Color Image Segmentation Based on Union Probability Density Function of Hue, Light and Saturation

摘要

While graylevel image segmentation has become quite a mature field, much less work has been done with regard to color image segmentation. In this paper, we propose a new color image segmentation method, which bases on union pdf of hue, light and saturation. In proposed methods, we first convert RGB space into HSI and segment color images depending mainly on hue. Then, we use light and saturation as supplement to extract details of the images, and compose the meaningful regions by eliminating noises. We demonstrate the performance of the proposed method on a variety of images. The experimental results show, the performance proposed method is robust.