计算机工程2017，Vol.43Issue(1)：280-286,7.DOI:10.3969/j.issn.1000-3428.2017.01.048
基于GLCM与自适应Gabor滤波器组的纹理图像分割
Texture Image Segmentation Based on GLCM and Self-adaptive Gabor Filter Bank
摘要
Abstract
To solve the problem of parameter selection in the algorithm of texture image segmentation based on Gabor filter,a texture image segmentation algorithm is proposed in this paper,which predicts the number of texture types and the parameters of Gabor filter bank.Firstly,the image is divided into regional blocks.Then,the number of texture types is predicted by the texture feature vector of regional blocks,and the parameters of Gabor filter bank are predicted by the characteristics of various texture features.Finally,texture features of the original image is extracted by using the predicted filter bank,and the image is clustered and segmented based on the predicted number of texture types.Experimental results show that the proposed algorithm can process the segmentation in the texture image with higher precision and faster speed,and is less affected by the number of texture types.关键词
Gabor滤波器/纹理图像/纹理类型/灰度共生矩阵/模糊C均值聚类Key words
Gabor filter/texture image/texture type/Gray Level Co-occurrence Matrix (GLCM)/Fuzzy C-means (FCM) clustering分类
信息技术与安全科学引用本文复制引用
闵永智,程天栋,殷超,岳彪,肖本郁,马宏锋..基于GLCM与自适应Gabor滤波器组的纹理图像分割[J].计算机工程,2017,43(1):280-286,7.基金项目
国家自然科学基金“基于机器视觉的铁路轨道表面缺陷快速识别与分类方法研究”(61461023) （61461023）
国家自然科学基金“铁路长大隧道路基表面沉降链式图像监测方法及模型”(61663022) （61663022）
甘肃省高原信息工程及控制重点实验室开放课题基金“钢轨表面缺陷机器视觉快速检测”(20161105). （20161105）
