关键词: 图像分割, 种子点, 区域生长, 肝脏分割

Abstract: Aiming to the problem of selecting the seed points manually and the accuracy and integrity of extracting the liver region, an automatic segmentation method based on the maximum inscribed circle is proposed in this paper. Firstly, the method uses the maximum area measurement method to lock the binarized liver region, and then automatically acquires the seed point position by finding the center of the largest inscribed circle that locks the liver region. Finally, the improved region growing algorithm is used for image segmentation. The experimental results show that the method effectively solves the problem of manually selecting seed points for regional growth, and can accurately and completely segment the liver tissue. In addition, it avoids subjective factors to select seed points at the wrong positions such as edges or noise.

Key words: image segmentation, seed point, region growth, liver segmentation