| 引用本文: | 王明威,万幼川,高贤君,等.纹理影像特征选择及K-means聚类优化方法.[J].国防科技大学学报,2017,39(6):152-159.[点击复制] |
| WANG Mingwei,WAN Youchuan,GAO Xianjun,et al.Texture image feature selection and optimization by using K-means clustering[J].Journal of National University of Defense Technology,2017,39(6):152-159[点击复制] |
|
| |
|
|
| 本文已被：浏览 7519次 下载 6797次 |
| 纹理影像特征选择及K-means聚类优化方法 |
| 王明威1, 万幼川1, 高贤君2, 叶志伟3 |
|
(1. 武汉大学 遥感信息工程学院， 湖北 武汉 430079;2. 长江大学 地球科学学院， 湖北 武汉 430100;3. 湖北工业大学 计算机学院， 湖北 武汉 430068)
|
| 摘要: |
| Gabor变换和K-means算法是最为常用的纹理分析方法。然而，采用Gabor变换得到的纹理特征向量具有较高的维数，影响算法的运行效率；K-means算法也易受初始类中心的影响而导致分类精度下降。因此，通过Relief算法对采用Gabor变换所提取的纹理特征进行选择，得到合适的纹理特征子集。进一步采用差分进化算法，对K-means算法的聚类中心进行优化从而提高纹理识别精度和效率。实验结果表明：提出的方法所需用到的纹理特征向量的维数相对于原始特征集有大幅降低，较之基本的K-means算法，纹理识别的精度也有较明显的提高。 |
| 关键词: 纹理识别 Gabor变换 K-means算法 Relief算法 差分进化算法 |
| DOI：10.11887/j.cn.201706022 |
| 投稿日期：2016-09-12 |
| 基金项目:国家科技支撑计划资助项目（2014BAL05B07）；国家自然科学基金资助项目（61301278）；长江大学青年基金资助项目（2016cqn04） |
|
| Texture image feature selection and optimization by using K-means clustering |
| WANG Mingwei1, WAN Youchuan1, GAO Xianjun2, YE Zhiwei3 |
|
(1. School of Remote Sensing and Information Engineering, Wuhan University, Wuhan 430079, China;2. School of Geoscience, Yangtze University, Wuhan 430100, China;3. School of Computer Science, Hubei University of Technology, Wuhan 430068, China)
|
| Abstract: |
| Gabor transform and K-means algorithm are two commonly used texture analysis methods. However, the texture feature vector has a high dimension by using Gabor transform, which will influence the operating efficiency. Meanwhile, K-means algorithm is affected by the initial clustering centers, and it may lead to the decrease of classification accuracy. Although, some optimization algorithms like genetic algorithm and particle swarm optimization algorithm could improve the performance of K-means algorithm to some extent, the optimization effect is difficult to guarantee as the increase of dimension. Hence, the Relief algorithm was applied to make a feature selection for Gabor texture feature, and to obtain a suitable texture feature subset. Furthermore, a differential evolution algorithm was used to optimize the clustering center of K-means algorithm, and enhance the accuracy and efficiency of texture recognition. Experimental results demonstrate that the dimension of texture feature vector by using the proposed method is obviously lower than that by using the original feature set, and the recognition accuracy is also apparently improved than the basic K-means algorithm. |
| Keywords: texture recognition Gabor transform K-means algorithm Relief algorithm differential evolution algorithm |
|
|
|
|
|
