Advances in Hyperspectral Image Classification Methods with Small Samples: A Review

Hyperspectral image (HSI) classification is one of the hotspots in remote sensing, and many methods have been continuously proposed in recent years. However, it is still challenging to achieve high accuracy classification in applications. One of the main reasons is the lack of labeled data. Due to the limitation of spatial resolution, manual labeling of HSI data is time-consuming and costly, so it is difficult to obtain a large amount of labeled data. In such a situation, many researchers turn their attention to the study of HSI classification with small samples. Focusing on this topic, this paper provides a systematic review of the research progress in recent years. Specifically, this paper contains three aspects. First, considering that the taxonomy used in previous review articles is not well-developed and confuses the reader, we propose a novel taxonomy based on the form of data utilization. This taxonomy provides a more accurate and comprehensive framework for categorizing the various approaches. Then, using the proposed taxonomy as a guideline, we analyze and summarize the existing methods, especially the latest research results (both deep and non-deep models) that were not included in the previous reviews, so that readers can understand the latest progress more clearly. Finally, we conduct several sets of experiments and present our opinions on current problems and future directions.

[1]  B. Berthelot,et al.  Active Learning for Hyperspectral Image Classification: A comparative review , 2022, IEEE Geoscience and Remote Sensing Magazine.

[2]  Yanning Zhang,et al.  Integrating Hybrid Pyramid Feature Fusion and Coordinate Attention for Effective Small Sample Hyperspectral Image Classification , 2022, Remote. Sens..

[3]  Liguo Wang,et al.  One-Shot Dense Network with Polarized Attention for Hyperspectral Image Classification , 2022, Remote. Sens..

[4]  Deyun Zhou,et al.  Graph-Based Deep Multitask Few-Shot Learning for Hyperspectral Image Classification , 2022, Remote. Sens..

[5]  Bing Liu,et al.  Unsupervised Meta Learning With Multiview Constraints for Hyperspectral Image Small Sample set Classification , 2022, IEEE Transactions on Image Processing.

[6]  Yiting Wang,et al.  Minimalistic fully convolution networks (MFCN): pixel-level classification for hyperspectral image with few labeled samples. , 2022, Optics express.

[7]  Hongtao Shi,et al.  An Investigation of a Multidimensional CNN Combined with an Attention Mechanism Model to Resolve Small-Sample Problems in Hyperspectral Image Classification , 2022, Remote. Sens..

[8]  Bing Liu,et al.  Small Sample Hyperspectral Image Classification Based on Cascade Fusion of Mixed Spatial-Spectral Features and Second-Order Pooling , 2022, Remote. Sens..

[9]  Bing Liu,et al.  Multiscale nested U-Net for small sample classification of hyperspectral images , 2022, Journal of Applied Remote Sensing.

[10]  Mingqiang Wei,et al.  Hyperspectral image classification on insufficient-sample and feature learning using deep neural networks: A review , 2021, Int. J. Appl. Earth Obs. Geoinformation.

[11]  Xueying Li,et al.  An overview of hyperspectral image feature extraction, classification methods and the methods based on small samples , 2021, Applied Spectroscopy Reviews.

[12]  Cong Bai,et al.  3D Octave and 2D Vanilla Mixed Convolutional Neural Network for Hyperspectral Image Classification with Limited Samples , 2021, Remote. Sens..

[13]  Aili Wang,et al.  Depthwise Separable Relation Network for Small Sample Hyperspectral Image Classification , 2021, Symmetry.

[14]  Wei Wei,et al.  Boosting Few-Shot Hyperspectral Image Classification Using Pseudo-Label Learning , 2021, Remote. Sens..

[15]  Bing Liu,et al.  Deep Multiview Learning for Hyperspectral Image Classification , 2021, IEEE Transactions on Geoscience and Remote Sensing.

[16]  F. Dell’acqua,et al.  A Review of Unsupervised Band Selection Techniques: Land Cover Classification for Hyperspectral Earth Observation Data , 2021, IEEE Geoscience and Remote Sensing Magazine.

[17]  Zenglin Xu,et al.  Rectifying the Shortcut Learning of Background for Few-Shot Learning , 2021, NeurIPS.

[18]  Ali Can Karaca.  Domain Transform Filter and Spatial-Aware Collaborative Representation for Hyperspectral Image Classification Using Few Labeled Samples , 2021, IEEE Geoscience and Remote Sensing Letters.

[19]  Mengdao Xing,et al.  Semi-supervised rotation forest based on ensemble margin theory for the classification of hyperspectral image with limited training data , 2021, Inf. Sci..

[20]  Biplab Banerjee,et al.  SPN: Stable Prototypical Network for Few-Shot Learning-Based Hyperspectral Image Classification , 2021, IEEE Geoscience and Remote Sensing Letters.

[21]  Bing Liu,et al.  Small sample classification of hyperspectral image using model-agnostic meta-learning algorithm and convolutional neural network , 2021, International Journal of Remote Sensing.

[22]  Zhijie Lin,et al.  A survey: Deep learning for hyperspectral image classification with few labeled samples , 2021, Neurocomputing.

[23]  Qian Du,et al.  Deep Cross-Domain Few-Shot Learning for Hyperspectral Image Classification , 2021, IEEE Transactions on Geoscience and Remote Sensing.

[24]  Ali Can Karaca.  Spatial aware probabilistic multi-kernel collaborative representation for hyperspectral image classification using few labelled samples , 2021 .

[25]  Jing Bai,et al.  Class Incremental Learning With Few-Shots Based on Linear Programming for Hyperspectral Image Classification , 2020, IEEE Transactions on Cybernetics.

[26]  Xiaorun Li,et al.  3D convolutional siamese network for few-shot hyperspectral classification , 2020, Journal of Applied Remote Sensing.

[27]  Ankush Gupta,et al.  CrossTransformers: spatially-aware few-shot transfer , 2020, NeurIPS.

[28]  Yushi Chen,et al.  Dual-Path Siamese CNN for Hyperspectral Image Classification With Limited Training Samples , 2020, IEEE Geoscience and Remote Sensing Letters.

[29]  Pengqiang Zhang,et al.  Deep Relation Network for Hyperspectral Image Few-Shot Classification , 2020, Remote. Sens..

[30]  Jonathan Cheung-Wai Chan,et al.  Collaborative learning of lightweight convolutional neural network and deep clustering for hyperspectral image semi-supervised classification with limited training samples , 2020 .

[31]  Jing Zhao,et al.  Hyperspectral imagery classification with cascaded support vector machines and multi-scale superpixel segmentation , 2020, International Journal of Remote Sensing.

[32]  Qiming Qin,et al.  Deep Quadruplet Network for Hyperspectral Image Classification with a Small Number of Samples , 2020, Remote. Sens..

[33]  Mercedes Eugenia Paoletti,et al.  Deep learning classifiers for hyperspectral imaging: A review , 2019 .

[34]  Jing Zhao,et al.  Spectral–Spatial Hyperspectral Image Classification Using Cascaded Markov Random Fields , 2019, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.

[35]  Jing Cui,et al.  Hyperspectral Image Classification With Small Training Sample Size Using Superpixel-Guided Training Sample Enlargement , 2019, IEEE Transactions on Geoscience and Remote Sensing.

[36]  Nicolas Audebert,et al.  Deep Learning for Classification of Hyperspectral Data: A Comparative Review , 2019, IEEE Geoscience and Remote Sensing Magazine.

[37]  James T. Kwok,et al.  Generalizing from a Few Examples , 2019, ACM Comput. Surv..

[38]  Gang Wan,et al.  Deep Few-Shot Learning for Hyperspectral Image Classification , 2019, IEEE Transactions on Geoscience and Remote Sensing.

[39]  Yu Li,et al.  Hyperspectral images classification with convolutional neural network and textural feature using limited training samples , 2019, Remote Sensing Letters.

[40]  J. Benediktsson,et al.  New Frontiers in Spectral-Spatial Hyperspectral Image Classification: The Latest Advances Based on Mathematical Morphology, Markov Random Fields, Segmentation, Sparse Representation, and Deep Learning , 2018, IEEE Geoscience and Remote Sensing Magazine.

[41]  Bing Liu,et al.  Deep convolutional recurrent neural network with transfer learning for hyperspectral image classification , 2018, Journal of Applied Remote Sensing.

[42]  Naoto Yokoya,et al.  Advances in Hyperspectral Image and Signal Processing: A Comprehensive Overview of the State of the Art , 2017, IEEE Geoscience and Remote Sensing Magazine.

[43]  Xiao Xiang Zhu,et al.  Deep Learning in Remote Sensing: A Comprehensive Review and List of Resources , 2017, IEEE Geoscience and Remote Sensing Magazine.

[44]  Zhenwei Shi,et al.  MugNet: Deep learning for hyperspectral image classification using limited samples , 2017, ISPRS Journal of Photogrammetry and Remote Sensing.

[45]  Jonathan Cheung-Wai Chan,et al.  Learning and Transferring Deep Joint Spectral–Spatial Features for Hyperspectral Classification , 2017, IEEE Transactions on Geoscience and Remote Sensing.

[46]  Binge Cui,et al.  A novel hyperspectral image classification approach based on multiresolution segmentation with a few labeled samples , 2017 .

[47]  Richard S. Zemel,et al.  Prototypical Networks for Few-shot Learning , 2017, NIPS.

[48]  Xiguang Yang,et al.  Estimating Soil Salinity Under Various Moisture Conditions: An Experimental Study , 2017, IEEE Transactions on Geoscience and Remote Sensing.

[49]  Sen Jia,et al.  Convolutional neural networks for hyperspectral image classification , 2017, Neurocomputing.

[50]  Przemysław Głomb,et al.  Semi-supervised hyperspectral classification from a small number of training samples using a co-training approach , 2016 .

[51]  Zhaohui Xue,et al.  Kernel Supervised Ensemble Classifier for the Classification of Hyperspectral Data Using Few Labeled Samples , 2016, Remote. Sens..

[52]  Marcin Andrychowicz,et al.  Learning to learn by gradient descent by gradient descent , 2016, NIPS.

[53]  Oriol Vinyals,et al.  Matching Networks for One Shot Learning , 2016, NIPS.

[54]  Peijun Du,et al.  Rotation-Based Support Vector Machine Ensemble in Classification of Hyperspectral Data With Limited Training Samples , 2016, IEEE Transactions on Geoscience and Remote Sensing.

[55]  Naoto Yokoya,et al.  Potential of Resolution-Enhanced Hyperspectral Data for Mineral Mapping Using Simulated EnMAP and Sentinel-2 Images , 2016, Remote. Sens..

[56]  Zhihao Qin,et al.  Estimation of Crop LAI using hyperspectral vegetation indices and a hybrid inversion method , 2015 .

[57]  David A. Clausi,et al.  Hyperspectral Image Classification With Limited Labeled Training Samples Using Enhanced Ensemble Learning and Conditional Random Fields , 2015, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.

[58]  Jocelyn Chanussot,et al.  A Physics-Based Unmixing Method to Estimate Subpixel Temperatures on Mixed Pixels , 2015, IEEE Transactions on Geoscience and Remote Sensing.

[59]  Maryam Imani,et al.  Band Clustering-Based Feature Extraction for Classification of Hyperspectral Images Using Limited Training Samples , 2014, IEEE Geoscience and Remote Sensing Letters.

[60]  Maryam Imani,et al.  Feature Extraction Using Attraction Points for Classification of Hyperspectral Images in a Small Sample Size Situation , 2014, IEEE Geoscience and Remote Sensing Letters.

[61]  Qingquan Li,et al.  A Two-Stage Feature Selection Framework for Hyperspectral Image Classification Using Few Labeled Samples , 2014, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.

[62]  J. Chanussot,et al.  Hyperspectral Remote Sensing Data Analysis and Future Challenges , 2013, IEEE Geoscience and Remote Sensing Magazine.

[63]  Fuchun Sun,et al.  A Fast and Robust Sparse Approach for Hyperspectral Data Classification Using a Few Labeled Samples , 2012, IEEE Transactions on Geoscience and Remote Sensing.

[64]  Chein-I Chang,et al.  Hyperspectral Data Exploitation , 2007 .

[65]  Lin Zhao,et al.  Hyperspectral Image Classification With Contrastive Self-Supervised Learning Under Limited Labeled Samples , 2022, IEEE Geoscience and Remote Sensing Letters.

[66]  Puhong Duan,et al.  Polygon Structure-Guided Hyperspectral Image Classification With Single Sample for Strong Geometric Characteristics Scenes , 2022, IEEE Transactions on Geoscience and Remote Sensing.

[67]  Peijun Du,et al.  S3Net: Spectral-Spatial Siamese Network for Few-Shot Hyperspectral Image Classification , 2022, IEEE Transactions on Geoscience and Remote Sensing.

[68]  Z. J. Wang,et al.  Cross-Domain Few-Shot Contrastive Learning for Hyperspectral Images Classification , 2022, IEEE Geoscience and Remote Sensing Letters.

[69]  Licheng Jiao,et al.  Nonoverlapped Sampling for Hyperspectral Imagery: Performance Evaluation and a Cotraining-Based Classification Strategy , 2022, IEEE Transactions on Geoscience and Remote Sensing.

[70]  Changhong Fu,et al.  DeconNet: End-to-End Decontaminated Network for Vision-Based Aerial Tracking , 2022, IEEE Transactions on Geoscience and Remote Sensing.

[71]  Licheng Jiao,et al.  Few-Shot Hyperspectral Image Classification Based on Adaptive Subspaces and Feature Transformation , 2022, IEEE Transactions on Geoscience and Remote Sensing.

[72]  Mengdao Xing,et al.  A Pixel Cluster CNN and Spectral-Spatial Fusion Algorithm for Hyperspectral Image Classification With Small-Size Training Samples , 2021, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.

[73]  Junping Zhang,et al.  Attention Multisource Fusion-Based Deep Few-Shot Learning for Hyperspectral Image Classification , 2021, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.

[74]  Xuchu Yu,et al.  Patch-Free Bilateral Network for Hyperspectral Image Classification Using Limited Samples , 2021, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.

[75]  Qiming Qin,et al.  Global Prototypical Network for Few-Shot Hyperspectral Image Classification , 2020, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.

[76]  Wei Wei,et al.  Intraclass Similarity Structure Representation-Based Hyperspectral Imagery Classification With Few Samples , 2020, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.

[77]  Bing Liu,et al.  Deep Induction Network for Small Samples Classification of Hyperspectral Images , 2020, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.