一种基于局部多结构元素形态学的边缘检测的修正算法

在前期工作的基础上 ,提出了一种基于局部多结构元素形态学的边缘检测的修正算法 .该算法根据图像中噪声和边缘形态的不同 ,建立了多个结构元素 .利用基于视觉模型的边缘阈值选择策略确定图像中梯度变化的像素点 ,并对其采用二值形态学的腐蚀操作 ,在进行腐蚀操作时 ,放宽了结构元素中相同灰度像素点的判别条件 ,从而更好的选择出图像边缘 ,滤除噪声 .实验表明 ,本文算法不仅具有很好的边缘提取能力 ,而且具有很强的抗噪能力

Modified Edge Detection Algorithm Based on Local Multiple Morphological Structuring Elements

Multiple morphological structuring elements are designed according to the differential pattern between the noise and the edge. By using the choice strategy for edge threshold based on visual perception model, a pixel the gradient varied obviously in the gray scale image is determined, and it is operated by using erosion of binary morphology to obtain the edge and remove the noise. Compared with the traditional morphological method by which all of the pixels of the image are dealt with, only partition pixels are operated in this algorithm, so the computational cost is decreased. On the other hand, the edge threshold is selected by using the choice strategy based on visual perception model, therefore, the detected image edge is more consistent with human visual perception. Compared with the traditional edge operator, the experiment results show that this algorithm not only has a good ability to detect the image edge, but also has a strong ability to suppress the noise in the image.