SI2FM: SID Isolation Double Forest Model for Hyperspectral Anomaly Detection
Hyperspectral image (HSI) anomaly detection (HSI-AD) has become a hot issue in hyperspectral information processing as a method for detecting undesired targets without a priori information against unknown background and target information, which can be better adapted to the needs of practical applic...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-01-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/15/3/612 |
| _version_ | 1797623358277287936 |
|---|---|
| author | Zhenhua Mu Ming Wang Yihan Wang Ruoxi Song Xianghai Wang |
| author_facet | Zhenhua Mu Ming Wang Yihan Wang Ruoxi Song Xianghai Wang |
| author_sort | Zhenhua Mu |
| collection | DOAJ |
| description | Hyperspectral image (HSI) anomaly detection (HSI-AD) has become a hot issue in hyperspectral information processing as a method for detecting undesired targets without a priori information against unknown background and target information, which can be better adapted to the needs of practical applications. However, the demanding detection environment with no prior and small targets, as well as the large data and high redundancy of HSI itself, make the study of HSI-AD very challenging. First, we propose an HSI-AD method based on the nonsubsampled shearlet transform (NSST) domain spectral information divergence isolation double forest (SI2FM) in this paper. Further, the method excavates the intrinsic deep correlation properties between NSST subband coefficients of HSI in two ways to provide synergistic constraints and guidance on the prediction of abnormal target coefficients. On the one hand, with the “difference band” as a guide, the global isolation forest and local isolation forest models are constructed based on the spectral information divergence (SID) attribute values of the difference band and the low-frequency and high-frequency subbands, and the anomaly scores are determined by evaluating the path lengths of the isolation binary tree nodes in the forest model to obtain a progressively optimized anomaly detection map. On the other hand, based on the relationship of NSST high-frequency subband coefficients of spatial-spectral dimensions, the three-dimensional forest structure is constructed to realize the co-optimization of multiple anomaly detection maps obtained from the isolation forest. Finally, the guidance of the difference band suppresses the background noise and anomaly interference to a certain extent, enhancing the separability of target and background. The two-branch collaborative optimization based on the NSST subband coefficient correlation mining of HSI enables the prediction of anomaly sample coefficients to be gradually improved from multiple perspectives, which effectively improves the accuracy of anomaly detection. The effectiveness of the algorithm is verified by comparing real hyperspectral datasets captured in four different scenes with eleven typical anomaly detection algorithms currently available. |
| first_indexed | 2024-03-11T09:27:47Z |
| format | Article |
| id | doaj.art-d6d848be7ab744bfa7ce2b84a9efac6d |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-03-11T09:27:47Z |
| publishDate | 2023-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-d6d848be7ab744bfa7ce2b84a9efac6d2023-11-16T17:51:43ZengMDPI AGRemote Sensing2072-42922023-01-0115361210.3390/rs15030612SI2FM: SID Isolation Double Forest Model for Hyperspectral Anomaly DetectionZhenhua Mu0Ming Wang1Yihan Wang2Ruoxi Song3Xianghai Wang4School of Geography, Liaoning Normal University, Dalian 116029, ChinaSchool of Computer and Information Technology, Liaoning Normal University, Dalian 116029, ChinaSchool of Computer and Information Technology, Liaoning Normal University, Dalian 116029, ChinaAerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100101, ChinaSchool of Geography, Liaoning Normal University, Dalian 116029, ChinaHyperspectral image (HSI) anomaly detection (HSI-AD) has become a hot issue in hyperspectral information processing as a method for detecting undesired targets without a priori information against unknown background and target information, which can be better adapted to the needs of practical applications. However, the demanding detection environment with no prior and small targets, as well as the large data and high redundancy of HSI itself, make the study of HSI-AD very challenging. First, we propose an HSI-AD method based on the nonsubsampled shearlet transform (NSST) domain spectral information divergence isolation double forest (SI2FM) in this paper. Further, the method excavates the intrinsic deep correlation properties between NSST subband coefficients of HSI in two ways to provide synergistic constraints and guidance on the prediction of abnormal target coefficients. On the one hand, with the “difference band” as a guide, the global isolation forest and local isolation forest models are constructed based on the spectral information divergence (SID) attribute values of the difference band and the low-frequency and high-frequency subbands, and the anomaly scores are determined by evaluating the path lengths of the isolation binary tree nodes in the forest model to obtain a progressively optimized anomaly detection map. On the other hand, based on the relationship of NSST high-frequency subband coefficients of spatial-spectral dimensions, the three-dimensional forest structure is constructed to realize the co-optimization of multiple anomaly detection maps obtained from the isolation forest. Finally, the guidance of the difference band suppresses the background noise and anomaly interference to a certain extent, enhancing the separability of target and background. The two-branch collaborative optimization based on the NSST subband coefficient correlation mining of HSI enables the prediction of anomaly sample coefficients to be gradually improved from multiple perspectives, which effectively improves the accuracy of anomaly detection. The effectiveness of the algorithm is verified by comparing real hyperspectral datasets captured in four different scenes with eleven typical anomaly detection algorithms currently available.https://www.mdpi.com/2072-4292/15/3/612hyperspectral imagenonsubsampled shearlet transformanomaly detectionspectral information divergenceisolation forestspatial-spectral dimensional forest |
| spellingShingle | Zhenhua Mu Ming Wang Yihan Wang Ruoxi Song Xianghai Wang SI2FM: SID Isolation Double Forest Model for Hyperspectral Anomaly Detection Remote Sensing hyperspectral image nonsubsampled shearlet transform anomaly detection spectral information divergence isolation forest spatial-spectral dimensional forest |
| title | SI2FM: SID Isolation Double Forest Model for Hyperspectral Anomaly Detection |
| title_full | SI2FM: SID Isolation Double Forest Model for Hyperspectral Anomaly Detection |
| title_fullStr | SI2FM: SID Isolation Double Forest Model for Hyperspectral Anomaly Detection |
| title_full_unstemmed | SI2FM: SID Isolation Double Forest Model for Hyperspectral Anomaly Detection |
| title_short | SI2FM: SID Isolation Double Forest Model for Hyperspectral Anomaly Detection |
| title_sort | si2fm sid isolation double forest model for hyperspectral anomaly detection |
| topic | hyperspectral image nonsubsampled shearlet transform anomaly detection spectral information divergence isolation forest spatial-spectral dimensional forest |
| url | https://www.mdpi.com/2072-4292/15/3/612 |
| work_keys_str_mv | AT zhenhuamu si2fmsidisolationdoubleforestmodelforhyperspectralanomalydetection AT mingwang si2fmsidisolationdoubleforestmodelforhyperspectralanomalydetection AT yihanwang si2fmsidisolationdoubleforestmodelforhyperspectralanomalydetection AT ruoxisong si2fmsidisolationdoubleforestmodelforhyperspectralanomalydetection AT xianghaiwang si2fmsidisolationdoubleforestmodelforhyperspectralanomalydetection |