An Unsupervised Deep Hyperspectral Anomaly Detector
Hyperspectral image (HSI) based detection has attracted considerable attention recently in agriculture, environmental protection and military applications as different wavelengths of light can be advantageously used to discriminate different types of objects. Unfortunately, estimating the background...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2018-02-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | http://www.mdpi.com/1424-8220/18/3/693 |
| _version_ | 1828370505884762112 |
|---|---|
| author | Ning Ma Yu Peng Shaojun Wang Philip H. W. Leong |
| author_facet | Ning Ma Yu Peng Shaojun Wang Philip H. W. Leong |
| author_sort | Ning Ma |
| collection | DOAJ |
| description | Hyperspectral image (HSI) based detection has attracted considerable attention recently in agriculture, environmental protection and military applications as different wavelengths of light can be advantageously used to discriminate different types of objects. Unfortunately, estimating the background distribution and the detection of interesting local objects is not straightforward, and anomaly detectors may give false alarms. In this paper, a Deep Belief Network (DBN) based anomaly detector is proposed. The high-level features and reconstruction errors are learned through the network in a manner which is not affected by previous background distribution assumption. To reduce contamination by local anomalies, adaptive weights are constructed from reconstruction errors and statistical information. By using the code image which is generated during the inference of DBN and modified by adaptively updated weights, a local Euclidean distance between under test pixels and their neighboring pixels is used to determine the anomaly targets. Experimental results on synthetic and recorded HSI datasets show the performance of proposed method outperforms the classic global Reed-Xiaoli detector (RXD), local RX detector (LRXD) and the-state-of-the-art Collaborative Representation detector (CRD). |
| first_indexed | 2024-04-14T06:38:54Z |
| format | Article |
| id | doaj.art-094c3f80a319437ab34ecfa410dfe86f |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-04-14T06:38:54Z |
| publishDate | 2018-02-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-094c3f80a319437ab34ecfa410dfe86f2022-12-22T02:07:23ZengMDPI AGSensors1424-82202018-02-0118369310.3390/s18030693s18030693An Unsupervised Deep Hyperspectral Anomaly DetectorNing Ma0Yu Peng1Shaojun Wang2Philip H. W. Leong3Department of Automatic Test and Control, School of Electrical Engineering and Automation, Harbin Institute of Technology, Harbin 150080, ChinaDepartment of Automatic Test and Control, School of Electrical Engineering and Automation, Harbin Institute of Technology, Harbin 150080, ChinaDepartment of Automatic Test and Control, School of Electrical Engineering and Automation, Harbin Institute of Technology, Harbin 150080, ChinaSchool of Electrical and Information Engineering, The University of Sydney, Sydney 2006, AustraliaHyperspectral image (HSI) based detection has attracted considerable attention recently in agriculture, environmental protection and military applications as different wavelengths of light can be advantageously used to discriminate different types of objects. Unfortunately, estimating the background distribution and the detection of interesting local objects is not straightforward, and anomaly detectors may give false alarms. In this paper, a Deep Belief Network (DBN) based anomaly detector is proposed. The high-level features and reconstruction errors are learned through the network in a manner which is not affected by previous background distribution assumption. To reduce contamination by local anomalies, adaptive weights are constructed from reconstruction errors and statistical information. By using the code image which is generated during the inference of DBN and modified by adaptively updated weights, a local Euclidean distance between under test pixels and their neighboring pixels is used to determine the anomaly targets. Experimental results on synthetic and recorded HSI datasets show the performance of proposed method outperforms the classic global Reed-Xiaoli detector (RXD), local RX detector (LRXD) and the-state-of-the-art Collaborative Representation detector (CRD).http://www.mdpi.com/1424-8220/18/3/693hyperspectral imagedeep learninganomaly detection |
| spellingShingle | Ning Ma Yu Peng Shaojun Wang Philip H. W. Leong An Unsupervised Deep Hyperspectral Anomaly Detector Sensors hyperspectral image deep learning anomaly detection |
| title | An Unsupervised Deep Hyperspectral Anomaly Detector |
| title_full | An Unsupervised Deep Hyperspectral Anomaly Detector |
| title_fullStr | An Unsupervised Deep Hyperspectral Anomaly Detector |
| title_full_unstemmed | An Unsupervised Deep Hyperspectral Anomaly Detector |
| title_short | An Unsupervised Deep Hyperspectral Anomaly Detector |
| title_sort | unsupervised deep hyperspectral anomaly detector |
| topic | hyperspectral image deep learning anomaly detection |
| url | http://www.mdpi.com/1424-8220/18/3/693 |
| work_keys_str_mv | AT ningma anunsuperviseddeephyperspectralanomalydetector AT yupeng anunsuperviseddeephyperspectralanomalydetector AT shaojunwang anunsuperviseddeephyperspectralanomalydetector AT philiphwleong anunsuperviseddeephyperspectralanomalydetector AT ningma unsuperviseddeephyperspectralanomalydetector AT yupeng unsuperviseddeephyperspectralanomalydetector AT shaojunwang unsuperviseddeephyperspectralanomalydetector AT philiphwleong unsuperviseddeephyperspectralanomalydetector |