Color image segmentation using watershed transform based on optimized grid
Xiuyan Wang, Tingting Cheng, Zongshuai Li · 2010 3rd International Congress on Image and Signal Processing · 2010
With regard to the over-segmentation of the traditional watershed transform, the paper proposes an algorithm using watershed transform based on optimized grid for color image segmentation. The algorithm transforms RGB color space to Lab color space and quantifys the color channels a and b in accordance with the grid-required. The watershed transform algorithm is applied to two-dimensional histogram. The regions are integrated by combining the region color information and space features. The experimental results show that the algorithm can solve over-segmentation problem, has good noise immunity and reduces the time complexity. The algorithm is effective measure for color image segmentation.