CMR-CNN: Cross-Mixing Residual Network for Hyperspectral Image Classification

Abstract

With the development of deep learning, various convolutional neural network (CNN) based methods have been proposed for the hyperspectral image (HSI) classification. Although most of them achieve good classification performance, there are still more misclassifications in the prediction map with fewer training samples. In order to address this shortcoming, this paper proposes to simultaneously use pixels' spatial information and spectral information for HSI classification. Briefly speaking, a new cross-mixing residual network denoted by CMR-CNN is developed, wherein one three-dimensional (3D) residual structure responsible for extracting the spectral characteristics, one twodimensional (2D) residual structure responsible for extracting the spatial characteristics, and one assisted feature extraction (AFE) structure responsible for linking the first two structures are respectively designed. With respect to experiments performed on five different datasets Indian Pines, University of Pavia, Salinas Scene, KSC, and Xuzhou in the case of different numbers of training samples show that, compared to some state-of-theart methods, CMR-CNN can achieve higher overall accuracy (OA), average accuracy (AA), and Kappa values. Particularly, compared with the newly proposed HSI classification methods OCT-MCNN, CMR-CNN respectively improves OA, AA and kappa by 4.13%, 3.67%, and 2.75% on average.

Network Collaborative Pruning Method for Hyperspectral Image Classification Based on Evolutionary Multi-Task Optimization

“…In this part, the proposed method is validated on more complex networks and larger HSI dataset. The proposed method is used to prune the complex network CMR-CNN [69] for HSI classification, the number of parameters is 28,779,784. A new cross-mixing residual network denoted by CMR-CNN is developed, wherein one three-dimensional (3D) residual structure responsible for extracting the spectral characteristics, one two-dimensional (2D) residual structure responsible for extracting the spatial characteristics, and one assisted feature extraction (AFE) structure responsible for linking the first two structures are designed.…”

Random Shuffling Data for Hyperspectral Image Classification with Siamese and Knowledge Distillation Network

“…Inspired by the Dense Convolutional Network, Wang et al [16] proposed a fast dense spectralspatial convolution (FDSSC) algorithm. In another work [17], a network named CMR-CNN adopted the 3D residual blocks followed by the 2D residual blocks together to capture the spatial-spectral feature information of the HSI. To reduce the network parameters and computational complexity, Meng et al [18] proposed a lightweight spectral-spatial convolution HSI classification module (LS2CM), and Li et al [19] designed a lightweight network architecture (LiteDenseNet).…”