Fuzzy local intensity clustering (FLIC) model for automatic medical image segmentation
| Autor: | Parviz Keshavarzi, Mohammad Rahmanimanesh, Saeed Mozaffari, Asieh Khosravanian |
|---|---|
| Rok vydání: | 2020 |
| Předmět: |
Fuzzy clustering
Computer science business.industry High Energy Physics::Lattice ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION 020207 software engineering Pattern recognition 02 engineering and technology Image segmentation Computer Graphics and Computer-Aided Design Fuzzy logic Computer graphics Computer Science::Computer Vision and Pattern Recognition 0202 electrical engineering electronic engineering information engineering Entropy (information theory) 020201 artificial intelligence & image processing Segmentation Computer Vision and Pattern Recognition Artificial intelligence business Cluster analysis Fuzzy method Software |
| Zdroj: | The Visual Computer. 37:1185-1206 |
| ISSN: | 1432-2315 0178-2789 |
| DOI: | 10.1007/s00371-020-01861-1 |
| Popis: | Intensity inhomogeneity is one of the main challenges in automatic medical image segmentation. In this paper, fuzzy local intensity clustering (FLIC), which is based on the combination of level set algorithm and fuzzy clustering, is proposed to mitigate the effect of intensity variation and noise contamination. For the FLIC method, the segmentation and bias modification are carried out in a fully automatic and simultaneous manner through the local clustering of intensity and selection of the initial contour by the fuzzy method. Besides, the local entropy is integrated into the FLIC function to improve the contour evolution. Experimental results on inhomogeneous medical images indicate the superiority of the FLIC model over the other state-of-the-art segmentation methods in terms of accuracy, robustness, and computational time. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Full text from SpringerLink