基于形态学滤波结合LOG算法的边缘检测

边缘检测是一种尝试从图像中提取有效部分的方法，主要捕获图像像素的急剧变化和检测重要的区域。为了更有效地进行边缘检测和抑制噪声，文章中对常规形态学边缘检测进行了改进，采用了多阈值分解对灰度图像进行二值化处理和分解，之后对图像进行了基于二阶拉普拉斯算子的LOG边缘检测，在初步提取出图像边缘后，又进行了多结构形态学滤波来实现对图像边缘进行进一步边界增强。实验结果表明，该法保留了更完善的边缘信息，有效消除了叠加噪声。

Edge detection based on morphological filtering combined with LOG algorithm

Edge detection is a method of trying to extract the effective part from the image, mainly detecting sharp changes in image brightness and capturing important areas. In order to perform edge detection and noise suppression more effectively, the conventional morphological edge detection is improved, and the image is binarized by threshold decomposition. The LOG edge detection of the Laplacian operator is further enhanced by multi-structural morphological filtering after the initial extraction of the image edges. The experimental results show that the method retains more edge information and effectively eliminates the superimposed noise.