Image segmentation via fuzzy object extraction and edge detection and its medical application

Issue title: Special Issue on Clinical Applications of Modern Imaging

Article type: Research Article

Authors: Lin, Yao | Tian, Jie | He, Huiguang

Affiliations: AI Lab, Institute of Automation, Chinese Academy of Sciences, Beijing POB 2728, 100080, China. Tel.: +86 10 82618465; Fax: +86 10 85191212; E-mail: [email protected]

Abstract: A new interactive segmentation method that combines fuzzy connected object extraction and edge detection is proposed. Fuzzy connectedness is a global fuzzy relation, which effectively captures fuzzy "hanging togetherness" of image elements. First, by selecting the seed point, fuzzy connectedness value between each image element and the seed point is computed via dynamic programming. Then, through adjusting the parameter, the different fuzzy connected object could be extracted. Finally, an edge detection method is adopted to detect the object boundary. The algorithm is demonstrated on some medical images: segmentation of knee tissues in CT image and segmentation of brain tissues in MR image. The results show that this method works well for some CT and MR images.

Journal: Journal of X-Ray Science and Technology, vol. 10, no. 1-2, pp. 95-106, 2002