鉴于传统高光谱影像分类大都采用监督学习方法,且仅利用了光谱信息,未考虑影像空间特征和流形结构。提出一种基于空-谱协同流形重构误差的高光谱影像分类方法,该算法基于高光谱影像中地物分布的空间一致性,利用少量标记的样本和大量的...鉴于传统高光谱影像分类大都采用监督学习方法,且仅利用了光谱信息,未考虑影像空间特征和流形结构。提出一种基于空-谱协同流形重构误差的高光谱影像分类方法,该算法基于高光谱影像中地物分布的空间一致性,利用少量标记的样本和大量的无标记空间近邻样本来进行半监督学习,并利用测试样本在每一子流形上的重构误差来表征相似性,实现鉴别分类。在Indian Pines和University of Pavia数据集上的实验结果表明,本文方法的分类精度在各种条件下要优于其他分类算法,其最高总体精度分别达到了95.67%和91.92%。该算法将高光谱遥感影像中的空间-光谱信息融入不同地物的子流形结构表征,在训练样本数量较少时仍能得到好的分类效果,有效提升了分类性能。展开更多
Aiming at the slow convergence and low accuracy problems of the traditional non-negative tensor factorization, a local hierarchical non-negative tensor factorization method is proposed by applying the local objective ...Aiming at the slow convergence and low accuracy problems of the traditional non-negative tensor factorization, a local hierarchical non-negative tensor factorization method is proposed by applying the local objective function theory to non- negative tensor factorization and combining the three semi-non- negative matrix factorization（NMF） model. The effectiveness of the method is verified by the facial feature extraction experiment. Through the decomposition of a series of an air compressor＇s vibration signals composed in the form of a bispectrum by this new method, the basis images representing the fault features and corresponding weight matrices are obtained. Then the relationships between characteristics and faults are analyzed and the fault types are classified by importing the weight matrices into the BP neural network. Experimental results show that the accuracy of fault diagnosis is improved by this new method compared with other feature extraction methods.展开更多
To minimize the low classification accuracy and low utilization of spatial information in traditional hyperspectral image classification methods, we propose a new hyperspectral image classification method, which is ba...To minimize the low classification accuracy and low utilization of spatial information in traditional hyperspectral image classification methods, we propose a new hyperspectral image classification method, which is based on the Gabor spatial texture features and nonparametric weighted spectral features, and the sparse representation classification method（Gabor–NWSF and SRC）, abbreviated GNWSF–SRC. The proposed（GNWSF–SRC） method first combines the Gabor spatial features and nonparametric weighted spectral features to describe the hyperspectral image, and then applies the sparse representation method. Finally, the classification is obtained by analyzing the reconstruction error. We use the proposed method to process two typical hyperspectral data sets with different percentages of training samples. Theoretical analysis and simulation demonstrate that the proposed method improves the classification accuracy and Kappa coefficient compared with traditional classification methods and achieves better classification performance.展开更多
Based on the observation that there exists multiple information in a pixel neighbor,such as distance sum and gray difference sum,local information enhanced LBP(local binary pattern)approach,i.e.LE-LBP,is presented.Geo...Based on the observation that there exists multiple information in a pixel neighbor,such as distance sum and gray difference sum,local information enhanced LBP(local binary pattern)approach,i.e.LE-LBP,is presented.Geometric information of the pixel neighborhood is used to compute minimum distance sum.Gray variation information is used to compute gray difference sum.Then,both the minimum distance sum and the gray difference sum are used to build a feature space.Feature spectrum of the image is computed on the feature space.Histogram computed from the feature spectrum is used to characterize the image.Compared with LBP,rotation invariant LBP,uniform LBP and LBP with local contrast,it is found that the feature spectrum image from LE-LBP contains more details,however,the feature vector is more discriminative.The retrieval precision of the system using LE-LBP is91.8%when recall is 10%for bus images.展开更多
文摘鉴于传统高光谱影像分类大都采用监督学习方法,且仅利用了光谱信息,未考虑影像空间特征和流形结构。提出一种基于空-谱协同流形重构误差的高光谱影像分类方法,该算法基于高光谱影像中地物分布的空间一致性,利用少量标记的样本和大量的无标记空间近邻样本来进行半监督学习,并利用测试样本在每一子流形上的重构误差来表征相似性,实现鉴别分类。在Indian Pines和University of Pavia数据集上的实验结果表明,本文方法的分类精度在各种条件下要优于其他分类算法,其最高总体精度分别达到了95.67%和91.92%。该算法将高光谱遥感影像中的空间-光谱信息融入不同地物的子流形结构表征,在训练样本数量较少时仍能得到好的分类效果,有效提升了分类性能。
基金The National Natural Science Foundation of China(No.50875078)the Natural Science Foundation of Jiangsu Province(No.BK2007115)the National High Technology Research and Development Program of China(863 Program)(No.2007AA04Z421)
文摘Aiming at the slow convergence and low accuracy problems of the traditional non-negative tensor factorization, a local hierarchical non-negative tensor factorization method is proposed by applying the local objective function theory to non- negative tensor factorization and combining the three semi-non- negative matrix factorization（NMF） model. The effectiveness of the method is verified by the facial feature extraction experiment. Through the decomposition of a series of an air compressor＇s vibration signals composed in the form of a bispectrum by this new method, the basis images representing the fault features and corresponding weight matrices are obtained. Then the relationships between characteristics and faults are analyzed and the fault types are classified by importing the weight matrices into the BP neural network. Experimental results show that the accuracy of fault diagnosis is improved by this new method compared with other feature extraction methods.
基金supported by the National Natural Science Foundation of China(No.61275010)the Ph.D.Programs Foundation of Ministry of Education of China(No.20132304110007)+1 种基金the Heilongjiang Natural Science Foundation(No.F201409)the Fundamental Research Funds for the Central Universities(No.HEUCFD1410)
文摘To minimize the low classification accuracy and low utilization of spatial information in traditional hyperspectral image classification methods, we propose a new hyperspectral image classification method, which is based on the Gabor spatial texture features and nonparametric weighted spectral features, and the sparse representation classification method（Gabor–NWSF and SRC）, abbreviated GNWSF–SRC. The proposed（GNWSF–SRC） method first combines the Gabor spatial features and nonparametric weighted spectral features to describe the hyperspectral image, and then applies the sparse representation method. Finally, the classification is obtained by analyzing the reconstruction error. We use the proposed method to process two typical hyperspectral data sets with different percentages of training samples. Theoretical analysis and simulation demonstrate that the proposed method improves the classification accuracy and Kappa coefficient compared with traditional classification methods and achieves better classification performance.
基金Project(61372176,51109112)supported by the National Natural Science Foundation of ChinaProject(2012M520277)supported by theChina Postdoctoral Science Foundation
文摘Based on the observation that there exists multiple information in a pixel neighbor,such as distance sum and gray difference sum,local information enhanced LBP(local binary pattern)approach,i.e.LE-LBP,is presented.Geometric information of the pixel neighborhood is used to compute minimum distance sum.Gray variation information is used to compute gray difference sum.Then,both the minimum distance sum and the gray difference sum are used to build a feature space.Feature spectrum of the image is computed on the feature space.Histogram computed from the feature spectrum is used to characterize the image.Compared with LBP,rotation invariant LBP,uniform LBP and LBP with local contrast,it is found that the feature spectrum image from LE-LBP contains more details,however,the feature vector is more discriminative.The retrieval precision of the system using LE-LBP is91.8%when recall is 10%for bus images.
