Study On Template Matching Algorithms Based On Gray Value

Template matching is to find the similar sub-image in another image according to the given template image. Template matching is one of the main research topics in graph and image processing, which has many important applications in many fields, such as machine vision and medical image processing, etc. Recently, the demand for the real-time template matching is becoming more urgent,, the research to improve the real-time performance of the traditional template matching algorithm is of great importance in theory and practice.This paper mainly studies the template matching algorithms based on the gray value. Template matching algorithms based on the gray value are very popular and widely used in template matching, but these algorithms still have some flaws, such as high time complexity and sensitive to the variation of image luminance and scale. To overcome these drawbacks, a new algorithm based on sequential similarity detection is proposed in this paper. Two main steps are involved in the improved algorithm:1. The coarse matching between matching image and template image. For any point in the image to be matched, the threshold value can be calculated by the corresponding search formula of the template image and the search image to be matched interlaced every other column, thus a rough match point can be obtained.2. Precise matching process. The precise matching based on the coarse matching is to find the optimal matching point to achieve the precise purpose, taking the relevant points obtained in coarse matching as the center point,,and the overall amount of calculation is thereby reduced.Experimental results show that the computing time and calculation of the improved template matching algorithm were reduced greatly as compared to the existing algorithms based on gray value.