Image matching under generalized hough transform

The paper analyzes the techniques of image template matching, and gives a matching framework based on generalized Hough transform, which is applicable to lots of current methods. With the framework’s guidance, extracting image features with specific characteristics and devising feature matching algorithms become easier and more efficient. We propose a new fast matching algorithm as evidence. It is robust for the linear transformation of image grey value and image noise. Its time complexity reduced to O(size of search image), the theoretic lower limit for image matching.