基于改进最小噪声分离变换的特征提取与分类  

Classification and feature extraction of hyperspectral images based on improved minimum noise fraction transformation

作　　者：白璘[1] 惠萌[1] 

机构地区：[1]长安大学电子与控制工程学院,陕西西安710064

出　　处：《计算机工程与科学》2015年第7期1344-1348,共5页Computer Engineering & Science

基　　金：中央高校基本科研业务费专项资金资助项目(CHD2011JC170);国家自然科学基金资助项目(41101357)

摘　　要：在最小噪声分离变换的基础上，引入核方法，采用小波核函数代替传统核函数对最小噪声分离变换予以改进。小波核函数的多分辨率分析特性可进一步提高算法的非线性映射能力。相关向量机高光谱图像分类是一种较新的高光谱图像分类方法，将新型核最小噪声分离变换方法与相关向量机相结合，对高光谱影像数据进行分类实验。仿真实验结果表明，基于小波核最小噪声分离变换的方法体现了高光谱影像的非线性特征，将所提出的方法应用于HYDICE系统在Washington DC Mall上空拍摄的数据，与对照算法相比，分类精度可提高3％～89／6，并可有效地提高小样本区域的分类精度。Based on minimum noise fraction transformation, we introduce a novel wavelet kernel method, which improves the minimum noise fraction transformation by replacing the traditional kernel function with the wavelet kernel function, for its feature of multi-resolution analysis can improve the nonlinear mapping capability of the kernel minimum noise fraction transformation method. The relevance vector machine classification of hyperspectral images is a new classification method which combines the novel kernel minimum noise fraction transformation with the relevance vector machine. Simulation re- sults show that, the wavelet kernel minimum noise fraction transformation method reflects the nonlinear characteristics of the hyperspectral images. The proposed method is applied to the HYDICE data （shoot over in Washington DC Mall）, and compared with the compare algorithm, its classification accuracy can be increased by 3%～8% and the classification precision of areas with small sample data can be improved effectively.

关 键 词：相关向量机 高光谱图像分类 核方法 最小噪声分离变换 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]