A new color adaptive mathematical morphology operator based on distance and threshold

In mathematical morphology, the fixed structure elements cannot adapt to the whole image. The key to morphology is selecting suitable structure elements according to the different attributes of image. Aiming at fixed threshold and selecting main color component based on experience in color adaptive morphology, an automatic threshold selection criterion and a new adaptive color image morphology operator are presented in this paper. Both the local features of image and the distance between pixels are taken into consideration. The new operator keeps the correlation of color image and a new method of vector sorting based on the new operator is proposed. The structure elements can automatically adjust the size and shape according to the automatic threshold. The experiment results show the new operator preserves image details better and the color changes little.