The Edge-Detection Algorithm Based on Enlarged Block-Range

摘要

Because the contrasts of different regions in an image are different, as it detecting the edge, it is very difficult to choose different detection thresholds in different image areas. So the edge which detected by the traditional algorithm is more fuzzy, its edge-line thickness is uneven, and there is more burr and fracture in the edge line. Firstly, the algorithm divides the image into same size of pixel blocks. To search for the maximum value b and the minimum value a in the pixel block, and obtained the range [a, b] of the pixel block. Then unify the range[a,b] of the pixel block to the total range[c,d] of the image. Finally, use the traditional detection operator to detect the image edge. The Algorithm can choose different detection thresholds in different image areas. The experimental results show that the detection edge detected by the algorithm is more clear and exquisite than that detected by the classical algorithm. And its edge-line thickness is even, and there is lesser burr and fracture in the edge line.