A Fuzzy C-Means Clustering Algorithm Based on Spatial Context Model for Image Segmentation
- PDF / 2,779,100 Bytes
- 17 Pages / 595.276 x 790.866 pts Page_size
- 84 Downloads / 219 Views
A Fuzzy C-Means Clustering Algorithm Based on Spatial Context Model for Image Segmentation Jindong Xu1
•
Tianyu Zhao1 • Guozheng Feng1 • Mengying Ni2 • Shifeng Ou2
Received: 7 September 2020 / Revised: 8 November 2020 / Accepted: 12 November 2020 Taiwan Fuzzy Systems Association 2020
Abstract An improved Fuzzy C-Means (FCM) algorithm, which is called Reliability-based Spatial context Fuzzy C-Means (RSFCM), is proposed for image segmentation in this paper. Aiming to improve the robustness and accuracy of the clustering algorithm, RSFCM integrates neighborhood correlation model with the reliability measurement to describe the spatial relationship of the target. It can make up for the shortcomings of the known FCM algorithm which is sensitive to noise. Furthermore, RSFCM algorithm preserves details of the image by balancing the insensitivity of noise and the reduction of edge blur using a new fuzzy measure indicator. Experimental data consisting of a synthetic image, a brain Magnetic Resonance (MR) image, a remote sensing image, and a traffic sign image are used to test the algorithm’s performance. Compared with the traditional fuzzy C-means algorithm, RSFCM algorithm can effectively reduce noise interference, and has better robustness. In comparison with state-of-the-art fuzzy C-means algorithm, RSFCM algorithm could improve pixel separability, suppress heterogeneity of intra-class objects effectively, and it is more suitable for image segmentation. Keywords Clustering Fuzzy c-means algorithm Image segmentation Spatial context
& Jindong Xu [email protected] 1
School of Computer and Control Engineering, Yantai University, Yantai 264005, China
2
School of Opto-Electronic Information Science and Technology, Yantai University, Yantai 264005, China
1 Introduction The image uncertainty becomes a major problem in image processing [1]. FCM (Please see Appendix 1 for details of the abbreviations in the paper) algorithm can manage the ambiguity of data classification [2] and is widely used in remote sensing image analysis, ITS, image segmentation, and auxiliary medical diagnosis [3–6]. However, FCM clustering algorithm only takes into account the gray value of the target image pixels, and ignores the correlation between the pixel and its neighboring pixels [7]. Moreover, conventional FCM often produces clustering maps containing noises, and it is sensitive to noise pollution. Thus, conventional FCM has bad robustness. It is easy to fall into local minimum and cannot obtain global optimal solutions [8]. Thereby, FCM needs to be improved for practical image segmentation [9]. To effectively suppress noise interference and improve image segmentation quality, researchers have incorporated local spatial information into the original FCM algorithm, and obtained a series of research results [10, 11]. Ahmed et al. [12] proposed an FCM_S method by adding spatial neighborhood terms to the objective function of FCM. In order to reduce the computational complexity of FCM_S, Chen and Zhang [13] developed FCM_S1 and F
Data Loading...
