【摘要】 针对tri_training协同训练算法在小样本的高光谱遥感影像半监督分类过程中,存在增选样本的误标记问题,提出一种基于空间邻域信息的半监督协同训练分类算法tri_training_SNI(tri_training based on Spatial Neighborhood Information)。首先利用分类器度量方法不一致度量和新提出的不一致精度度量从MLR(Multinomial Logistic Regression)、KNN(k-Nearest Neighbor)、ELM(Extreme Learning Machine)和RF(Random Forest)4个分类器中选择3分类性能差异性最大的3个分类器;然后在样本选择过程中,采用选择出来的3个分类器,在两个分类器分类结果相同的基础上,加入初始训练样本的8邻域信息进行未标记样本的二次筛选和标签的确定,提高了半监督学习的样本选择精度。通过对AVIRIS和ROSIS两景高光谱遥感影像进行分类实验,结果表明与传统的tri_training协同算法相比,该算法在分类精度方面有明显提高。更多还原

【Abstract】 In the process of hyperspectral image classification using the tri_training algorithm,the labels of unlabeled samples have error labels when the amount of initial training samples is small.In this paper,we propose a novel tri_training based on spatial neighborhood information(tri_training_SNI)to solve the problem for the tri_training algorithm.Firstly,we choose three basic classifiers from MLR(Multinomial Logistic Regression),KNN(k-Nearest Neighbor),ELM(Extreme Learning Machine)and RF(Random Forest)classifier based on disagreement measure and disagreement-accuracy.These classifiers are redefined using unlabeled samples in the tri_training_SNI process.In detail,in each round of tri_training_SNI,unlabeled samples are labeled for a classifier by the following two steps.Step 1:the first selection of unlabeled samples is constructed under certain conditions that the other two classifiers have the same labels.Step 2:spatial Neighborhood Information of initial training samples based on 8-neighborhood is applied in this proposed approach to construct the secondary selection of unlabeled samples and the labels of unlabeled samples.Then the final classification results are produced via majority voting by the classification results of three classifiers.Experiments on two real hyperspectral data indicate that the proposed approach can effectively improve classification performance.更多还原