Improved Canny Edge Detection Operator

Edge detection is an important method in image processing, which can not only extract the important contour features of the image, but also greatly reduce the workload of the computer in processing. Among them, the canny operator has been widely used among many algorithms with its certain advantages, but the processing effect is easily affected by the noise. In order to remove the Gaussian noise in the image and to retain the edge information better, this paper proposes an improved canny edge detection algorithm. The algorithm discards the previous Gaussian filtering and adopts bilateral filtering to improve the smoothing process, replaces the original gradient template in the gradient calculation, and adds 45 angle and 135 angle gradient templates to reduce the problem of missed detection and false detection, then uses linear interpolation to calculate the sub-pixel gradient value in the non-maximum suppression to makes the non-maximum suppression result more accurate. At last, the Otsu is used for threshold selection in the dual-threshold screening stage to achieve its adaptivity. After comparison, it is found that the improved canny operator has some advantages in processing images with Gaussian noise compared with the traditional canny operator.