Matching Algorithm Based On Edge Features To Improve

Global search for the image matching the speed of the impact of matching problems, this paper, we introduce a genetic algorithm, with the Hausdorff distance will be a combination of image matching to optimize the search process. First, the reference images and real-time image pre-processing, in order to select adaptive threshold Canny operator to extract the edge image as a feature unit. And then deal with the introduction of genetic algorithm to match the image registration operation. Genetic manipulation in the use of standard deviation modified Hausdorff distance improved genetic algorithm to construct the fitness function, genetic algorithm to determine the optimal transformation parameters, the completion of the reference images and real-time image matching. Finally, through experiment simulation the effectiveness of the algorithm.Hausdorff distance between two images as an evaluation of the relationship between the location of the quantitative criteria, with its strong anti-interference ability and fault tolerance are widely used in image matching. Hausdorff distance, however simple on the noise and isolated points are more sensitive, resulting in a higher rate of false matches. Various types of improvement in the form of Hausdorff distance matching a specific environment to overcome these deficiencies, but in the face of the complexity of the issue of distortion of the imaging noise and complex situation, they can not achieve an ideal effect match. This article takes a first standard deviation modified Hausdorff distance (STMHD) sort, for their part of Seoul after the mean improvements in the form of Hausdorff distance. STMHD this improved to better overcome the noise, and some pseudo-edge matching block of the impact of precision and stability in the speed and accuracy of registration to achieve more desirable results.