为了解决传统sobel算子在边缘检测中的不足,提出了一种改进算法。在传统算法的基础上增加了45°、135°、180°、225°、270°以及315°六个方向模板,并在模板权值的选取上进行改进,调整各个模板参数。用改进后的算法对实验图像进行边缘提取,由仿真实验可以看出,改进后算法提高了边缘检测的精度,并具有优于传统算法的抗噪性能。
Aiming at deficiency of traditional sobel operator in edge detection, an improved algorithm is proposed. We add 45°, 135°, 180° ,225°, 270° ,315° six direction pattern based on traditonal operator, and improving the method of weight selection, adjusting the parameter of every template. Using the improved algorithm to extract the edge of the experimental image. Simulation experiments show that, the improved algorithm has improved the precision of edge detection and the anti noise performance.