Abstract
Oil spill monitoring in remote sensing field has become a very popular technology to detect the spatial distribution of polluted regions. However, previous studies mainly focus on the supervised detection technologies, which requires a large number of high-quality training set. To solve this problem, we propose a self-supervised learning method to learn the deep neural network from unlabelled hyperspectral data for oil spill detection, which consists of three parts: data augmentation, unsupervised deep feature learning, and oil spill detection network. First, the original image is augmented with spectral and spatial transformation to improve robustness of the self-supervised model. Then, the deep neural networks are trained on the augmented data without label information to produce the high-level semantic features. Finally, the pre-trained parameters are transferred to establish a neural network classifier to obtain the detection result, where a contrastive loss is developed to fine-tune the learned parameters so as to improve the generalization ability of the proposed method. Experiments performed on ten oil spill datasets reveal that the proposed method obtains promising detection performance with respect to other state-of-the-art hyperspectral detection approaches.
Article PDF
Similar content being viewed by others
Explore related subjects
Discover the latest articles, news and stories from top researchers in related subjects.Avoid common mistakes on your manuscript.
References
Wang X Z, Liu D, Cheng G, et al. Solar heating assisted rapid cleanup of viscous crude oil spills using reduced graphene oxide-coated sponges. Sci China Tech Sci, 2020, 63: 1487–1496
Gu Y F, Jin X D, Xiang R Z, et al. UAV-based integrated multi-spectral-LiDAR imaging system and data processing. Sci China Tech Sci, 2020, 63: 1293–1301
Qin F K, Chen S T, Chen R, et al. Leakage detection of oil tank using terahertz spectroscopy. Sci China Tech Sci, 2021, 64: 1947–1952
Duan P, Ghamisi P, Kang X, et al. Fusion of dual spatial information for hyperspectral image classification. IEEE Trans Geosci Remote Sens, 2021, 59: 7726–7738
Zhang S, Kang X, Duan P, et al. Polygon structure-guided hyperspectral image classification with single sample for strong geometric characteristics scenes. IEEE Trans Geosci Remote Sens, 2022, 60: 1–12
Kang X, Duan P, Xiang X, et al. Detection and correction of mislabeled training samples for hyperspectral image classification. IEEE Trans Geosci Remote Sens, 2018, 56: 5673–5686
Lu Y C, Shi J, Hu C M, et al. Optical interpretation of oil emulsions in the ocean. Part II: Applications to multi-band coarse-resolution imagery. Remote Sens Environ, 2020, 242: 111778
Pelta R, Carmon N, Ben-Dor E. A machine learning approach to detect crude oil contamination in a real scenario using hyperspectral remote sensing. Int J Appl Earth Observ GeoInf, 2019, 82: 101901
Yang J, Ren G, Ma Y, et al. Oil spill AISA+ hyperspectral data detection based on different sea surface glint suppression methods. Int Arch Photogramm Remote Sens Spatial Inf Sci, 2018, XLII-3: 2083–2087
Löw F, Stieglitz K, Diemar O. Terrestrial oil spill mapping using satellite earth observation and machine learning: A case study in South Sudan. J Environ Manage, 2021, 298: 113424
Duan P, Lai J, Kang J, et al. Texture-aware total variation-based removal of sun glint in hyperspectral images. ISPRS J Photogram Remote Sens, 2020, 166: 359–372
Liu D, Han L. Spectral curve shape matching using derivatives in hyperspectral images. IEEE Geosci Remote Sens Lett, 2017, 14: 504–508
Liu D, Zhang J, Wang X. Reference spectral signature selection using density-based cluster for automatic oil spill detection in hyperspectral images. Opt Express, 2016, 24: 7411
Song D M, Liu B, Chen S C, et al. Classification of the different thickness of the oil film based on wavelet transform spectrum information. Aquat Procedia, 2015, 3: 133–143
Liu B, Li Y, Chen P, et al. Extraction of oil spill information using decision tree based minimum noise fraction transform. J Ind Soc Remote Sens, 2016, 44: 421–426
Song M P, Chang M, An J B, et al. Active contour segmentation for hyperspectral oil spill remote sensing. In: Proceedings of the Society of Photo-Optical Instrumentation Engineers (SPIE) Conference Series. Beijing, 2013. 891026
Wang B, Shao Q, Song D, et al. A spectral-spatial features integrated network for hyperspectral detection of marine oil spill. Remote Sens, 2021, 13: 1568
Zhu X, Li Y, Zhang Q, et al. Oil film classification using deep learning-based hyperspectral remote sensing technology. Int J Geo-Infor, 2019, 8: 181
Yang J F, Wan J H, Ma Y, et al. Oil spill hyperspectral remote sensing detection based on DCNN with multi-scale features. J Coast Res, 2019, 90: 332
Lan M, Zhang Y, Zhang L, et al. Global context based automatic road segmentation via dilated convolutional neural network. Inf Sci, 2020, 535: 156–171
Sun X, Qu Y, Gao L, et al. Target detection through tree-structured encoding for hyperspectral images. IEEE Trans Geosci Remote Sens, 2021, 59: 4233–4249
Yue J, Fang L, Rahmani H, et al. Self-supervised learning with adaptive distillation for hyperspectral image classification. IEEE Trans Geosci Remote Sens, 2022, 60: 1–13
Hong D, Gao L, Yao J, et al. Graph convolutional networks for hyperspectral image classification. IEEE Trans Geosci Remote Sens, 2021, 59: 5966–5978
Zheng K, Gao L, Liao W, et al. Coupled convolutional neural network with adaptive response function learning for unsupervised hyperspectral super resolution. IEEE Trans Geosci Remote Sens, 2021, 59: 2487–2502
Hong D, Gao L, Yokoya N, et al. More diverse means better: multimodal deep learning meets remote-sensing imagery classification. IEEE Trans Geosci Remote Sens, 2021, 59: 4340–4354
Oord van den A, Li Y, Vinyals O. Representation learning with contrastive predictive coding. arXiv: 1807.03748
He K, Fan H, Wu Y, et al. Momentum contrast for unsupervised visual representation learning. arXiv: 1911.05722
Joseph K J, Khan S, Khan F S, et al. Towards open world object detection. In: Proceedings of the IEEE Conference Comput Vis Pattern Recognit. Nashville, 2021. 5826–5836
Ben Hamida A, Benoit A, Lambert P, et al. 3-D deep learning approach for remote sensing image classification. IEEE Trans Geosci Remote Sens, 2018, 56: 4420–4434
Gillis N, Kuang D, Park H. Hierarchical clustering of hyperspectral images using rank-two nonnegative matrix factorization. IEEE Trans Geosci Remote Sens, 2015, 53: 2066–2078
Zhang Y, Du B, Zhang L, et al. A low-rank and sparse matrix decomposition-based mahalanobis distance method for hyperspectral anomaly detection. IEEE Trans Geosci Remote Sens, 2016, 54: 1376–1389
Li S, Zhang K, Duan P, et al. Hyperspectral anomaly detection with kernel isolation forest. IEEE Trans Geosci Remote Sens, 2020, 58: 319–329
Fawcett T. An introduction to ROC analysis. Pattern Recognit Lett, 2006, 27: 861–874
Author information
Authors and Affiliations
Corresponding author
Additional information
This work was supported by the National Natural Science Foundation of China (Grant No. 61890962 and 61871179), the Scientific Research Project of Hunan Education Department (Grant No. 19B105), the Natural Science Foundation of Hunan Province (Grant Nos. 2019JJ50036 and 2020GK2038), the National Key Research and Development Project (Grant No. 2021YFA0715203), the Hunan Provincial Natural Science Foundation for Distinguished Young Scholars (Grant No. 2021JJ022), and the Huxiang Young Talents Science and Technology Innovation Program (Grant No. 2020RC3013).
Rights and permissions
About this article
Cite this article
Duan, P., Xie, Z., Kang, X. et al. Self-supervised learning-based oil spill detection of hyperspectral images. Sci. China Technol. Sci. 65, 793–801 (2022). https://doi.org/10.1007/s11431-021-1989-9
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11431-021-1989-9