Shadow Removal Adaptive Edge Detection Based On Canny Theory

Edge detection is fundamental task for computer vision systems. It has been used extensively as a preprocessing step for a myriad of image processing algorithms, including image enhancement, object detection/recognition, compression, and digital watermarking algorithms. Such algorithms, in turn, have been used for medical, military, security, consumer applications, and industrial detection. As many systems rely on edge detection, the development of accurate edge detection in both clean and noisy environments is a must. Currently, no single edge detection method has been produced whose performance is superior for all applications. This is largely due to the subjective nature of the edge detection problem. Secondly, a reliable, unbiased measure to objectively assess the performance of edge detector outputs has not been developed which can be used universally.There are many traditional edge detection algorithms, such as Roberts, Sobel, Prewitt, Kirsch, and Laplace. The fundamental of them is that constructing an edge detection algorithm with a small neighborhood in each pixel of the original image, and then carry out with first differential or second differential operator to obtain the maximum gradient or the zero-crossing point of the second derivative, finally select an appropriate threshold to extract the edge. But these algorithms share the same shortcomings, such as, they are sensitive to noise, and can't select threshold adaptively, and the detection results are as well as we hope.According to the application background of this paper, we firstly discuss the image preprocessing about how to eliminate shadows on the edge detection effect with global threshold, then based on analyzing the classical Canny algorithm, we propose an adaptive improved Canny algorithm. On account of the parameters of traditional algorithm should be set artificially, We put forward a improved filter according to characteristics of image region, which can automatically identify scale parameter of Gaussian filter function; Secondly we improve the traditional gradient value calculation; Finally we propose a method which can identify the double adaptive threshold using Otsu method combined image gradient map, and it avoids the effect of subjective factor. Experiments show that the method of this paper performs better than classical Canny edge detection algorithm, and more automatic.