A novel weighted spatial T‐spherical fuzzy C‐means algorithms with bias correction for image segmentation. Issue 2 (19th September 2021)

• Record Type: Journal Article
• Authors: Xian, Sidong; Cheng, Yue; Chen, Kaiyuan
• Is Part Of: International journal of intelligent systems. Volume 37:Issue 2(2022)
• Page Start: 1239
• Page End: 1272
• Publication Date: 2021-09-19
• Subjects: bias correction -- image segmentation -- spatial T‐spherical fuzzy c‐means -- T‐spherical fuzzy sets; Artificial intelligence -- Periodicals; Expert systems (Computer science) -- Periodicals; Intelligence artificielle -- Périodiques; Systèmes experts (Informatique) -- Périodiques; 006.3
• DOI: 10.1002/int.22668
• Languages: English
• ISSNs: 0884-8173

Abstract

Abstract: Fuzzy c‐means (FCM) is a time‐honored method for its simplicity of calculation and ease of understanding. However, the previous image segmentation work exploiting FCM could not achieve the ideal effect in image edge recognition of the fuzzy information or antinoise of the pixel. This paper exploits the spatial T‐spherical fuzzy c‐means model with bias correction ( E M ‐ s T S F C M p q ) to improves the effect of antinoise and image edge recognition. First, the image is transformed into a T‐spherical fuzzy set by a novel T‐spherical fuzzification technology with the processing of the bias fields. Second, the membership degree of the image is updated through the proposed T‐spherical FCM algorithm. Third, a novel calculation method of weighted combination is exploited to integrate spatial information to overcome the noise. The proposed method is applied to three‐dimensional MR image segmentation and the Berkeley segmentation datasets to illustrate the accuracy and efficiency.