A fast dynamic graph convolutional network and CNN parallel network for hyperspectral image classification
원문 초록
Deep learning has achieved impressive results on hyperspectral image (HSI) classification. Among them, both convolutional neural networks (CNNs) and graph neural networks (GNNs) have great potential for HSI classification. Supervised CNNs can efficiently extract hierarchical spatial–spectral features of HSIs, but these methods face the problem of high time complexity as the number of network layers increases. Semisupervised GNNs can rapidly capture the structural information of HSIs, while they cannot be well extended to HSI applications because of the process of adjacency matrix consuming large amount of memory resources. In this article, we propose a fast dynamic graph convolutional network and CNN (FDGC) parallel network for HSI classification. We first obtain two classification features by flattening and pooling operations on the results of the convolution layers, which fully exploits the spatial–spectral information contained in the hyperspectral data cube. Then, a dynamic graph convolution module is applied to extract the intrinsic structural information of each patch. Finally, we can obtain the HSI classification results based on these spatial, spectral, and structural features. By using three branches, FDGC can parallelly process multiple features of HSI in a supervised learning manner. In addition, regularization techniques, such as DropBlock and label smoothing, are applied to further improve the generalization capability of the model. Experimental results on three datasets show that our proposed algorithm is comparable with the state-of-the-art supervised learning models in terms of accuracy while also significantly outperforming in terms of training and inference time.
함께 살펴볼 논문
공통 태그를 근거로 연결했습니다. 실제 인용 관계나 논문의 우수성을 뜻하지 않습니다.
MVNet: Hyperspectral Remote Sensing Image Classification Based on Hybrid Mamba-Transformer Vision Backbone Architecture
Hyperspectral Image Classification via Transformer-based Spectral-Spatial Attention Decoupling and Adaptive Gating
Spectral-Spatial Self-Supervised Learning for Few-Shot Hyperspectral Image Classification
Dynamic 3D KAN Convolution with Adaptive Grid Optimization for Hyperspectral Image Classification
Boosting Adversarial Transferability for Hyperspectral Image Classification Using 3D Structure-invariant Transformation and Intermediate Feature Distance
Parameter-Efficient Fine-Tuning of Multispectral Foundation Models for Hyperspectral Image Classification
토론
이 논문과 연결된 포럼 글이 아직 없습니다.