摘要
研究基于最小二乘小波孪生支持向量机(least squares wavelet twin support vector machines,简称LS-WTSVM)的遥感多光谱影像云检测.首先根据云在不同波段中大气的辐射特点,结合Landsat7 ETM+影像数据的光谱特性获得云像元的光谱特征,再通过提取每个图像块的灰度共生矩阵得到相应像元的纹理结构特征,根据像元的光谱特性和纹理结构特征构造特征向量,最后利用最小二乘小波孪生支持向量机多分类算法进行Landsat7 ETM+影像像元的云检测,实现不同类型云区的多分类识别.仿真实验结果表明,该算法能准确地检测出多光谱影像中的厚云和薄云.
In this paper, a novel cloud detection method for remote sensing multi-spectral image based on least squares wavelet twin support vector machines （LS-WTSVM） was proposed. Firstly, the spectral feature of the cloud pixel was acquired on the basis of the atmospheric radiation characteristics of cloud in different bands and the spectral characteristics of Landsat7 ETM ＋ image data. Then the texture feature of the corresponding pixel was obtained by extracting the gray level co-occurrence matrix of the each image block. Using the spectral properties and texture feature of pixels to construct the feature vectors and training LS-WTSVM multi-classification algorithm to detect the Landsat7 ETM＋ image cloud pixels, different types of cloud was multi-classified and recognized. Experimental results showed that this method could detect the thick cloud and thin cloud of multi- spectral image accurately.
出处
《安徽大学学报（自然科学版）》
CAS
北大核心
2014年第1期48-55,共8页
Journal of Anhui University(Natural Science Edition)
基金
国家自然科学基金资助项目(61172127)
安徽省教育厅重点科研计划资助项目(KJ2010A021)
安徽省自然科学基金资助项目(1208085QF104)
关键词
Landsat7影像
云检测
多分类
最小二乘小波孪生支持向量机
小波核
Landsat7 images
cloud detection
multi - classification
least squares wavelet twinsupport vector machines
wavelet kernel
