Hyperspectral Image Classification With Contrastive Self-Supervised Learning Under Limited Labeled Samples

Article Geochemistry & Geophysics

Heterogeneous Few-Shot Learning for Hyperspectral Image Classification

Yan Wang et al.

Summary: This study proposes a heterogeneous few-shot learning (HFSL) method for hyperspectral image (HSI) classification, which utilizes few-shot learning and knowledge transfer on different datasets to make full use of spectral and spatial information, improving classification accuracy.

IEEE GEOSCIENCE AND REMOTE SENSING LETTERS (2022)

Add to Collection

Article Geochemistry & Geophysics

Physically Constrained Transfer Learning Through Shared Abundance Space for Hyperspectral Image Classification

Ying Qu et al.

Summary: Hyperspectral image (HSI) classification is an active research topic, and recent advances in deep learning have greatly improved its performance. Transfer learning-based methods can help address the issue of domain discrepancies between training and testing images, and the proposed physically constrained transfer learning through shared abundance space (PCTL-SAS) method demonstrates superior results compared to state-of-the-art approaches. This new method has the potential to facilitate the deployment of HSI classification in real-world scenarios.

IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING (2021)

Add to Collection

Article Geochemistry & Geophysics

Compact Band Weighting Module Based on Attention-Driven for Hyperspectral Image Classification

Lin Zhao et al.

Summary: The novel CBW module presented in this article evaluates different spectral bands based on their contributions to a given classification task using an attention-driven mechanism. Compared to existing BW modules, the CBW is lightweight with only 20 parameters, reducing model complexity and time cost significantly. Experimental results demonstrate the superiority of the proposed algorithm over current state-of-the-art BW methods.

IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING (2021)

Add to Collection

Article Geochemistry & Geophysics

Hyperspectral Image Classification Based on 3-D Octave Convolution With Spatial-Spectral Attention Network

Xu Tang et al.

Summary: In this article, a 3-D octave convolution with spatial-spectral attention network (3DOC-SSAN) is proposed to improve the classification accuracy of hyperspectral images (HSIs). By integrating spatial and spectral information through an information complement model, important spatial and spectral features can be fully utilized for classification tasks. Comparing with existing classifiers, this method shows competitive performance on benchmark datasets.

IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING (2021)

Add to Collection

Article Geochemistry & Geophysics

Deep Multiview Learning for Hyperspectral Image Classification

Bing Liu et al.

Summary: This article proposes a deep multiview learning method to address the small sample problem in hyperspectral image classification. By constructing multiple views and designing a deep residual network, the method improves classification accuracy in an unsupervised learning setting.

IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING (2021)

Add to Collection

Article Engineering, Electrical & Electronic

Hyperspectral Image Classification Using a Hybrid 3D-2D Convolutional Neural Networks

Saeed Ghaderizadeh et al.

Summary: The study introduces the application of 3-D-CNN in hyperspectral image classification, discusses the current issues and solutions, and proposes a hybrid CNN method to address feature extraction and model complexity problems.

IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING (2021)

Add to Collection

Article Geochemistry & Geophysics