Spatial-Spectral Joint Classification of Hyperspectral Image With Locality and Edge Preserving

Abstract

Hyperspectral image (HSI) classification is an important part of its processing and application. Aiming at the problems of high data dimensionality and high spatial neighborhood correlation in HSI classification, we propose a spatial-spectral joint classification method of HSI with locality and edge preserving in this article. First, the input HSI is normalized, and the feature is extracted by principal component analysis. The first principal component image is taken as the guidance image. Second, guided filtering is used to extract the spatial features of each band separately. Then, the extracted spatial features are superimposed, and low-dimensional embedding is completed through local Fisher discriminant analysis. Finally, the obtained low-dimensional embedded features are input into a random forest classifier to get classification results. The experimental results of two HSI show that the proposed method achieves higher classification accuracy than other related methods. In the case of randomly selecting 10% and 1% samples from each class of ground object as training samples, the overall classification accuracy is improved to 99.57% and 97.79%, respectively. This method effectively uses the spatial and local information of the image in low dimensional embedding, and preserves the boundaries of the ground objects, thus improving the classification effect.

A Novel Band Selection and Spatial Noise Reduction Method for Hyperspectral Image Classification

“…To highlight the superiority of the proposed NGNMF-E2DSSA method, in this subsection, we compare a few stateof-the-art spectral-spatial methods, i.e. the SuperPCA [55], GF-LFDA [36], IAPs [56] and SSMRPE [17], while the raw data with SVM was set as the baseline method (abbreviated as "SVM"). These state-of-the-art methods can jointly utilize the spectral and spatial information and reduce the dimensions of the HSI data.…”

“…These state-of-the-art methods can jointly utilize the spectral and spatial information and reduce the dimensions of the HSI data. We downloaded the source code or wrote the code of each compared method and set the parameters optimally as suggested in [17,36,55,56] to generate the results. The embedding window of E2DSSA was set to 81 for all three datasets.…”

“…In [35], a novel ICA and EMAP based spectral-spatial DR method was proposed, in which spectral features extracted by ICA was suitable for representing each class, whilst the improved and reduced AP (rAP) could further address the spatial redundancy of EMAP. In [36], the guided filter (GF) was used to extract spatial context information, with the LFDA for lowdimensional embedding. Zheng et.al [18] performed graphbased feature extraction on the selected band subset, which achieved superior classification results yet with a high computational cost.…”

Spectral–Spatial Joint Feature Extraction for Hyperspectral Image Based on High-Reliable Neighborhood Structure

“…Finally, manifold reconstruction was performed and the low-dimensional discriminative features are extracted for classification. Zhang et al [38] first get the features by applying PCA on the normalized input HSI and utilized guided filtering to extract the spatial features of each band separately. Then, the extracted spatial features are superimposed, and low-dimensional embedding is completed through LFDA for spectral-spatial joint classification of HSI.…”