Research of Liver Segmentation Based on Distance Transform and Iterative Watersheds

摘要：

Watershed has a strong edge detection capability and is also more sensitive to weak edges. However, traditional watershed algorithms easily lead to oversegmentation, especially in abdominal CT images where the gray value is similar. Therefore, this paper puts forward a new method of liver segmentation, using a multi-step iterative watershed segmentation method, combined with complete Euclidean distance transform, mathematical morphology and other methods to solve the over-segmentation problem. We verify the validity of this algorithm by the actual liver segmentation experiments in abdominal CT images.

关键词： image segmentation liver watershed distance transform

作者: Yanan Zhang Huiyan Jiang Mao Zong Xiangying Liu

作者单位: Sino-Dutch College, Northeastern University, Liaoning Province, Shenyang, China, 110004 Software College, Northeastern University, Liaoning Province, Shenyang, China, 110004 Key Laboratory Software College, Northeastern University, Liaoning Province, Shenyang, China, 110004

会议类型: 国际会议

会议名称: 2010 International Conference on Image Analysis and Signal Processing(2010 图像分析与信号处理国际会议 IASP 10)

会议地点: 厦门

会议语种:英文

页码: 274-277

在线出版日期: 2010-04-12（万方平台首次上网日期，不代表论文的发表时间）