Nonlocal Band Attention Network for Hyperspectral Image Band Selection

Li, Tiancong; Cai, Yaoming; Cai, Zhihua; Liu, Xiaobo; Hu, Qiubo

doi:10.1109/jstars.2021.3065687

Cited by 21 publications

(4 citation statements)

References 41 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To reveal DMuCA in key bands perception, we further perform HSI classification by DMuCA with full or selected bands (reference the results from [40,41]) as the input, and compare it with the ResNet base (Base). The results in Table 6 show that DMuCA achieves a significant increase over the base when performing classification with the full spectral bands, and a slight increase with the selected bands.…”

Section: Ablation Studymentioning

confidence: 99%

A Dual Multi-Head Contextual Attention Network for Hyperspectral Image Classification

Liang

et al. 2022

Remote Sensing

View full text Add to dashboard Cite

To learn discriminative features, hyperspectral image (HSI), containing 3-D cube data, is a preferable means of capturing multi-head self-attention from both spatial and spectral domains if the burden in model optimization and computation is low. In this paper, we design a dual multi-head contextual self-attention (DMuCA) network for HSI classification with the fewest possible parameters and lower computation costs. To effectively capture rich contextual dependencies from both domains, we decouple the spatial and spectral contextual attention into two sub-blocks, SaMCA and SeMCA, where depth-wise convolution is employed to contextualize the input keys in the pure dimension. Thereafter, multi-head local attentions are implemented as group processing when the keys are alternately concatenated with the queries. In particular, in the SeMCA block, we group the spatial pixels by evenly sampling and create multi-head channel attention on each sampling set, to reduce the number of the training parameters and avoid the storage increase. In addition, the static contextual keys are fused with the dynamic attentional features in each block to strengthen the capacity of the model in data representation. Finally, the decoupled sub-blocks are weighted and summed together for 3-D attention perception of HSI. The DMuCA module is then plugged into a ResNet to perform HSI classification. Extensive experiments demonstrate that our proposed DMuCA achieves excellent results over several state-of-the-art attention mechanisms with the same backbone.

show abstract

Section: Ablation Studymentioning

confidence: 99%

A Dual Multi-Head Contextual Attention Network for Hyperspectral Image Classification

Liang

et al. 2022

Remote Sensing

View full text Add to dashboard Cite

show abstract

“…In order to better capture the correlation between spatial pixels and spectral bands in HSI, the network-driven methods began to introduce attention mechanisms to various tasks of hyperspectral images [31]- [33]. Hu et al proposed squeeze-and-excitation(SE) attention to adjust the weights of different feature map channels by learning the relationship between channels, therefore improving the expressive ability of the network [34].…”

Section: Attention Mechanismmentioning

confidence: 99%

Hyperspectral Snapshot Compressive Imaging With Dense Back-Projection Joint Attention Network

Sun

Huang

Zhao

et al. 2022

IEEE J. Sel. Top. Appl. Earth Observations Remote Sensing

View full text Add to dashboard Cite

Hyperspectral Snapshot Compressive Imaging (SCI) system encodes three-dimensional hyperspectral images into a single two-dimensional snapshot measurement and then decodes the underlying 3D hyperspectral images by solving the compressive sensing reconstruction problem. Practical applications of SCI imaging systems require fast and high-quality reconstruction.To meet this requirement, we propose a novel encoder and decoder network with dense back-projection joint attention for hyperspectral SCI. The main contributions of our network lie in two aspects. First, we propose a dense back-projection module and deploy it in an encoder with five scales. It computes the back-projection between each scale and all its preceding scales, thereby fusing complementary information between different scales for efficient reconstruction. Second, we design a spatialspectral attention module and deploy it in the decoder to boost reconstruction quality. By exploiting a cascade of spatialspectral attention, it can efficiently capture spatial and spectral correlations in hyperspectral images with a low volume of parameters. In addition, a compound loss, including the reconstruction loss and the spatial-spectral total variation loss, is designed to guide network learning in an end-to-end manner. Intensive experiments on simulation and real data show that our method has obvious advantages over multiple state-of-the-art methods, achieving a significant improvement in reconstruction quality and a substantial reduction in running time.

show abstract

“…Among them, HSIs can provide rich spectral information (i.e., about 10-nm spectral resolution), which makes it possible to identify the ground objects with different characteristics by means of the spectral information [6]. Based on the above advantage, the HSIs are widely employed in the field of hyperspectral image classification [7]- [8], hyperspectral unmixing [9], [10], hyperspectral pansharpening [11], [12], band selection [13], [14], hyperspectral anomaly detection (HAD) [15], [16] and hyperspectral target detection [17], [18], etc. The HAD, which aims to search for the pixels whose spectral signatures are deviated from the surrounding background pixels without priori knowledge about anomalies, has attracted extensive attention in the military and civilian fields.…”

Section: Introductionmentioning

confidence: 99%

Hyperspectral Anomaly Detection via Sparse Representation and Collaborative Representation

Lin

Zhang

Cheng

et al. 2023

IEEE J. Sel. Top. Appl. Earth Observations Remote Sensing

View full text Add to dashboard Cite

Sparse representation (SR)-based approaches and collaborative representation (CR)-based methods are proved to be effective to detect the anomalies in a hyperspectral image (HSI). Nevertheless, the existing methods for achieving hyperspectral anomaly detection (HAD) generally only consider one of them, failing to comprehensively exploit them to further promote the detection performance. To address the issue, a novel HAD method, which integrates both sparse representation and collaborative representation (SRCR), is proposed in this paper. To be specific, a SR model, whose overcomplete dictionary is generated by means of the density-based clustering algorithm and superpixel segmentation method, is firstly constructed for each pixel in an HSI. Then, for each pixel in an HSI, the used atoms in SR model are sifted to form the background dictionary corresponding to the CR model. To fully exploit both SR and CR information, we further combine the residual features obtained from both SR and CR model by the nonlinear transformation function to generate the response map. Finally, to preserve contour information of the objects, a postprocessing operation with guided filter is imposed into the response map to acquire the detection result. Experiments conducted on simulated and real data sets demonstrate that the proposed SRCR outperforms the state-of-the-art methods.

show abstract

Nonlocal Band Attention Network for Hyperspectral Image Band Selection

Cited by 21 publications

References 41 publications

A Dual Multi-Head Contextual Attention Network for Hyperspectral Image Classification

A Dual Multi-Head Contextual Attention Network for Hyperspectral Image Classification

Hyperspectral Snapshot Compressive Imaging With Dense Back-Projection Joint Attention Network

Hyperspectral Anomaly Detection via Sparse Representation and Collaborative Representation

Contact Info

Product

Resources

About