Abstract: The traditional methods of color image edge detection are mainly based on gray images.Color image is converted to gray image，then edges are detected by grayscale edge detector.These methods make use of luminance information to detect edges，but don’t take into account the chroma information of color image.Therefore，some edges cannot be detected.In this paper，a method of SUSAN color image edge detection based on CIELAB space is presented.The RGB color image is converted to CIELAB space，then SUSAN operator based on color difference is applied to detect edges.The experimental results show that the proposed method can effectively detect the edges of color image.Compared with the traditional grayscale edge detetion，the method achieves better performance on edge preservation.

Key words: color image, edge detection, color difference, CIELAB, SUSAN operator

摘要： 传统的彩色图像边缘检测方法主要是基于灰度图像的，先将彩色图像转化为灰度图像，然后用灰度图像边缘检测方法检测边缘。这些方法利用彩色图像的亮度信息进行边缘检测，没有考虑其色度信息。因此部分边缘不能被检测出来。提出了一种基于CIELAB空间的SUSAN彩色图像边缘检测方法。该方法首先将彩色图像从RGB空间转换到CIELAB空间，然后用基于色差的SUSAN算子检测边缘。实验结果表明：此方法能有效地检测出彩色图像的边缘。在保留图像边缘方面，性能优于基于灰度图像的边缘检测方法。

关键词: 彩色图像, 边缘检测, 色差, CIELAB空间, SUSAN算子