高光谱遥感图像分类算法中的应用研究

摘要

针对高光谱遥感图像数据量大、维数高、数据之间冗余量大的特点,提出一种基于决策边界特征提取( Decision Boundary Feature Extraction,DBFE)的SVM高光谱遥感图像分类算法.首先采用DBFE对高光谱遥感图像进行特征提取,消除特征之间相关性,并降低特征维数,然后采用GA对SVM参数进行优化,找到最优分类模型参数,最后采用最优分类模型对待分类的高光谱遥感图像进行分类.仿真结果表明,高光谱遥感图像分类算法提高了高光谱遥感图像分类的效率和分类正确率,说明分类方法是有效、可行的.%Because hyperspectral remote sensing images have large amount of data, high dimensions, and redundancy among the characteristics, this paper proposed one hyperspectral remote sensing image classification algorithm based on the Decision Boundary Feature Extraction (DBFE ) and SVM. Firstly, the DBFE was used for hyperspectral remote sensing image feature extraction, removing features between correlation and reducing the feature dimension. Then, GA was used to optimize SVM parameters and find the optimal classification model parameters. Finally, the optimal classification model was used to classify hyperspectral remote sensing image. Simulation experimental results show that the proposed algorithm can improve the classification efficiency and accuracy of hyperspectral remote sensing image, and is effective and feasible.