考虑到无损检测图像具有易受噪声干扰且目标占图像面积较小的特性,以图像块为基本单元,提出一种较强鲁棒性和自适应性的抑制式模糊C均值算法用于无损检测图像的分割.首先,对图像块内像素的权重进行自适应确定,其权重受图像块内像素的空间距离和灰度值大小的影响;然后构建了图像块的模糊不确定性模型,并以图像块为基本单元将其引入至新的目标函数并进行求解,给出算法的执行流程;最后采用无损检测图像进行实验,结果显示所提出的算法具有较好的鲁棒性和有效性.
Considering that the non-destructive testing(NDT)image is easy to be disturbed by noise and the target occupies a small area of the image, a robust and adaptive suppressed fuzzy C-means algorithm based on the image patch(SFCMP)is proposed to segment NDT image. Firstly, the weight of the pixels in the image patch is adaptively determined, which is affected by the spatial distance and gray value of each pixel in the image patch. Then the fuzzy uncertainty model of image patch is constructed, and it is introduced into the new objective function with image patch as the basic unit, and the implementation process of the SFCMP is given. Finally, experiments are carried out on the NDT images, and the results show that the SFCMP has good robustness and effectiveness.
[1] 李喜孟.无损检测[M].北京:机械工业出版社, 2011.
[2] 侯旺,钟立军,张小虎,等.红外目标分割方法研究[J].国防科技大学学报, 2013, 35(2): 173- 178.
[3] FAN J L, ZHEN W Z, XIE W X. Suppressed fuzzy c-means clustering algorithm[J]. Pattern Recognit Lett, 2003, 24(9/10): 1607-1612. DOI: 10.1016/S0167-8655(02)00401-4.
[4] HUNG W L, YANG M S, CHEN D H. Parameter selection for suppressed fuzzy c-means with an application to MRI segmentation[J]. Pattern Recognit Lett, 2006, 27(5): 424-438. DOI: 10.1016/j.patrec.2005.09.005.
[5] 朱占龙,刘永军,赵战民,等.用于分割无损检测图像的改进的抑制式模糊C均值聚类算法[J].仪器仪表学报, 2019, 40(8): 110- 118. DOI: 10.19650/j.cnki.cjsi.J1905078.
[6] LIU Y, HOU T, LIU F. Improving fuzzy c-means method for unbalanced dataset[J]. Electron Lett, 2015, 51(23): 1880-1882. DOI: 10.1049/el.2015.1541.
[7] NOORDAM J C, VAN DEN BROEK W H A M, BUYDENS L M C. Multivariate image segmentation with cluster size insensitive Fuzzy C-means[J]. Chemom Intell Lab Syst, 2002, 64(1): 65-78. DOI: 10.1016/S0169-7439(02)00052-7.
[8] JI Z X, XIA Y, CHEN Q, et al. Fuzzy c-means clustering with weighted image patch for image segmentation[J]. Appl Soft Comput, 2012, 12(6): 1659-1667. DOI: 10.1016/j.asoc.2012.02.010.
[9] LIU Y L, WANG J, CHEN X, et al. A robust and fast non-local means algorithm for image denoising[J]. J Comput Sci & Tech, 2008, 23(2): 270-279. DOI: 10.1007/s11390-008-9129-8.
[10] 朱占龙,王军芬.基于自适应模糊C均值与后处理的图像分割算法[J].激光与光电子学进展, 2018, 55(1): 213-222. DOI: 10.3788/lop55.011004.
[11] 陆海青,葛洪伟.自适应灰度加权的鲁棒模糊C均值图像分割[J].智能系统学报, 2018, 13(4): 584-593. DOI: 10.11992/tis.201701008.
[12] 潘金艳,高朋,高云龙,等.基于可靠性的鲁棒模糊聚类[J].控制理论与应用, 2021, 38(4): 516-528. DOI: 10.7641/CTA.2020.00480. (
