摘要： 当图像中噪声与边缘强度相差不大时，用LFFD算法检测边缘时会扩大噪声。针对该问题，给出一种抗噪声的边缘检测算法(EFFD)。该改进算法通过使用模糊熵来抑制噪声扩大，用分形维度来描述图像的局部特征。通过对不带噪声和带有椒盐噪声的图像的边缘检测，说明EFFD在带噪声的图像中可以抑制噪声扩大，获得较好的边缘特征。

关键词: 边缘检测, 计盒维, 局部模糊分形维, 模糊熵

Abstract: If the difference of intensity of noise and edge strength is not significant in image, Local Fuzzy Fractal Dimension(LFFD) can make noise larger. For this problem, EFFD algorithm which can reduce image noise availably is proposed. The improved algorithm uses fuzzy entropy to suppress the noise increased, and uses the fractal dimension to describe the image local characteristics. Through edge detection of the noise and saltpepper noise images, experimental results show that the algorithm can suppress the noise expanded to obtain better edge features in the noise image.

Key words: edge detection, boxcounting fractal, local fuzzy fractal dimension, fuzzy entropy

中图分类号: 

• TP391.41