Template matching of occluded object under low PSNR

A robust template matching using occlusion-free correlation (OFC) coefficient is presented in this paper for the purpose of locating objects partially occluded under low PSNR (Peak Signal to Noise Ratio) environment. The OFC coefficient can effectively eliminate the negative effect of occlusion on matching score, resulting in providing better accuracy in recognizing and locating objects under bad environment. This algorithm provides a closed-loop scheme through a reliability assessment of occlusion detection which is able to check if the occlusion detection is well made. Extensive experimental results through occluded images with various noise levels demonstrate that the proposed algorithm is robust against objects partially occluded under low PSNR and superior in terms of the false alarm and miss rates in the identification problem. (C) 2012 Elsevier Inc. All rights reserved.