Hyperspectral anomaly detection based on the weakly supervised robust autoencoder
ZHANG Guojian1(), LIU Shengzhen2(), SUN Yingjun1, YU Kaijie3, LIU Lina4
1. College of Surveying and Geo-Informatics, Shandong Jianzhu University, Jinan 250101, China 2. First Geodetic Survey Team, Ministry of Natural Resources, Xi’an 710054, China 3. Zhejiang Zhongcexintu Geographic Information Technology Co., Huzhou 313200, China 4. Ningbo Yinzhou District Surveying and Mapping Institute, Ningbo 315041, China
Hyperspectral anomaly detection has received particular attention due to its unsupervised detection of targets. Moreover, autoencoder (AE), together with its variants, can automatically extract deep features and detect anomalous targets. However, AE is highly generalizable due to the existence of anomalies in the training set, thus suffering a reduced ability to distinguish anomalies from the background. This study proposed an anomaly detection algorithm based on the weakly supervised robust AE (WSRAE). First, this study developed a salient category search strategy and used probability-based category thresholds to label coarse samples in order to make preparation for network-based weakly supervised learning. Moreover, this study constructed a robust AE framework constrained jointly by l2,1 norm and anomaly-background spectral distances. This framework was robust with regard to noise and anomalies during training. Finally, this study detected anomalous targets based on the reconstruction errors obtained from all samples. Experiments on four hyperspectral datasets show that the WSRAE algorithm has greater detection performance than other state-of-the-art anomaly detection algorithms.
Cheng B Z. Study and progress of anomaly target detection in hyperspectral imagery[J]. Remote Sensing for Land and Resources, 2014, 26(3):1-7.doi:10.6046/gtzyyg.2014.03.01.
doi: 10.6046/gtzyyg.2014.03.01
Hou Z F, Liu R Y, Yan B K, et al. Hyperspectral imagery anomaly detection based on band selection and learning dictionary[J]. Remote Sensing for Land and Resources, 2019, 31(1),33-41.doi:10.6046/gtzyyg.2019.01.05.
doi: 10.6046/gtzyyg.2019.01.05
[5]
Reed I S, Yu X. Adaptive multiple-band CFAR detection of an optical pattern with unknown spectral distribution[J]. IEEE Transactions on Acoustics,Speech,and Signal Processing, 1990, 38(10):1760-1770.
doi: 10.1109/29.60107
[6]
Kwon H, Nasrabad N M. Kernel RX-algorithm:A nonlinear anomaly detector for hyperspectral imagery[J]. IEEE Transactions on Geoscience and Remote Sensing, 2005, 43(2):388-397.
doi: 10.1109/TGRS.2004.841487
[7]
Guo Q, Zhang B, Ran Q, et al. Weighted-RXD and linear filter-based RXD:Improving background statistics estimation for anomaly detection in hyperspectral imagery[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2014, 7(6):2351-2366.
doi: 10.1109/JSTARS.4609443
[8]
Du B, Zhang L. Random-selection-based anomaly detector for hyperspectral imagery[J]. IEEE Transactions on Geoscience and Remote sensing, 2010, 49(5):1578-1589.
doi: 10.1109/TGRS.2010.2081677
[9]
Li W, Du Q. Collaborative representation for hyperspectral anomaly detection[J]. IEEE Transactions on Geoscience and Remote Sensing, 2014, 53(3):1463-1474.
doi: 10.1109/TGRS.2014.2343955
Liu W J, Wu X J, Qu H C, et al. Improved collaborative representation for hyperspectral imagery anomaly detection algorithm[J]. Application Research of Computers, 2018 (12):68.
[11]
Zhang Y, Du B, Zhang L, et al. A low-rank and sparse matrix decomposition-based Mahalanobis distance method for hyperspectral anomaly detection[J]. IEEE Transactions on Geoscience and Remote Sensing, 2015, 54(3):1376-1389.
doi: 10.1109/TGRS.2015.2479299
[12]
Tao R, Zhao X, Li W, et al. Hyperspectral anomaly detection by fractional Fourier entropy[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2019, 12(12):4920-4929.
doi: 10.1109/JSTARS.2019.2940278
[13]
Xie W, Liu B, Li Y, et al. Autoencoder and adversarial-learning-based semisupervised background estimation for hyperspectral anomaly detection[J]. IEEE Transactions on Geoscience and Remote Sensing, 2020, 58(8):5416-5427.
doi: 10.1109/TGRS.36
Xue B, Wang Y Z, Liu S H, et al. Change detection of high-resolution remote sensing images based on Siamese network[J]. Remote Sensing for Land and Resources, 2022, 34(1):61-66.doi:10.6046/zrzyyg.2021122.
doi: 10.6046/zrzyyg.2021122
Han Y L, Cui P X, Yang S H, et al. Classification of hyperspectral images based on feature fusion of residual network[J]. Remote Sensing for Land and Resources, 2021, 33(2):11-19.doi:10.6046/gtzyyg.2020209.
doi: 10.6046/gtzyyg.2020209
[16]
Xie W, Lei J, Liu B, et al. Spectral constraint adversarial autoencoders approach to feature representation in hyperspectral anomaly detection[J]. Neural Networks, 2019, 119:222-234.
doi: S0893-6080(19)30229-1
pmid: 31472289
[17]
Lu X, Zhang W, Huang J. Exploiting embedding manifold of autoencoders for hyperspectral anomaly detection[J]. IEEE Transactions on Geoscience and Remote Sensing, 2019, 58(3):1527-1537.
doi: 10.1109/TGRS.36
[18]
Zhao C, Zhang L. Spectral-spatial stacked autoencoders based on low-rank and sparse matrix decomposition for hyperspectral anomaly detection[J]. Infrared Physics & Technology, 2018, 92:166-176.
[19]
Parimala M, Lopez D, Senthilkumar N C. A survey on density based clustering algorithms for mining large spatial databases[J]. International Journal of Advanced Science and Technology, 2011, 31(1):59-66.
[20]
Ma Y, Li C, Mei X, et al. Robust sparse hyperspectral unmixing with l2,1 norm[J]. IEEE Transactions on Geoscience and Remote Sensing, 2016, 55(3):1227-1239.
doi: 10.1109/TGRS.2016.2616161
[21]
Gong D, Liu L, Le V, et al. Memorizing normality to detect anomaly:Memory-augmented deep autoencoder for unsupervised anomaly detection[C]// Proceedings of the IEEE/CVF International Conference on Computer Vision. 2019:1705-1714.
2023, Vol. 35 
), 刘胜震2(
