Robust multi‐view representation for spatial–spectral domain in application of hyperspectral image classification

Abstract

Spatial-spectral representation plays an important role in hyperspectral images (HSIs) classification. However, many of the existing local feature algorithms for HSIs are based on the two-dimensional image and do not take full advantage of the information hidden in HSI, such as spatial-spectral locality correlation information, thereby reducing the robustness of these algorithms. In response to these problems, this study presents a robust multi-view spatial-spectral representation method with the characteristics of HSIs. There are two key techniques in this representation method, called spatial-spectral locality constrained linear coding (SSLLC) and spatial-spectral pyramid matching model (SSPM). Firstly, SSLLC applies the locality information of the feature points and visual words and uses the discriminant information provided by the nearest-neighbouring spatial-spectral feature points in HSIs. Secondly, SSPM works by partitioning the image into increasingly fine sub-cubes and uses the cubes to match the local features of the HSIs. The multi-view representation is tolerant to illumination change, image rotation, affine distortion etc. To assess the validity of authors' algorithm, the authors compared their results with several existing approaches, including a deep learning method. The experimental results show that this representation method can effectively improve the accuracy of HSIs classification.

High Spatial Resolution Remote Sensing Data Classification Method Based on Spectrum Sharing

“…In the formula, the key point of feature extraction is to capture the location of the central pixel, design and extract the tunnel to focus the position of the central pixel Z ij on the T ij , convert the spectral feature vectors after the input and convolution data output to one-dimensional vectors, and then take r as the radius input data, then take the pixel as the center point and Z ij � R k×k (k � 2R + 1) as the initial data input to the designed feature extraction tunnel, and then realize the spectral feature extraction in the region with the central pixel Z ij as the target [30]. e design formula is as follows:…”

CNN‐combined graph residual network with multilevel feature fusion for hyperspectral image classification

“…It could, respectively, learn the global and local features from the whole HSI and the most critical area so as to make full use of this information. Again, more classifying HSI methods via deep learning have also emerged in the following literature, including spectral‐spatial LSTM networks [37], generative adversarial network [38], robust multiview representation for spatial‐spectral domain [39], regularised transfer learning [40] and feature‐grouped network with spectral‐spatial connected attentions [41] etc.…”