A novel graph convolution and frequency domain filtering approach for hyperspectral anomaly detection
Abstract This paper introduces a novel algorithm for hyperspectral anomaly detection (HAD) that combines graph-based representations with frequency domain filtering techniques. In this approach, hyperspectral images (HSIs) are modeled as graphs, where each pixel is treated as a node with spectral fe...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Springer
2025-01-01
|
| Series: | Complex & Intelligent Systems |
| Subjects: | |
| Online Access: | https://doi.org/10.1007/s40747-024-01738-z |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832571203943923712 |
|---|---|
| author | Yang Ding Hao Yan Jingyuan He Juanjuan Yin A. Ruhan |
| author_facet | Yang Ding Hao Yan Jingyuan He Juanjuan Yin A. Ruhan |
| author_sort | Yang Ding |
| collection | DOAJ |
| description | Abstract This paper introduces a novel algorithm for hyperspectral anomaly detection (HAD) that combines graph-based representations with frequency domain filtering techniques. In this approach, hyperspectral images (HSIs) are modeled as graphs, where each pixel is treated as a node with spectral features, and the edges capture pixel correlations based on the K-Nearest Neighbor (KNN) algorithm. Graph convolution is employed to extract spatial structural features, enhancing the understanding of spatial relationships within the data. Additionally, the algorithm addresses the ’right-shift’ phenomenon in the spectral domain, often associated with anomalies, by using a beta wavelet filter for efficient spectral filtering and anomaly detection. The key contributions of this work include: 1) the use of a graph-based model for HSI that effectively integrates both spatial and spectral dimensions, 2) employing KNN for edge construction to include distant pixels and mitigate noise, 3) spatial feature extraction via graph convolution to provide detailed insights into spatial interconnections and variations, enhancing the detection process, and 4) leveraging the beta wavelet filter to handle the ’right-shift’ spectral phenomenon and reduce computational complexity. Experimental evaluations on four benchmark datasets show that the proposed method achieves outstanding performance with AUC scores of 0.9986, 0.9975, 0.9859, and 0.9988, significantly outperforming traditional and state-of-the-art anomaly detection techniques. |
| format | Article |
| id | doaj-art-e272f40b51dd42588801553a4eea569a |
| institution | Kabale University |
| issn | 2199-4536 2198-6053 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | Springer |
| record_format | Article |
| series | Complex & Intelligent Systems |
| spelling | doaj-art-e272f40b51dd42588801553a4eea569a2025-02-02T12:49:56ZengSpringerComplex & Intelligent Systems2199-45362198-60532025-01-0111112310.1007/s40747-024-01738-zA novel graph convolution and frequency domain filtering approach for hyperspectral anomaly detectionYang Ding0Hao Yan1Jingyuan He2Juanjuan Yin3A. Ruhan4Xi’an Peihua UniversityDalian University of TechnologyYan’an UniversityNorthwest UniversityXi’an Peihua UniversityAbstract This paper introduces a novel algorithm for hyperspectral anomaly detection (HAD) that combines graph-based representations with frequency domain filtering techniques. In this approach, hyperspectral images (HSIs) are modeled as graphs, where each pixel is treated as a node with spectral features, and the edges capture pixel correlations based on the K-Nearest Neighbor (KNN) algorithm. Graph convolution is employed to extract spatial structural features, enhancing the understanding of spatial relationships within the data. Additionally, the algorithm addresses the ’right-shift’ phenomenon in the spectral domain, often associated with anomalies, by using a beta wavelet filter for efficient spectral filtering and anomaly detection. The key contributions of this work include: 1) the use of a graph-based model for HSI that effectively integrates both spatial and spectral dimensions, 2) employing KNN for edge construction to include distant pixels and mitigate noise, 3) spatial feature extraction via graph convolution to provide detailed insights into spatial interconnections and variations, enhancing the detection process, and 4) leveraging the beta wavelet filter to handle the ’right-shift’ spectral phenomenon and reduce computational complexity. Experimental evaluations on four benchmark datasets show that the proposed method achieves outstanding performance with AUC scores of 0.9986, 0.9975, 0.9859, and 0.9988, significantly outperforming traditional and state-of-the-art anomaly detection techniques.https://doi.org/10.1007/s40747-024-01738-zHyperspectral anomaly detection (HAD)Graph neural networks (GNN)Graph convolutional networks(GCN)Right-shift phenomenonBeta wavelet |
| spellingShingle | Yang Ding Hao Yan Jingyuan He Juanjuan Yin A. Ruhan A novel graph convolution and frequency domain filtering approach for hyperspectral anomaly detection Complex & Intelligent Systems Hyperspectral anomaly detection (HAD) Graph neural networks (GNN) Graph convolutional networks(GCN) Right-shift phenomenon Beta wavelet |
| title | A novel graph convolution and frequency domain filtering approach for hyperspectral anomaly detection |
| title_full | A novel graph convolution and frequency domain filtering approach for hyperspectral anomaly detection |
| title_fullStr | A novel graph convolution and frequency domain filtering approach for hyperspectral anomaly detection |
| title_full_unstemmed | A novel graph convolution and frequency domain filtering approach for hyperspectral anomaly detection |
| title_short | A novel graph convolution and frequency domain filtering approach for hyperspectral anomaly detection |
| title_sort | novel graph convolution and frequency domain filtering approach for hyperspectral anomaly detection |
| topic | Hyperspectral anomaly detection (HAD) Graph neural networks (GNN) Graph convolutional networks(GCN) Right-shift phenomenon Beta wavelet |
| url | https://doi.org/10.1007/s40747-024-01738-z |
| work_keys_str_mv | AT yangding anovelgraphconvolutionandfrequencydomainfilteringapproachforhyperspectralanomalydetection AT haoyan anovelgraphconvolutionandfrequencydomainfilteringapproachforhyperspectralanomalydetection AT jingyuanhe anovelgraphconvolutionandfrequencydomainfilteringapproachforhyperspectralanomalydetection AT juanjuanyin anovelgraphconvolutionandfrequencydomainfilteringapproachforhyperspectralanomalydetection AT aruhan anovelgraphconvolutionandfrequencydomainfilteringapproachforhyperspectralanomalydetection AT yangding novelgraphconvolutionandfrequencydomainfilteringapproachforhyperspectralanomalydetection AT haoyan novelgraphconvolutionandfrequencydomainfilteringapproachforhyperspectralanomalydetection AT jingyuanhe novelgraphconvolutionandfrequencydomainfilteringapproachforhyperspectralanomalydetection AT juanjuanyin novelgraphconvolutionandfrequencydomainfilteringapproachforhyperspectralanomalydetection AT aruhan novelgraphconvolutionandfrequencydomainfilteringapproachforhyperspectralanomalydetection |