Intelligent Medical Image Segmentation Approach Based on Wavelet-based Multi-resolution Analysis and SOFM
This paper makes a commitment to find a universal medical image segmentation algorithm. Directing towards the particularity and diversity of medical images, an improved segmentation algorithm based on wavelet multi-resolution analysis and neural network is proposed. The original image undergone wavelet multi-resolution analysis, and then the results of the low-frequency information were sent into the self-organizing feature map (SOFM) for further cluster analysis. Only one (global threshold) or a group of (local threshold) optimal threshold can be obtained intelligently based on number of categories. By processing computed tomography medical images, we have not only got a complete region of interest, but also obtained clear edge features. Experimental results show that the method can well take into account the applicability and accuracy of segmentation, and have broad application prospects.
medical images image segmentation wavelet transform multi-resolution analysis SOFM
Wencang Zhao Junbo Zhang
College of Automation and Electronic Engineering Qingdao University of Science & Technology Qingdao 266042 China
国际会议
秦皇岛
英文
425-428
2010-11-05（万方平台首次上网日期，不代表论文的发表时间）