An Improved Canny Edge Detector Against Impulsive Noise Based on CIELAB Space

The traditional Canny edge detector is designed to detect the edges of gray image, the color image must be converted to gray image, this method only uses the luminance information of color image, and don´t take into account the chroma information, therefore, it misses some edges. In this paper, an improved Canny edge detector against impulsive noise based on CIELAB space is proposed, the color image is first converted from RGB space to CIELAB space, it uses switching vector median filter to remove impulsive noise, and then calculates the color difference and direction, non-maximal suppression is used to refine the color difference image, finally, it extracts edges by double-threshold method. The experimental results show that the proposed method has better performance on impulsive noise suppression and detects more edges than the traditional Canny edge detector.