Cascaded Recurrent Neural Networks for Hyperspectral Image Classification

By Renlong Hang, Qingshan Liu, Danfeng Hong, Pedram Ghamisi. IEEE Transactions on Geoscience and Remote Sensing, 2017

680 citations recorded by the source corpus.

By considering the spectral signature as a sequence, recurrent neural networks (RNNs) have been successfully used to learn discriminative features from hyperspectral images (HSIs) recently. However, most of these models only input the whole spectral bands into RNNs directly, which may not fully explore the specific properties of HSIs. In this paper, we propose a cascaded RNN model using gated recurrent units to explore the redundant and complementary information of HSIs. It mainly consists of two RNN layers. The first RNN layer is used to eliminate redundant information between adjacent spectral bands, while the second RNN layer aims to learn the complementary information from nonadjacent spectral bands. To improve the discriminative ability of the learned features, we design two strategies for the proposed model. Besides, considering the rich spatial information contained in HSIs, we further extend the proposed model to its spectral–spatial counterpart by incorporating some convolutional layers. To test the effectiveness of our proposed models, we conduct experiments on two widely used HSIs. The experimental results show that our proposed models can achieve better results than the compared models.

Subfields: Computer Vision and Pattern Recognition, Media Technology

View Paper: https://www.semanticscholar.org/paper/9ed92e11c78d1f96bf364ced84ae8a07b56ac321

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

In Media Technology

Citations in the corpus

A Spectral-Spatial-Dependent Global Learning Framework for Insufficient and Imbalanced Hyperspectral Image Classification (2021, methodology): “ConvLSTM was proposed to extract dependent features of the spectrums and geometric [32], [33].”
Spectral–Spatial Feature Tokenization Transformer for Hyperspectral Image Classification (2022, methodology): “Moreover, as a feedforward neural network, the recurrent neural network (RNN) [49] was studied for the HSI clas-siﬁcation task.”
Hyperspectral Image Classification—Traditional to Deep Models: A Survey for Future Prospects (2021, methodology): “Recently, [207], introduced a cascade RNN for HSIC which consist of two layers of GRU-based RNN, the first layer is used to reduce the redundant spectral bands and the second layer is used to learn the features from HSI, furthermore, few convolutional layers are employed to”
Deep learning in remote sensing: a review (2017, background): “RECURRENT NEURAL NETWORKS FOR HyPERSPECTRAL IMAGE CLASSIFICATION In [36], the authors propose an RNN model with a new activation function and modified gated recurrent unit for hyperspectral image classification that can effectively analyze hyperspectral pixels as sequential data and then determine i”
SpectralFormer: Rethinking Hyperspectral Image Classification With Transformers (2021, methodology): “For example, 1) although transformers perform outstandingly in solving the problem of long-term dependencies of spectral signatures, they lose the power of capturing locally contextual or semantic characteristics (cf. CNNs or RNNs); 2) as mentioned in [27], the skip connection plays a crucial role i”
Deep Few-Shot Learning for Hyperspectral Image Classification (2019, background): “deep belief networks [35], recurrent neural networks [36], 1-D CNN [37], 2-D CNN [38]–[41], and 3-D CNN [42]–[44]”
Spatial-Spectral Transformer for Hyperspectral Image Classification (2021, methodology): “Therefore, using CNN can lead to information loss when dealing with hyperspectral pixel vectors [36].”
Deep Learning for Hyperspectral Image Classification: An Overview (2019, neutral): “[106].”
Feature Extraction for Hyperspectral Imagery: The Evolution From Shallow to Deep: Overview and Toolbox (2020, background): “Different from them, Hang et al. proposed a serial framework to fuse CNNs and RNNs [122].”
The color out of space: learning self-supervised representations for Earth Observation imagery (2020, background): “Among works focusing on how to exploit spectral bands, [17] devises Recurrent Neural Networks (RNNs) to handle the redundancy underlying adjacent spectral channels.”
Learning Spectral-Spatial-Temporal Features via a Recurrent Convolutional Neural Network for Change Detection in Multispectral Imagery (2018, methodology): “and RNN [22], [23], [37]–[39] are well-established techniques for remote sensing applications, to the best of our”
Multiscale Dynamic Graph Convolutional Network for Hyperspectral Image Classification (2019, methodology): “address this defect, deep learning [15], [16], [17], [18] has been extensively employed for hyperspectral image classification and has attracted increasing attention for its strong representation ability.”
Hyperspectral Image Classification With Deep Learning Models (2018, background): “[41] proposed a novel RNN model for hyperspectral image classification, which could effectively analyzed hyperspectral pixels as sequential data and then determined information categories via network reasoning.”