基于数学形态学细化算子的改进Canny算法研究

Research on improved Canny algorithm based on thinning operator of mathematical morphology

边缘检测是图像处理领域研究的一个重要内容。本文基于数学形态学改进现有Canny算子。该方法首先用Canny算子进行滤波,然后采用非极大抑制技术,将强边缘图像和弱边缘图像的边缘进行连接,再对提取的边缘利用形态学细化算子细化。针对不同图像采用不同门限比例,可取得不同的边缘提取效果。本文采用峰值信噪比、均方误差、平均绝对误差三个评价指标对边缘检测算法的优越性进行度量。实验分析表明,本文算法优于Sobel算子、Roberts算子、Log算子及传统Canny算法。

Edge detection is an important part in the field of image processing. In this paper, a new edge detection method was put forward by improving the Canny operator based on mathematical morphology. The method filtered the image by Canny operator, and then nonmaxima suppression technology was applied for fitting the edges of strong edge image and weak edge image together. Finally, thinning operator of mathematical morphology was introduced to thin the edge of result image. Different effects on edge detection could be achieved by different images using corresponding threshold ratios. With three evaluation indexes such as PSNR, MSE, MAE, the results showed that proposed approach was obviously superior to Sobel operator, Roberts operator, Log operator and Canny operator.