GPU-Based Parallel Design of the Hyperspectral Signal Subspace Identification by Minimum Error (HySime) | |
Wu, Xin1; Huang, Bormin1; Wang, Lizhe1; Zhang, Jianqi1 | |
刊名 | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
2016 | |
卷号 | 9期号:9页码:4400-4406 |
关键词 | LAND-SURFACE PHENOLOGY VEGETATION GREEN-UP TIME-SERIES GLOBAL CHANGE LAST DECADE SNOW COVER AVHRR DATA INDEX MODIS TEMPERATURE |
英文摘要 | Signal subspace identification provides a performance improvement in hyperspectral applications, such as target detection, spectral unmixing, and classification. The HySime method is a well-known unsupervised approach for hyperspectral signal subspace identification. It computes the estimated noise and signal correlation matrices from which a subset of eigenvectors is selected to best represent the signal subspace in the least square sense. Depending on the complexity and dimensionality of the hyperspectral scene, the HySime algorithm may be computationally expensive. In this paper, we propose a massively parallel design of the HySime method for acceleration on NVIDIA's graphics processing units (GPUs). Our pure GPU-based implementation includes the optimal use of the page-locked host memory, block size, and the number of registers per thread. The proposed implementation was validated in terms of accuracy and performance using the NASA AVIRIS hyperspectral data. The benchmark with the NVIDIA GeForce GTX 580 and Tesla K20 GPUs shows significant speedups with regards to the optimized CPU-based serial counterpart. This new fast implementation of the HySime method demonstrates good potential for real-time hyperspectral applications. © 2016 IEEE. |
学科主题 | Engineering; Physical Geography; Remote Sensing; Imaging Science & Photographic Technology |
类目[WOS] | Engineering, Electrical & Electronic ; Geography, Physical ; Remote Sensing ; Imaging Science & Photographic Technology |
收录类别 | SCI ; EI |
语种 | 英语 |
WOS记录号 | WOS:20163102673643 |
内容类型 | 期刊论文 |
源URL | [http://ir.radi.ac.cn/handle/183411/39525] |
专题 | 遥感与数字地球研究所_SCI/EI期刊论文_期刊论文 |
作者单位 | 1. School of Physics and Optoelectronic Engineering, Xidian University, Xi'an 2.710071, China 3. Xidian University, Xi'an 4.710071, China 5. Space Science and Engineering Center, University of Wisconsin, Madison 6.WI 7.53706, United States 8. Institute of Remote Sensing and Digital Earth, Chinese Academy of Sciences, Beijing 9.100094, China |
推荐引用方式 GB/T 7714 | Wu, Xin,Huang, Bormin,Wang, Lizhe,et al. GPU-Based Parallel Design of the Hyperspectral Signal Subspace Identification by Minimum Error (HySime)[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing,2016,9(9):4400-4406. |
APA | Wu, Xin,Huang, Bormin,Wang, Lizhe,&Zhang, Jianqi.(2016).GPU-Based Parallel Design of the Hyperspectral Signal Subspace Identification by Minimum Error (HySime).IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing,9(9),4400-4406. |
MLA | Wu, Xin,et al."GPU-Based Parallel Design of the Hyperspectral Signal Subspace Identification by Minimum Error (HySime)".IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing 9.9(2016):4400-4406. |
个性服务 |
查看访问统计 |
相关权益政策 |
暂无数据 |
收藏/分享 |
除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。
修改评论