Barron, Fleet, and Beauchemin (BFB) over various weighting factor (α) concerning with the feedback in Peak Signal to Noise Ratio (PSNR) for the best performance on each frame of video sequence for comparison. We also investigate over confidence based optical flow algorithm for high reliability (CBOF) under the best forward and backward optical flow in PSRN of each reconstructed frame and relationship with the different on master images sequences for evaluation. Experimental results of the maximum and minimum of the best average in PSNR for each reconstructed video sequence are demonstrated for performance evaluation. These experiment results are comprehensively tested on several standard sequences such as AKIYO, COASTGUARD, CONTAINER, and FOREMAN that have difference foreground and background movement characteristic.