Triple-Attention-Based Parallel Network for Hyperspectral Image Classification

Abstract

Convolutional neural networks have been highly successful in hyperspectral image classification owing to their unique feature expression ability. However, the traditional data partitioning strategy in tandem with patch-wise classification may lead to information leakage and result in overoptimistic experimental insights. In this paper, we propose a novel data partitioning scheme and a triple-attention parallel network (TAP-Net) to enhance the performance of HSI classification without information leakage. The dataset partitioning strategy is simple yet effective to avoid overfitting, and allows fair comparison of various algorithms, particularly in the case of limited annotated data. In contrast to classical encoder–decoder models, the proposed TAP-Net utilizes parallel subnetworks with the same spatial resolution and repeatedly reuses high-level feature maps of preceding subnetworks to refine the segmentation map. In addition, a channel–spectral–spatial-attention module is proposed to optimize the information transmission between different subnetworks. Experiments were conducted on three benchmark hyperspectral datasets, and the results demonstrate that the proposed method outperforms state-of-the-art methods with the overall accuracy of 90.31%, 91.64%, and 81.35% and the average accuracy of 93.18%, 87.45%, and 78.85% over Salinas Valley, Pavia University and Indian Pines dataset, respectively. It illustrates that the proposed TAP-Net is able to effectively exploit the spatial–spectral information to ensure high performance.

Application of convolutional neural network in fusion and classification of multi-source remote sensing data

“…Multi-source remote sensing data includes hyperspectral data (HSI) and lidar data (LiDAR), due to their different types and applicable directions, there are certain challenges in fusion and classification (Qu et al, 2021 ). Therefore, the research uses CNN to extract its features, and proposes a dual-branch convolutional neural network (DB-CNN), which is convenient for organically combining multiple data sources.…”

Towards On-Board Hyperspectral Satellite Image Segmentation: Understanding Robustness of Deep Learning through Simulating Acquisition Conditions

“…The recent advancements in this field consist in proposing deeper architectures capable of extracting more informative features. This encompasses the use of densely connected CNNs [38], attention mechanisms [39], or multi-branch networks [40,41]. Additionally, many attempts are aimed at elaborating lightweight models [42] that are suitable for on-board processing.…”

Localization and recognition of specific targets in land-based hyperspectral images based on combined spatial and spectral information

“…Traditional hyperspectral image target detection technology is mostly based on spectral information, that is, the spectral reflectance information of object surface; Spatial target detection technology relies too much on shape information, and it is difficult to break through the twodimensional space limitation. Therefore, comprehensive utilization of spatial and spectral information to complete classification and detection tasks has become a research hotspot [9][10]. However, in practical applications, especially in the aspect of military target precision attack, it is necessary to complete accurate target iden-tification while target positioning, which must rely on the joint information of space and spectrum.…”