I. Introduction
Pattern matching, also called template matching, is the procedure of seeking a required pattern or template in a given signal. For pattern matching in images, this procedure is illustrated in Fig. 1. Pattern matching has applications in manufacturing for quality control [1], image-based rendering [2], image compression [3], object detection [4], super resolution [5], texture synthesis [6], block matching in motion estimation [7], [8], image denoising [9]–[11], road/path tracking [12], mouth tracking [13], image matching [14], action recognition [15], and tone mapping [16].
Pattern matching in image ‘couple.’ We represent the pattern by vector and represent the candidate windows by vectors .