基于纹理特征MNF变换的多光谱遥感影像分类  

Multi-Spectral Remote Sensing Image Classification Based on Texture Feature MNF Transform

作　　者：李亚坤[1] 关洪军[1] 孙传亮[2] 敖志刚[1] LI Ya-kun GUAN Hong-jun SUN Chuan-liang AO Zhi-gang(Field Engineering Institute, PLA University of Science and Technology, Nanjing 210007, China College of Resources and Environmental Sciences, Nanjing Agricultural University, Nanjing 210095, China)

机构地区：[1]解放军理工大学野战工程学院,南京210007 [2]南京农业大学资源与环境科学学院,南京210095

出　　处：《兵器装备工程学报》2017年第2期113-117,131,共6页Journal of Ordnance Equipment Engineering

摘　　要：提出了一种基于纹理特征最小噪声分离(Minimum Noise Fraction,MNF)变换的多光谱遥感影像分类方法。利用灰度共生矩阵对每个光谱波段进行纹理特征提取,对纹理特征进行MNF变换,将集中了大部分纹理特征信息的MNF分量与光谱信息协同进行分类。基于CBERS-04遥感影像对郎玛村地区进行岩土分类实验。结果表明,该方法的分类精度高于传统的基于光谱主成分纹理特征的多光谱遥感影像分类方法,其分类结果具有更好的区域一致性和较少的小图斑噪声。A classification method based on Minimum Noise Fraction （MNF） transform of texture feature was proposed. Firstly, the gray level co-occurrence matrix was used to extract texture features of every spectral band. Secondly, the texture features were transformed by using MNF. The MNF components, which have the majority of the texture feature information, would be used to assist spectral information in Classifying. Geotechnical classification was studied based on CBERS-04 remote sensing image in Langma village area. Results show that the classification accuracy of the proposed method is higher than that of traditional classification method based on spectral principal component texture features, and the classification results have better regional consistency and less noise of small map spot.

关 键 词：纹理特征 最小噪声分离(MNF) 多光谱遥感影像 岩土分类 CBERS-04 

分 类 号：TJ02[兵器科学与技术—兵器发射理论与技术] TP751[自动化与计算机技术—检测技术与自动化装置]