Location of 2-Dim code in image by corner detection

We report on the development and implementation of a robust algorithm for extracting 2-Dim code Area in video images. The algorithm concludes two main steps, firstly, convert the RGB images into gray-scale images and location the point of max local radio in the gray image. These points are so-called corner points. Secondly we used the points to clustering and the results of the clustering can be used to find one area that conform the shape of a 2-Dim code area. The experiment results shows that 2-Dim code area in the 2-Dim code images with rectangle or circle element, bad contrast in illumination or with adjacent noise in the image could be find correctly.