Abstract
The large-scale labeled data is very crucial to train a classification model with strong generalization ability. However, the collection of large-scale labeled data is very expensive, especially in the remote sensing fields. The available of labeled data is very limited for the hyperspectal image classification. To address such a challenge, active learning and semi-supervised learning are two popular techniques in machine learning community. In this paper, we integrate active learning and semi-supervised learning into a framework by improving the quality of pseudo-labels for hyperspectral remote sensing images. In the proposed method, the collaboration of the spatial features and spectral features are adopted to improve the ability of classifier. Specifically, we train two classifiers with spatial feature and spectral feature respectively based on the labeled data. Then the prediction probabilities of the two classifiers are combined for strong prediction. With active learning technique, we can select a batch of the most informative samples and obtain a new labeled dataset. Two classifiers based on the new labeled dataset can be obtained. With these two classifiers, another prediction results by combining their predictions can be obtained. To guarantee the quality of the pseudo-labels, the samples that are predicted with the same labels before and after active learning are assigned with pseudo-labels. The samples that can not be assigned with high confident samples are regarded as the candidate pool for active learning. The final predictions are obtained by the classification models trained on the pseudo-labeled samples and the labeled samples with both the spatial features and spectral features. The experiments on two popular hyperspectral images show that the proposed method outperforms the state-of-the-art and baseline methods.
Similar content being viewed by others
References
Benediktsson J, Palmason J, Sveinsson J (2005) Classification of hyperspectral data from urban areas based on extended morphological profiles. IEEE Trans Geosci Remote Sens 43(3):480–491
Chen L, Shang S, Yang C, Li J (2020) Spatial keyword search: a survey. GeoInformatica 24(1):85–106
Cui B, Cui J, Lu Y, Guo N, Gong M (2020) A sparse representation-based sample pseudo-labeling method for hyperspectral image classification. Remote Sens 12(4):664
Demir B, Persello C, Bruzzone L (2011) Batch-mode active-learning methods for the interactive classification of remote sensing images. IEEE Trans Geosci Remote Sens 49(3):1014–1031
Fang B, Li Y, Zhang H, Chan JCW (2020) Collaborative learning of lightweight convolutional neural network and deep clustering for hyperspectral image semi-supervised classification with limited training samples. ISPRS J Photogramm Remote Sens 161:164–178
Fauvel M, Chanussot J, Benediktsson JA, Sveinsson JR (2007) Spectral and spatial classification of hyperspectral data using svms and morphological profiles. In: 2007 IEEE International geoscience and remote sensing symposium, pp 4834–4837
Gao M, Zhang Z, Yu G, Arık SÖ, Davis LS, Pfister T (2020) Consistency-based semi-supervised active learning: Towards minimizing labeling cost. In: Vedaldi A, Bischof H, Brox T, Frahm JM (eds) Computer Vision – ECCV 2020, pp 510–526
Han P, Shang S, Sun A, Zhao P, Zheng K, Kalnis P (2019) Auc-mf: point of interest recommendation with auc maximization. In: 2019 IEEE 35th international conference on data engineering (ICDE), pp 1558–1561
He Z, Xia K, Li T, Zu B, Yin Z, Zhang J (2021) A constrained graph-based semi-supervised algorithm combined with particle cooperation and competition for hyperspectral image classification. Remote Sens 13 (2):193
Hong D, Wu X, Ghamisi P, Chanussot J, Yokoya N, Zhu XX (2020) Invariant attribute profiles: A spatial-frequency joint feature extractor for hyperspectral image classification. IEEE Trans Geosci Remote Sens 58(6):3791–3808
Hu J, Hong D, Zhu XX (2019) Mima: Mapper-induced manifold alignment for semi-supervised fusion of optical image and polarimetric sar data. IEEE Trans Geosci Remote Sens 57(11):9025–9040
Huo L, Tang P (2014) A batch-mode active learning algorithm using region-partitioning diversity for svm classifier. IEEE J Sel Top Appl Earth Obs Remote Sens 7(4):1036–1046
Imani M, Ghassemian H (2020) An overview on spectral and spatial information fusion for hyperspectral image classification: Current trends and challenges. Information Fusion 59:59–83
Jamshidpour N, Safari A, Homayouni S (2020) A ga-based multi-view, multi-learner active learning framework for hyperspectral image classification. Remote Sens 12(2):297
Liu C, Li J, He L (2018) Superpixel-based semisupervised active learning for hyperspectral image classification. IEEE J Sel Top Appl Earth Obs Remote Sens 12(1):357–370
Lv J, Zhang H, Yang M, Yang W (2020) A novel spectral–spatial based adaptive minimum spanning forest for hyperspectral image classification. GeoInformatica 24:827–848
Ma KY, Chang CI (2021) Iterative training sampling coupled with active learning for semisupervised spectral-spatial hyperspectral image classification. IEEE Trans Geoscience Remote Sens
Patra S, Bruzzone L (2014) A novel som-svm-based active learning technique for remote sensing image classification. IEEE Trans Geosci Remote Sens 52 (11):6899–6910
Platt J, et al. (1999) Probabilistic outputs for support vector machines and comparisons to regularized likelihood methods. Advances in Large Margin Classifiers 10(3):61–74
Shang S, Chen L, Wei Z, Jensen CS, Zheng K, Kalnis P (2018) Parallel trajectory similarity joins in spatial networks. The VLDB Journal 27(3):395–420
Shang S, Shen J, Wen JR, Kalnis P. (2020) Deep understanding of big geospatial data for self-driving cars
Tuia D, Ratle F, Pacifici F, Kanevski MF, Emery WJ (2009) Active learning methods for remote sensing image classification. IEEE Trans Geosci Remote Sens 47(7):2218–2232
Tuia D, Volpi M, Copa L, Kanevski M, Munoz-Mari J (2011) A survey of active learning algorithms for supervised remote sensing image classification. IEEE J Sel Top Signal Process 5(3):606–617
Tuia D, Volpi M, Copa L, Kanevski M, Munoz-Mari J (2011) A survey of active learning algorithms for supervised remote sensing image classification. IEEE J Sel Top Signal Process 5(3):606–617
Van Engelen JE, Hoos HH (2020) A survey on semi-supervised learning. Mach Learn 109(2):373–440
Wan L, Tang K, Li M, Zhong Y, Qin AK (2015) Collaborative active and semisupervised learning for hyperspectral remote sensing image classification. IEEE Trans Geosci Remote Sens 53(5):2384–2396
Wang Z, Du B, Guo Y (2020) Domain adaptation with neural embedding matching. IEEE Trans Neural Netw Learn Sys 31(7):2387–2397
Wang Z, Du B, Shi Q, Tu W (2019) Domain adaptation with discriminative distribution and manifold embedding for hyperspectral image classification. IEEE Geosci Remote Sens Lett 16(7):1155–1159
Wang Z, Du B, Zhang L, Zhang L, Jia X (2017) A novel semisupervised active-learning algorithm for hyperspectral image classification. IEEE Trans Geosci Remote Sens 55(6):3071–3083
Wu H, Prasad S (2017) Semi-supervised deep learning using pseudo labels for hyperspectral image classification. IEEE Trans Image Process 27 (3):1259–1270
Zhang Y, Cao G, Li X, Wang B, Fu P (2019) Active semi-supervised random forest for hyperspectral image classification. Remote Sensing 11(24)
Zhao Y, Shang S, Wang Y, Zheng B, Nguyen QVH, Zheng K (2018) Rest: A reference-based framework for spatio-temporal trajectory compression. In: Proceedings of the 24th ACM SIGKDD International conference on knowledge discovery & data mining, pp 2797–2806
Zhou X, Prasad S (2017) Active and semisupervised learning with morphological component analysis for hyperspectral image classification. IEEE Geosci Remote Sens Lett 14(8):1348–1352
Zhou X, Prasad S (2017) Active and semisupervised learning with morphological component analysis for hyperspectral image classification. IEEE Geosci Remote Sens Lett 14(8):1348–1352
Zhu R, Dornaika F, Ruichek Y (2020) Semi-supervised elastic manifold embedding with deep learning architecture. Pattern Recogn 107:107425
Acknowledgements
This work was supported in part by the National Natural Science Foundation of China under Grants 62006176, the Science and Technology Major Project of Hubei Province (Next- Generation AI Technologies) under Grant 2019AEA170, the Natural Science Foundation of Hubei Province under Grants 2020CFB241 and by the National Post-Doctoral Program for Innovative Talents under Grant BX20190250, Projected funded by China Postdoctoral Science Foundation 2020M672413, and the National Dam Safety Research Center Program (CX2020B06).
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Wang, Z., Du, B. Unified active and semi-supervised learning for hyperspectral image classification. Geoinformatica 27, 23–38 (2023). https://doi.org/10.1007/s10707-021-00443-0
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10707-021-00443-0