Hyperspectral anomaly detection based on the weakly supervised robust autoencoder

ZHANG Guojian; LIU Shengzhen; SUN Yingjun; YU Kaijie; LIU Lina

doi:10.6046/zrzyyg.2022093

2023 Vol. 35, No. 2

Article Contents

Article navigation > Remote Sensing for Natural Resources > 2023 > 35(2): 167-175

Next Article Previous Article

ZHANG Guojian, LIU Shengzhen, SUN Yingjun, YU Kaijie, LIU Lina. 2023. Hyperspectral anomaly detection based on the weakly supervised robust autoencoder. Remote Sensing for Natural Resources, 35(2): 167-175. doi: 10.6046/zrzyyg.2022093

Citation:

ZHANG Guojian, LIU Shengzhen, SUN Yingjun, YU Kaijie, LIU Lina. 2023. Hyperspectral anomaly detection based on the weakly supervised robust autoencoder. Remote Sensing for Natural Resources, 35(2): 167-175. doi: 10.6046/zrzyyg.2022093

Hyperspectral anomaly detection based on the weakly supervised robust autoencoder

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

More Information

Corresponding author: LIU Shengzhen

Received Date: 21 March 2022

Revised Date: 16 August 2022

Publish Date: 07 July 2023

Abstract

Abstract

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 l_2,1 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.
- hyperspectral image /
- anomaly detection /
- salient category search /
- robust AE