Spectral-Spatial Attention Rotation-Invariant Classification Network for Airborne Hyperspectral Images

Abstract

An airborne hyperspectral imaging system is typically equipped on an aircraft or unmanned aerial vehicle (UAV) to capture ground scenes from an overlooking perspective. Due to the rotation of the aircraft or UAV, the same region of land cover may be imaged from different viewing angles. While humans can accurately recognize the same objects from different viewing angles, classification methods based on spectral-spatial features for airborne hyperspectral images exhibit significant errors. The existing methods primarily involve incorporating image or feature rotation angles into the network to improve its accuracy in classifying rotated images. However, these methods introduce additional parameters that need to be manually determined, which may not be optimal for all applications. This paper presents a spectral-spatial attention rotation-invariant classification network for the airborne hyperspectral image to address this issue. The proposed method does not require the introduction of additional rotation angle parameters. There are three modules in the proposed framework: the band selection module, the local spatial feature enhancement module, and the lightweight feature enhancement module. The band selection module suppresses redundant spectral channels, while the local spatial feature enhancement module generates a multi-angle parallel feature encoding network to improve the discrimination of the center pixel. The multi-angle parallel feature encoding network also learns the position relationship between each pixel, thus maintaining rotation invariance. The lightweight feature enhancement module is the last layer of the framework, which enhances important features and suppresses insignificance features. At the same time, a dynamically weighted cross-entropy loss is utilized as the loss function. This loss function adjusts the model’s sensitivity for samples with different categories according to the output in the training epoch. The proposed method is evaluated on five airborne hyperspectral image datasets covering urban and agricultural regions. Compared with other state-of-the-art classification algorithms, the method achieves the best classification accuracy and is capable of effectively extracting rotation-invariant features for urban and rural areas.

Generalized Category Discovery in Aerial Image Classification via Slot Attention

“…While these techniques have demonstrated effectiveness in aerial image classification, they are limited in their capacity to encapsulate global semantic information, owing to the reliance on handcrafted local features for representing images. With the recent advancements in convolutional neural networks (CNNs), a plethora of methodologies have emerged within the domain of aerial image classification [1][2][3][4][5]. CNNs exhibit commendable prowess in capturing both global and local representations of complex aerial images without necessitating additional manual intervention, thus significantly augmenting the efficacy of aerial image classification.…”

“…Deep neural networks have demonstrated excellent performance in aerial image classification [1][2][3][4][5]. However, existing methodologies rely on the closed-set assumption, meaning that the image categories in the test set must be a subset of the categories in the training set.…”

Prior-based collaborative representation with global adaptive weight for hyperspectral anomaly detection

“…1 Compared with visible and multispectral imagery, hyperspectral images (HSIs) contain hundreds of continuous spectral bands that provide spectral-spatial information. On this basis, HSI interpretation techniques have been widely applied in many fields, such as classification, [2][3][4][5] change detection, [6][7][8][9] and object detection. [10][11][12] As an important subbranch of hyperspectral target location, hyperspectral anomaly detection (HAD) aims to distinguish objects of interest by capturing subtle spectral differences of ground objects without any prior information or supervision.…”

Multi-Prior Graph Autoencoder with Ranking-Based Band Selection for Hyperspectral Anomaly Detection

“…As opposed to visible and multispectral images, hyperspectral images (HSIs) have been regarded as a remarkable invention in the field of remote sensing imaging sciences, due to their practical capacity to capture high-dimensional spectral information from different scenes on the Earth's surface [1]. HSIs consist of innumerable contiguous spectral bands that span the electromagnetic spectrum, providing rich and detailed physical attributes of land covers, which facilitate the development of various applications such as change detection [2][3][4][5], land-cover classification [6][7][8], retrieval [9], scene classification [10,11] and anomaly detection [12,13].…”