A Weighted Spatial-Spectral Kernel RX Algorithm and Efficient Implementation on GPUs
The kernel RX (KRX) detector proposed by Kwon and Nasrabadi exploits a kernel function to obtain a better detection performance. However, it still has two limits that can be improved. On the one hand, reasonable integration of spatial-spectral information can be used to further improve its detection...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2017-02-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | http://www.mdpi.com/1424-8220/17/3/441 |
| _version_ | 1811301613331021824 |
|---|---|
| author | Chunhui Zhao Jiawei Li Meiling Meng Xifeng Yao |
| author_facet | Chunhui Zhao Jiawei Li Meiling Meng Xifeng Yao |
| author_sort | Chunhui Zhao |
| collection | DOAJ |
| description | The kernel RX (KRX) detector proposed by Kwon and Nasrabadi exploits a kernel function to obtain a better detection performance. However, it still has two limits that can be improved. On the one hand, reasonable integration of spatial-spectral information can be used to further improve its detection accuracy. On the other hand, parallel computing can be used to reduce the processing time in available KRX detectors. Accordingly, this paper presents a novel weighted spatial-spectral kernel RX (WSSKRX) detector and its parallel implementation on graphics processing units (GPUs). The WSSKRX utilizes the spatial neighborhood resources to reconstruct the testing pixels by introducing a spectral factor and a spatial window, thereby effectively reducing the interference of background noise. Then, the kernel function is redesigned as a mapping trick in a KRX detector to implement the anomaly detection. In addition, a powerful architecture based on the GPU technique is designed to accelerate WSSKRX. To substantiate the performance of the proposed algorithm, both synthetic and real data are conducted for experiments. |
| first_indexed | 2024-04-13T07:11:55Z |
| format | Article |
| id | doaj.art-1270687a68144d6d9c15473d256bbe21 |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-04-13T07:11:55Z |
| publishDate | 2017-02-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-1270687a68144d6d9c15473d256bbe212022-12-22T02:56:50ZengMDPI AGSensors1424-82202017-02-0117344110.3390/s17030441s17030441A Weighted Spatial-Spectral Kernel RX Algorithm and Efficient Implementation on GPUsChunhui Zhao0Jiawei Li1Meiling Meng2Xifeng Yao3College of Information and Communication Engineering, Harbin Engineering University, Harbin 150001, ChinaCollege of Information and Communication Engineering, Harbin Engineering University, Harbin 150001, ChinaCollege of Information and Communication Engineering, Harbin Engineering University, Harbin 150001, ChinaCollege of Information and Communication Engineering, Harbin Engineering University, Harbin 150001, ChinaThe kernel RX (KRX) detector proposed by Kwon and Nasrabadi exploits a kernel function to obtain a better detection performance. However, it still has two limits that can be improved. On the one hand, reasonable integration of spatial-spectral information can be used to further improve its detection accuracy. On the other hand, parallel computing can be used to reduce the processing time in available KRX detectors. Accordingly, this paper presents a novel weighted spatial-spectral kernel RX (WSSKRX) detector and its parallel implementation on graphics processing units (GPUs). The WSSKRX utilizes the spatial neighborhood resources to reconstruct the testing pixels by introducing a spectral factor and a spatial window, thereby effectively reducing the interference of background noise. Then, the kernel function is redesigned as a mapping trick in a KRX detector to implement the anomaly detection. In addition, a powerful architecture based on the GPU technique is designed to accelerate WSSKRX. To substantiate the performance of the proposed algorithm, both synthetic and real data are conducted for experiments.http://www.mdpi.com/1424-8220/17/3/441anomaly detectiongraphics processing units (GPUs)hyperspectral imagingkernel mappingspatial-spectral informationparallel processing |
| spellingShingle | Chunhui Zhao Jiawei Li Meiling Meng Xifeng Yao A Weighted Spatial-Spectral Kernel RX Algorithm and Efficient Implementation on GPUs Sensors anomaly detection graphics processing units (GPUs) hyperspectral imaging kernel mapping spatial-spectral information parallel processing |
| title | A Weighted Spatial-Spectral Kernel RX Algorithm and Efficient Implementation on GPUs |
| title_full | A Weighted Spatial-Spectral Kernel RX Algorithm and Efficient Implementation on GPUs |
| title_fullStr | A Weighted Spatial-Spectral Kernel RX Algorithm and Efficient Implementation on GPUs |
| title_full_unstemmed | A Weighted Spatial-Spectral Kernel RX Algorithm and Efficient Implementation on GPUs |
| title_short | A Weighted Spatial-Spectral Kernel RX Algorithm and Efficient Implementation on GPUs |
| title_sort | weighted spatial spectral kernel rx algorithm and efficient implementation on gpus |
| topic | anomaly detection graphics processing units (GPUs) hyperspectral imaging kernel mapping spatial-spectral information parallel processing |
| url | http://www.mdpi.com/1424-8220/17/3/441 |
| work_keys_str_mv | AT chunhuizhao aweightedspatialspectralkernelrxalgorithmandefficientimplementationongpus AT jiaweili aweightedspatialspectralkernelrxalgorithmandefficientimplementationongpus AT meilingmeng aweightedspatialspectralkernelrxalgorithmandefficientimplementationongpus AT xifengyao aweightedspatialspectralkernelrxalgorithmandefficientimplementationongpus AT chunhuizhao weightedspatialspectralkernelrxalgorithmandefficientimplementationongpus AT jiaweili weightedspatialspectralkernelrxalgorithmandefficientimplementationongpus AT meilingmeng weightedspatialspectralkernelrxalgorithmandefficientimplementationongpus AT xifengyao weightedspatialspectralkernelrxalgorithmandefficientimplementationongpus |