ABSTRACT
With the increasing number of hyperspectral remote sensing images, it is of great research significance to utilize labeled source domain images to achieve accurate classification of unlabeled target domain images quickly. However, due to variations in atmospheric conditions, lighting conditions, temporal factors, and spatial morphologies of materials within the same object category, the spectral shift frequently occurs, leading to significant data distribution differences between the source and target domains. This phenomenon hinders the adaptability of the classification model trained on source domain data to the target domain data, thereby seriously affecting the effectiveness of domain adaptation methods. To address the issue of significant distribution differences and lack of labels between domains, this paper proposes a hyperspectral image classification method based on attention subdomain adaptation(ASDA). Firstly, a spatial-spectral feature extractor is used to extract features, aiming to make full use of the information in HSI. Then, a attention-based conditional distribution alignment method is proposed to align the features of interest, which can alleviate the negative impact of redundant pixels and redundant bands on subdomain alignment. Specifically, spatial attention is used to remove the influence of spatial interference pixels, while channel attention is employed to eliminate redundant bands in the spectrum. Finally, the subdomains are aligned using the Local Maximum Mean Discrepancy (LMMD) criterion, which focuses on aligning more effective features of the attention network. This alignment enables accurate classification of the target domain data. The effectiveness of this method is validated on two datasets.
- Ghamisi P, 2017. Advances in hyperspectral image and signal processing: A comprehensive overview of the state of the art. IEEE Geosci. Remote Sens. Mag, vol. 5, no. 4, 37–78 pages, Dec 2017.Google ScholarCross Ref
- Kouw W. M and Loog M.2021. A review of domain adaptation without target labels. IEEE Trans. Pattern Anal. Mach. Intell, vol. 43, no. 3, pp. 766–785, Mar. 2021, https:// doi: 10.1109/TPAMI.2019.2945942.Google ScholarCross Ref
- Farahani A, Voghoei S, Rasheed K and Arabnia H. R.2021. A brief review of domain adaptation. Adv. Data Sci. Inf. Eng, 877–894 pages.Google Scholar
- Peng, J, Huang, Y, Sun, W, Chen, N, Ning, Y, & Du, Q. 2022. Domain Adaptation in Remote Sensing Image Classification: A Survey. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 15, 9842–9859. https://doi.org/10.1109/jstars.2022.3220875.Google ScholarCross Ref
- Yao Y and Doretto G. 2010. Boosting for transfer learning with multiple sources. in Proc. IEEE Comput. Soc. Conf. Comput. Vis. Pattern Recognit., Jun. 2010, 1855–1862 pages, https://doi: 10.1109/CVPR.2010.5539857.Google ScholarCross Ref
- Li H, Li J, Zhao Y, Gong M, Zhang Y and Liu T. 2021. Cost-sensitive self-paced learning with adaptive regularization for classification of image time series. IEEE J. Sel. Topics Appl. Earth Observ. Remote Sens, vol. 14, 11713–11727 pages .Google ScholarCross Ref
- Zhong S and Zhang Y. 2021. An iterative training sample updating approach for domain adaptation in hyperspectral image classification. IEEE Geosci. Remote Sens. Lett., vol. 18, no. 10, pp. 1821–1825, Oct. 2021.Google ScholarCross Ref
- Wang Y, Chen Y, Hao S, Feng W and Shen Z. 2017. Balanced distribution adaptation for transfer learning. in Proc. IEEE Int. Conf. Data Mining (ICDM), Nov. 2017, pp. 1129–1134, https://doi: 10.1109/ICDM.2017.150.Google ScholarCross Ref
- Zhang Y, Li W, Tao R, Peng J, Du Q.and Cai Z. 2021. Cross-scene hyperspectral image classification with discriminative cooperative alignment. IEEE Trans. Geosci. Remote Sens, vol. 59, no. 1,9646–9660 pages, Nov. 2021, https://doi: 10.1109/TGRS.2020.3046756.Google ScholarCross Ref
- Wang Z, Du B, Shi Q and Tu W. 2019. Domain adaptation with discriminative distribution and manifold embedding for hyperspectral image classification. IEEE Geosci. Remote Sens. Lett., vol. 16, no. 7, 1155–1159 pages, Jul. 2019, https://doi: 10.1109/LGRS.2018.2889967.Google ScholarCross Ref
- Nguyen Minh Trieu and Nguyen Truong Thinh, "A Study of Combining KNN and ANN for Classifying Dragon Fruits Automatically," Journal of Image and Graphics, Vol. 10, No. 1, pp. 28-35, March 2022.Google Scholar
- Sarunya Kanjanawattana*, Worawit Teerawatthanaprapha, Panchalee Praneetpholkrang, Gun Bhakdisongkhram, and Suchada Weeragulpiriya, "Pineapple Sweetness Classification Using Deep Learning Based on Pineapple Images," Journal of Image and Graphics, Vol. 11, No. 1, pp. 47-52, March 2023.Google Scholar
- Wambugu N, 2021. Hyperspectral image classification on insufficientsample and feature learning using deep neural networks: A review. Int. J. Appl. Earth Observ. Geoinformation, vol. 105, p. 102603.Google ScholarCross Ref
- Long M, Cao Y, Wang J and Jordan M. 2015. Learning transferable features with deep adaptation networks. in Proc. Int. Conf. Mach. Learn, 97–105 pages.Google Scholar
- Ma X, Mou X, Wang J, Liu X, Wang H and Yin B. 2019. Cross-data set hyperspectral image classification based on deep domain adaptation. IEEE Trans. Geosci. Remote Sens., vol. 57, no. 12, 10164–10174 pages , Dec. 2019.Google ScholarCross Ref
- Zhu, Y., Zhuang, F., Wang, J., Ke, G., Chen, J., Bian, J., Xiong, H., & He, Q. (2021). Deep Subdomain Adaptation Network for Image Classification.IEEE Transactions on Neural Networks and Learning Systems, 32(4), 1713–1722. https://doi.org/10.1109/tnnls.2020.2988928Google ScholarCross Ref
- Ganin Y, Ustinova E, Ajakan H, 2016. Domain-adversarial training of neural networks. J. The journal of machine learning research, 2016, 17(1): 2096-30.Google Scholar
- Liu Z, Ma L, Du Q. 2020. Class-wise distribution adaptation for unsupervised classification of hyperspectral remote sensing images. J. IEEE Transactions on Geoscience and Remote Sensing, 2020, 59(1): 508-21.Google Scholar
- Deng B, Jia S, Shi D. 2019. Deep metric learning-based feature embedding for hyperspectral image classification. J. IEEE Transactions on Geoscience and Remote Sensing, 2019, 58(2): 1422-35.Google Scholar
- Li S, Lv F, Xie B, 2021, Bi-classifier determinacy maximization for unsupervised domain adaptation. C. Proceedings of the American Association for Artificial Intelligence.Google ScholarCross Ref
- Xu Y, Li Z, Li W, 2021. Dual-channel residual network for hyperspectral image classification with noisy labels. J. IEEE Transactions on Geoscience and Remote Sensing, 2021, 60: 1-11.Google Scholar
- He K, Zhang X, Ren S, 2016. Deep residual learning for image recognition. C. Proceedings of the IEEE conference on computer vision and pattern recognition.Google ScholarCross Ref
- Woo S, Park J, Lee J-Y, 2018. Cbam: Convolutional block attention module. C. Proceedings of the European conference on computer vision (ECCV).Google ScholarDigital Library
Index Terms
- Attention-Based Subdomain Adaptation Network for Hyperspectral Image Classification
Recommendations
Active multi-kernel domain adaptation for hyperspectral image classification
We propose a multi-kernel domain adaption framework for HIS classification.The multiple kernels can help capture the connection among domains.The method can adaptively migrate the information across domains.We design a nested actively updating process ...
A Fast Domain Adaptation Network for Image Super-Resolution
Image and GraphicsAbstractMost previous super-resolution (SR) methods are based on high-resolution (HR) images and corresponding low-resolution (LR) images obtained artificially through bicubic downsampling. However, in real scenes, LR images are usually obtained with ...
Instance Similarity-based Adversarial Domain Adaptation Network for Hyperspectral Image Classification
ICIGP '24: Proceedings of the 2024 7th International Conference on Image and Graphics ProcessingIn recent years, deep domain adaptation (DA) technology has been widely used in cross-domain hyperspectral image (HSI) classification problems. However, most existing deep DA algorithms aim to align the distributions of domains while ignoring sample ...
Comments