期刊
IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING
卷 10, 期 10, 页码 4387-4398出版社
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/JSTARS.2017.2713126
关键词
Classification; multicore and many-core architectures; parallel optimization; performance analysis; support vector machine (SVM)
类别
资金
- National Natural Science Foundation of China [61303003, 41374113]
- National High-Tech R&D (863) Program of China [2013AA01A208]
- Tsinghua University Initiative Scientific Research Program [20131089356]
- National Key Research and Development Plan of China [2016YFA0602200]
Support Vector Machine (SVM) is a classification method that has been widely used in the domain of remote sensing for decades. Although SVM-based classification method achieves good performance for classification accuracy in many studies, it can become very time-consuming in some remote sensing applications such as hyperspectral image classification or large-scale land cover mapping. To improve the efficiency for SVM training and classification in remote sensing applications, we designed and implemented a highly efficient multiclass support vector machine (MMSVM) for x 86-based multicore and many-core architectures such as the Ivy Bridge CPUs and the Intel Xeon Phi coprocessor (MIC) based on our previous MIC-SVM library. Various analysis methods and optimization strategies are employed to fully utilize the multilevel parallelism of our studied architectures. We select several real-world remote sensing datasets to evaluate the performance of our proposed MMSVM. Compared with the widely used serial LIBSVM, our MMSVM achieves 6.3-31.1 x (in training) and 4.9-32.2 x(in classification) speedups on MIC, and 6.9-14.9x (in training) and 5.5-22.1x (in classification) speedups on the Ivy Bridge CPUs. We also conduct a performance comparison analysis on different platforms and provide some ideas on how to select the most suitable architecture for specific re-mote sensing classification problems in order to achieve the best performance.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据