Image segmentation based on the HSI color space and an improved mean shift

A novel image segmentation algorithm based on the HSI color space and an improved mean shift is proposed. The current work is discussed form two perspectives, i.e., the color space more suitable for human being's perception is selected and the segmentation algorithm is improved sufficiently. Unlike many of the traditional algorithms, the proposed algorithm is nonparametric and does not require any training base or prior knowledge, operates on the image adaptively. Experimental results reveal that the proposed algorithm can achieve better segmentation performance and has a stronger self-adaptability but less complexity than the comparison methods. Moreover, it is shown that the proposed algorithm can be a promising approach and be extended to more related research areas.