Color edge detection based on mathematical morphology in HSI space

This paper proposes an adaptive algorithm of color image edge detection which is based on multiple structure and multi-scale elements in HSI space. Firstly, morphological edge detection is executed using different structure and different scale elements to hue, saturation and intensity. Then the color edge information is obtained by combining component´s edge information according to the weight derived from information entropy. The experimental results show that proposed algorithm can adequately make use of the hue, saturation and intensity information, effectively eliminate the noise and adaptively extract the complete edge information.