高光谱图像稀疏约束与自编码器特征提取相结合的异常检测方法

doi:10.11947/j.AGCS.2023.20210604

测绘学报 ›› 2023, Vol. 52 ›› Issue (6): 932-943.doi: 10.11947/j.AGCS.2023.20210604

高光谱图像稀疏约束与自编码器特征提取相结合的异常检测方法

宋尚真¹, 杨怡欣², 王会峰¹, 王晓艳¹, 荣生辉³, 周慧鑫⁴

1. 长安大学电子与控制工程学院, 陕西西安 710061;
2. 西安邮电大学通信与信息工程学院, 陕西西安 710121;
3. 中国海洋大学电子工程学院, 山东青岛 266100;
4. 西安电子科技大学物理与光电工程学院, 陕西西安 710071

收稿日期:2021-10-28 修回日期:2022-05-11 发布日期:2023-07-08
通讯作者: 王会峰 E-mail:hfwang@chd.edu.cn
作者简介:宋尚真(1991-),男,博士,讲师,研究方向为高光谱图像处理与智能解译。E-mail:szsong@chd.edu.cn
基金资助:
国家自然科学基金(52172324);陕西省重点研发计划(2021GY-285;2021SF-483)

Hyperspectral anomaly detection combining sparse constraint and feature extraction via stacked autoencoder

SONG Shangzhen¹, YANG Yixin², WANG Huifeng¹, WANG Xiaoyan¹, RONG Shenghui³, ZHOU Huixin⁴

1. School of Electronics and Control Engineering, Chang'an University, Xi'an 710064, China;
2. School of Communications and Information Engineering, Xi'an Uiversity of Posts &Telecommunications, Xi'an 710121, China;
3. School of Electronic Engineering, Ocean University of China, Qingdao 266100, China;
4. School of Physics and Radio and Television Engineering, Xi'an University of Electronic Science and Technology, Xi'an 710068, China

Received:2021-10-28 Revised:2022-05-11 Published:2023-07-08
Supported by:
The National Natural Science Foundation of China (No. 52172324); The Key Research and Development Program of Shaanxi Province (Nos. 2021GY-285; 2021SF-483)

摘要/Abstract

摘要： 高光谱图像的异常检测在军事、农业、勘探、防火等领域具有重要的应用价值。传统的高光谱图像异常检测算法未能有效地挖掘图像光谱的深层特征,而深度学习方法具有良好的提取深层特征信息的能力。由于异常检测问题一般无法获取地物先验信息,因此无监督网络相比于监督网络要更为适用。而现有的基于自编码器的异常检测算法没有对局部信息进行有效利用,导致检测效果受限。针对这一问题,本文提出一种基于稀疏表示约束的自编码器深度特征提取方法。首先通过栈式自编码器得到深层次语义信息;然后利用稀疏表示作为约束与编码器进行有效结合,挖掘了潜在隐藏空间中的特征元素的局部表示特性;最后采用分数傅里叶变换,通过空间-频率表示获得原始光谱与其傅里叶变换的中间域中的特征,进一步增强了背景和异常的光谱区分度,且能有效去除噪声的影响。在Hymap、AVIRIS、ROSIS、HYDICE这4种光谱仪采集的5幅高光谱遥感影像上进行了性能验证,得到的曲线下覆盖面积(area under curve,AUC)分别为0.990 5、0.998 3、0.999 0、0.992 8和0.911 0,相比于对比算法都有了不同程度的效果提升。结果表明本文方法具有更好的检测精度。

关键词: 高光谱影像, 异常检测, 深度学习, 自编码器, 稀疏表示, 傅里叶变换

Abstract: Anomaly detection of hyperspectral images has important application value in military, agriculture, exploration, fire protection and other fields. Traditional algorithms of hyperspectral image (HSI) anomaly detection (AD) do not effectively mine the deep features of the image spectrum, while the deep learning method has good ability to extract deep feature information. Since the AD problem generally cannot obtain the prior information in advance, the unsupervised network is more suitable. Existing AD algorithms based on autoencoder (AE) does not make effective use of the local information, resulting in limited detection effect. To overcome this shortcoming, the paper proposes an AD method based on sparse representation (SR) constraints for stacked autoencoder (SAE). Firstly, the semantic information is obtained by SAE. Secondly, the SR is used as a constraint to effectively combine with the encoder, and the local characteristics of the feature elements in the potential hidden space are mined. Finally, the fractional Fourier transform is utilized, and the characteristics of the original spectrum and its intermediate domain of Fourier transform are obtained by spatial-frequency representation. Consequently, the spectral discrimination between background and anomalies is further enhanced, and the effect of noise is also removed. The experiment performs verification on 5 HSIs collected by 4 spectrometers including Hymap, AVIRIS, ROSIS, and HYDICE. The area under curve (AUC) values are 0.990 5, 0.998 3, 0.999 0, 0.992 8 and 0.911 0, respectively. Compared with compared algorithms, the effect of the proposed algorithm can be improved.

Key words: hyperspectral imagery, anomaly detection, deep learning, autoencoder, sparse representation, Fourier transform

中图分类号:

P227

宋尚真, 杨怡欣, 王会峰, 王晓艳, 荣生辉, 周慧鑫. 高光谱图像稀疏约束与自编码器特征提取相结合的异常检测方法[J]. 测绘学报, 2023, 52(6): 932-943.

SONG Shangzhen, YANG Yixin, WANG Huifeng, WANG Xiaoyan, RONG Shenghui, ZHOU Huixin. Hyperspectral anomaly detection combining sparse constraint and feature extraction via stacked autoencoder[J]. Acta Geodaetica et Cartographica Sinica, 2023, 52(6): 932-943.

参考文献

[1] KATTENBORN T, LEITLOFF J, SCHIEFER F, et al. Review on convolutional neural networks (CNN) in vegetation remote sensing[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2021, 173: 24-49.
[2] LI Yansheng, MA Jiayi, ZHANG Yongjun. Image retrieval from remote sensing big data: a survey[J]. Information Fusion, 2021, 67: 94-115.
[3] 李敏, 朱国康, 张学武, 等. 基于多孔径映射的高光谱异常检测算法[J]. 测绘学报, 2016, 45(10): 1222-1230.DOI: 10.11947/j.AGCS.2016.20160119. LI Min, ZHU Guokang, ZHANG Xuewu, et al. An anomaly detector based on multi-aperture mapping for hyperspectral data[J]. Acta Geodaetica et Cartographica Sinica, 2016, 45(10): 1222-1230.DOI: 10.11947/j.AGCS.2016.20160119.
[4] 黄远程, 薛园园, 李朋飞. 高光谱影像子空间分析孤立森林异常目标探测方法[J]. 测绘学报, 2021, 50(3): 416-425.DOI: 10.11947/j.AGCS.2021.20200036. HUANG Yuancheng, XUE Yuanyuan, LI Pengfei. Subspace analysis isolation forest for hyperspectral anomaly detection[J]. Acta Geodaetica et Cartographica Sinica, 2021, 50(3): 416-425.DOI: 10.11947/j.AGCS.2021.20200036.
[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.
[6] CHEN J Y, REED I S. A detection algorithm for optical targets in clutter[J]. IEEE Transactions on Aerospace and Electronic Systems, 1987, 23(1): 46-59.
[7] SCHAUM A. Joint subspace detection of hyperspectral targets[C]//Proceedings of 2004 IEEE Aerospace Conference. Big Sky: IEEE, 2004.
[8] KWON H, NASRABADI 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.
[9] WANG Kaidong, WANG Yao, ZHAO Xile, et al. Hyperspectral and multispectral image fusion via nonlocal low-rank tensor decomposition and spectral unmixing[J]. IEEE Transactions on Geoscience and Remote Sensing, 2020, 58(11): 7654-7671.
[10] LI Wei, DU Qian. Collaborative representation for hyperspectral anomaly detection[J]. IEEE Transactions on Geoscience and Remote Sensing, 2015, 53(3): 1463-1474.
[11] SUN Weiwei, LIU Chun, LI Jialin, et al. Low-rank and sparse matrix decomposition-based anomaly detection for hyperspectral imagery[J]. Journal of Applied Remote Sensing, 2014, 8(1): 083641.
[12] XU Yang, WU Zebin, LI Jun, et al. Anomaly detection in hyperspectral images based on low-rank and sparse representation[J]. IEEE Transactions on Geoscience and Remote Sensing, 2016, 54(4): 1990-2000.
[13] QU Ying, WANG Wei, GUO Rui, et al. Hyperspectral anomaly detection through spectral unmixing and dictionary-based low-rank decomposition[J]. IEEE Transactions on Geoscience and Remote Sensing, 2018, 56(8): 4391-4405.
[14] LEE H, KWON H. Going deeper with contextual CNN for hyperspectral image classification [J]. IEEE Transactions on Image Processing, 2017, 26(10): 4843-4855.
[15] MEI Shaohui, YUAN Xin, JI Jingyu, et al. Hyperspectral image spatial super-resolution via 3D full convolutional neural network[J]. Remote Sensing, 2017, 9(11): 1139.
[16] 蓝朝桢, 卢万杰, 于君明, 等. 异源遥感影像特征匹配的深度学习算法[J]. 测绘学报, 2021, 50(2): 189-202.DOI: 10.11947/j.AGCS.2021.20200048. LAN Chaozhen, LU Wanjie, YU Junming, et al. Deep learning algorithm for feature matching of cross modality remote sensing images[J]. Acta Geodaetica et Cartographica Sinica, 2021, 50(2): 189-202.DOI: 10.11947/j.AGCS.2021.20200048.
[17] ZHANG Xiangrong, SUN Yujia, ZHANG Jingyan, et al. Hyperspectral unmixing via deep convolutional neural networks[J]. IEEE Geoscience and Remote Sensing Letters, 2018, 15(11): 1755-1759.
[18] QIAN Yinlong, DONG Jing, WANG Wei, et al. Learning and transferring representations for image steganalysis using convolutional neural network[C]//Proceedings of 2016 IEEE International Conference on Image Processing. Phoenix: IEEE, 2016: 2752-2756.
[19] SHI Yanzi, LEI Jie, YIN Yaping, et al. Discriminative feature learning with distance constrained stacked sparse autoencoder for hyperspectral target detection[J]. IEEE Geoscience and Remote Sensing Letters, 2019, 16(9): 1462-1466.
[20] ZHAO Chunhui, LI Chuang, FENG Shou, et al. Spectral-spatial anomaly detection via collaborative representation constraint stacked autoencoders for hyperspectral images[J]. IEEE Geoscience and Remote Sensing Letters, 2022, 19: 1-5.
[21] DONOHO D L, TSAIG Y, DRORI I, et al. Sparse solution of underdetermined systems of linear equations by stagewise orthogonal matching pursuit[J]. IEEE Transactions on Information Theory, 2012, 58(2): 1094-1121.
[22] BENGIO Y, LAMBLIN P, POPOVICI D, et al. Greedy layer-wise training of deep networks[C]//Proceedings of the 19th International Conference on Neural Information Processing Systems. Hong Kong: IEEE, 2006: 153-160.
[23] REN X, LIN Z. Linearized alternating direction method with adaptive penalty and warm starts for fast solving transform invariant low-rank textures[J]. International journal of computer vision, 2013, 104: 1-14.
[24] CANDAN C, KUTAY M A, OZAKTAS H M. The discrete fractional Fourier transform[J]. IEEE Transactions on Signal Processing, 2000, 48(5): 1329-1337.
[25] TAO Ran, ZHAO Xudong, LI Wei, 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.
[26] WANG Shuihua, ZHANG Yudong, YANG Xiaojun, et al. Pathological brain detection by a novel image feature—fractional Fourier entropy[J]. Entropy, 2015, 17(12): 8278-8296.
[27] SNYDER D, KEREKES J, FAIRWEATHER I, et al. Development of a web-based application to evaluate target finding algorithms[C]//Proceedings of 2008 IEEE International Geoscience and Remote Sensing Symposium.Boston: IEEE, 2009: II-915.
[28] STEFANOU M S, KEREKES J P. A method for assessing spectral image utility[J]. IEEE Transactions on Geoscience and Remote Sensing, 2009, 47(6): 1698-1706.
[29] KEREKES J. Receiver operating characteristic curve confidence intervals and regions[J]. IEEE Geoscience and Remote Sensing Letters, 2008, 5(2): 251-255.

高光谱图像稀疏约束与自编码器特征提取相结合的异常检测方法

Hyperspectral anomaly detection combining sparse constraint and feature extraction via stacked autoencoder

RichHTML

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

本文评价

[1]	邓敏, 罗斌, 唐建波, 姚志鹏, 刘国平, 温翔, 胡润波, 柴华, 胡文柯. 顾及轨迹密度分布异质性的道路交叉口提取方法[J]. 测绘学报, 2023, 52(6): 1000-1009.
[2]	马梦锴, 董箭, 唐露露, 彭认灿, 周寅飞, 王芳. 兼顾要素正确分类及精准定位的栅格海图水深注记自动提取方法[J]. 测绘学报, 2023, 52(6): 1022-1036.
[3]	刘帅, 李笑迎, 于梦, 邢光龙. 高分辨率遥感图像双解耦语义分割网络模型[J]. 测绘学报, 2023, 52(4): 638-647.
[4]	张永显, 马国锐, 崔志祥, 张志军. 面向大视角差的无人机影像序列学习型特征匹配[J]. 测绘学报, 2023, 52(2): 230-243.
[5]	卜丽静, 吴文玉, 张正鹏, 杨银. 先验引导的多降质特征“珞珈一号”夜光遥感影像质量提升方法[J]. 测绘学报, 2023, 52(2): 244-259.
[6]	孙根云, 王鑫, 安娜, 张爱竹. 基于多源遥感的大尺度高分辨率不透水面深度学习提取方法[J]. 测绘学报, 2023, 52(2): 272-282.
[7]	禹文豪, 魏铖, 陈佳鑫. 面向不平衡POI类别分布的电子地图多层次标签预测模型[J]. 测绘学报, 2023, 52(2): 318-328.
[8]	杨敏, 陈果, 李连营, 黄浩然, 苗静, 晏雄锋. 序列卷积神经网络支持下线状地图目标的分段方法[J]. 测绘学报, 2023, 52(1): 108-116.
[9]	眭海刚, 刘畅, 干哲, 江政杰, 徐川. 多模态遥感图像匹配方法综述[J]. 测绘学报, 2022, 51(9): 1848-1861.
[10]	张良培, 何江, 杨倩倩, 肖屹, 袁强强. 数据驱动的多源遥感信息融合研究进展[J]. 测绘学报, 2022, 51(7): 1317-1337.
[11]	程涛, 张洋, James Haworth. 基于网络和图的时空智能——概念、方法和应用[J]. 测绘学报, 2022, 51(7): 1629-1639.
[12]	龚健雅, 宦麟茜, 郑先伟. 影像解译中的深度学习可解释性分析方法[J]. 测绘学报, 2022, 51(6): 873-884.
[13]	张勤, 赵超英, 陈雪蓉. 多源遥感地质灾害早期识别技术进展与发展趋势[J]. 测绘学报, 2022, 51(6): 885-896.
[14]	龚健雅, 张觅, 胡翔云, 张展, 李彦胜, 姜良存. 智能遥感深度学习框架与模型设计[J]. 测绘学报, 2022, 51(4): 475-487.
[15]	王蓝星, 王群明, 童小华. 融合多光谱影像的高光谱影像厚云去除方法[J]. 测绘学报, 2022, 51(4): 612-621.