This paper presents a novel classification method of hyperspectral image(HSI) based on EMAPs and SMLR. Firstly, we adopt EMAPs(Extended Morphological multi-Attribute Profiles) algorithm to extract the spatial information of HSI efficiently, and combine the spectral information to form spatial-spectral features fusion model. EMAPs can replace simple structural elements with multiple attributes structure and cascade them to obtain attributes feature of multiple structures. Then we utilize SMLR(Sparse Multinomial Logistic Regression) for HSI classification. SMLR is applicable to high-dimensional and large data sets. A multiclass classifier based on MLR is adopted, and a fast algorithm is used to learn a sparse multiclass classifier. Compared with other methods in HSI experiments, our method provides an excellent result.
A rotation SVM ensemble learning method based on dimension reduction combination is proposed aiming at the problem of dimensionality disaster and low classification accuracy of hyperspectral remote sensing images. PCA algorithm is used to reduce dimension firstly. This method can not only eliminate data redundancy and retain the main information but also realize non-singularity of intra-class distance matrix that LDA required. Then LDA method is used to reduce dimension based on projection secondly. The twice dimension reduction produce the minimum intra-class distance, the maximum inter-class distance and the best discrimination of samples. Then, the sample after twice dimension reduction is classified by Rotation SVM ensemble learning algorithm. SVM is use to be base classifier because of its high classification accuracy. In order to enhance the samples diversity of ensemble classifiers, rotation matrix is constructed firstly, and then SVM classifier is trained on each rotation matrix. Above steps are performed repeatedly, classification result is produced by voting method finally. Experimental results based on Indian Pines hyperspectral images show that the reduction dimension effect and classification accuracy of the proposed method are better than other classification methods.
In this paper, we proposed a new classification method based on support vector machine (SVM) combined with multi-scale segmentation. The proposed method obtains satisfactory segmentation results which are based on both the spectral characteristics and the shape parameters of segments. SVM method is used to label all these regions after multiscale segmentation. It can effectively improve the classification results. Firstly, the homogeneity of the object spectra, texture and shape are calculated from the input image. Secondly, multi-scale segmentation method is applied to the RS image. Combining graph theory based optimization with the multi-scale image segmentations, the resulting segments are merged regarding the heterogeneity criteria. Finally, based on the segmentation result, the model of SVM combined with spectrum texture classification is constructed and applied. The results show that the proposed method can effectively improve the remote sensing image classification accuracy and classification efficiency.
Spectral unmixing technique is one of the key techniques to identify and classify the material in the hyperspectral image processing. A novel robust spectral unmixing method based on nonnegative matrix factorization(NMF) is presented in this paper. This paper used an edge-preserving function as hypersurface cost function to minimize the nonnegative matrix factorization. To minimize the hypersurface cost function, we constructed the updating functions for signature matrix of end-members and abundance fraction respectively. The two functions are updated alternatively. For evaluation purpose, synthetic data and real data have been used in this paper. Synthetic data is used based on end-members from USGS digital spectral library. AVIRIS Cuprite dataset have been used as real data. The spectral angle distance (SAD) and abundance angle distance(AAD) have been used in this research for assessment the performance of proposed method. The experimental results show that this method can obtain more ideal results and good accuracy for spectral unmixing than present methods.