Automatic Intestinal Canal Segmentation Based Region Growing with Multi-Scale Entropy
Xin Hua, Jide Qian, Hengjun Zhao, Lipei Liu, Li Liu, Yi Wu · 2018
To reduce the manual participation involved in existing digitized human intestinal canal segmentation, an algorithm based on region growing using multi-scale entropy as the regional growth standard is proposed. In this algorithm, the color digitized human slice image is converted from the RGB color space to the HSV color space firstly. Then the seed points for region growing are automatic generated by the hue component histogram in the HSV space. According to the similarity of entropy vectors with different radius, the intestinal canal in digitized human slice image can be automatic segmented by region growing method finally. The proposed algorithm is tested on Chinese Visible Human Dataset, and the qualitative analysis of experimental results shows the effectiveness of the proposed algorithm.