Recently, deep learning has achieved remarkable breakthroughs in hyperspectral image (HSI) classification tasks, particularly with methods based on convolutional neural networks (CNNs) and transformers. However, these methods have several limitations: 1) the limited receptive field inherent in the convolutional layer greatly hampers capturing feature contextual information on a large scale and 2) transformers cannot establish strong local relationships, making it challenging to characterize complex dependencies between distant pixels and different bands in HSIs. Moreover, as the network complexity increases, so does the number of network parameters. We propose a novel network called the spectral-spatial convolutional fusion long-range dependence transformer network (LRDTN) for HSI classification to address these challenges. LRDTN comprises three key components: dynamic-dependent convolutional (DDC) module, the multiscale enhanced fusion (MsEF) module, and the local–global perception transformer (LGPT). Specifically, the DDC dynamically models local features, while the MsEF integrates information from different scales to capture contextual relationships in HSI features effectively. Additionally, the ability to mine and utilize HSI local-global features and complex long-range dependencies is enhanced by the proposed transformer variant, LGPT. Ultimately, through the ingeniously designed structure of the LRDTN, the model effectively maintains its performance while reducing the number of network parameters. Extensive experiments conducted on four typical HSI datasets, including urban areas, agricultural areas, and swamps, demonstrate the superiority of LRDTN over other state-of-the-art networks. The code is available at https://github.com/ybyangjing/LRDTN .

Deep learning has been utilized for hyperspectral image (HSI) classification in recent years, with notable performance improvements. In particular, convolutional neural networks (CNNs) methods have achieved major advancements in this area. However, there are some drawbacks to the existing CNN-based HSI classification approaches: 1) the lack of effective and simple feature representations in CNNs, which overlook the effects of spectral differences and spatial contextual information; 2) the classification model has an enormous network complexity as a result of its numerous training parameters and high computational requirements; and 3) the category samples in HSI data exhibit a significant long tail distribution issue, which affects the classification performance of HSI. To address these issues, we propose a lightweight spectral-spatial neural architecture with multi-attention feature extraction (LSSMA) for HSI classification. The main work consists of three areas:1) a spectral feature extraction and fusing module (SFEF) is created to facilitate the fusing of spectral-spatial features while reducing the number of trainable parameters and computational complexity of the model. This module uses convolutions of various kernel sizes for residual connection and feature fusion, and introduces group convolution to achieve efficient feature representation; 2)to utilize the spectral-spatial correlation of HSI data to its fullest, a multi-scale convolutional activation guided attention mechanism (MsCA) is designed and the position attention module (PAM) is referenced, which can capture spectral differences and spatial contextual relationships between ground objects; and 3) focal loss (FL) is applied in computer vision tasks to the LSSMA model to enhance its capacity to handle category imbalance. Experimental results on four publicly available hyperspectral datasets show that the method obtains better classification performance at a lower computational cost.

Using convolutional neural networks (CNNs) in classifying hyperspectral images (HSIs) has achieved quite good results in recent years. It is widely used in agricultural remote sensing, geological exploration, environmental monitoring, and marine remote sensing. Unfortunately, the complexity of network structures used for hyperspectral image classification challenges the efficient delivery of HSI data extremely, and existing methods suffer from a large amount of redundancy in the network weight parameters during training, as they either require huge computational resources or make inefficient use of storage space when designing the network structure, and many of the parameters that waste computational resources contribute less to the rich spectral and spatial information transfer in HSI. So we introduce LCTCS, a better low-memory and less-parametric network approach. LCTCS aims to improve the efficiency of computational resource utilization with advanced classification performance and lower levels of computational resources. Unlike the conventional 2D and 3D convolution used previously, we use simple and efficient 3D grouped convolution as a vehicle to convey the semantic features of HSIs. More specifically, we design a novel two-channel sparse network to classify HSIs since grouped 3D convolution conveys the properties of hyperspectral data well in the time and space domains.We have compared LCTCS with eight widely used network methods on four publicly available hyperspectral datasets for learning HSI information. A series of experiments shows that the model architecture designed has $$65.89 \%$$65.89 % less storage space than the DBDA method, consumes $$67.36 \%$$67.36 % fewer computational resources than the SSRN method on the IP dataset, and accomplishes a highly accurate classification task with the number of parameters accounting for only $$1.99 \%$$1.99 % that of the DBMA method.