| 基于多GPU的Harris角点检测并行算法 |
| |
| 引用本文： | 肖汉, 周清雷, 张祖勋. 基于多GPU的Harris角点检测并行算法[J]. 武汉大学学报 ( 信息科学版), 2012, 37(7): 876-881. |
| |
| 作者姓名： | 肖汉 周清雷 张祖勋 |
| |
| 作者单位： | 1郑州大学信息工程学院,郑州市科学大道100号450001;2郑州师范学院信息科学与技术学院,郑州市英才街6号450044;3武汉大学遥感信息工程学院,武汉市珞喻路129号430079 |
| |
| 基金项目： | 国家自然科学基金资助项目，国家863计划资助项目，中国博士后科学基金资助项目，河南省高等学校青年骨干教师资助项目 |
| |
| 摘 要： | 提出了一种基于多图形处理器(graphic processing unit,GPU)设计思想的Harris角点检测并行算法,使用众多线程将计算中耗时的影像高斯卷积平滑滤波部分改造成单指令多线程(single instruction multi-ple thread,SIMT)模式,并采用GPU中共享存储器、常数存储器和锁页内存机制在统一计算设备架构(com-pute unified device archetecture,CUDA)上完成影像角点检测的全过程。实验结果表明,基于多GPU的Har-ris角点检测并行算法比CPU上的串行算法可获得最高达60倍的加速比,其执行效率明显提高,对于大规模数据处理呈现出良好的实时处理能力。
|
| 关 键 词： | 图形处理器 统一计算设备架构 单指令多线程 角点检测 Harris算子 |
| 收稿时间： | 2012-04-28 |
Parallel Algorithm of Harris Corner Detection Based on Multi-GPU |
| |
| XIAO Han, ZHOU Qinglei, ZHANG Zuxun. Parallel Algorithm of Harris Corner Detection Based on Multi-GPU[J]. Geomatics and Information Science of Wuhan University, 2012, 37(7): 876-881. |
| |
| Authors: | XIAO Han ZHOU Qinglei ZHANG Zuxun |
| |
| Affiliation: | 1 School of Information Engineering,Zhengzhou University,100 Science Road,Zhengzhou 450001,China;2 School of Information Science and Technology,Zhengzhou Normal University, 6 Yingcai Street,Zhengzhou 450044,China;3 School of Remote Sensing and Information Engineering,Wuhan University,129 Luoyu Road,Wuhan 430079,China |
| |
| Abstract: | Parallel algorithm of Harris corner detection based on the core concept of Multi-GPU is proposed,so that time-consuming Gaussian image convolution filtering part during the whole image corner detection process can be implemented by many parallel threads.Finally,implementation of this SIMT parallel algorithm using GPU mechanism of shared memory and constant memory and pinned host memory in CUDA is detailed.The Experiments show that the parallel algorithm of Harris corner detection based on Multi-GPU demonstrates substantial improvement up to 60 times speedup than the serial algorithm running in the CPU,is with high efficiency compared with CPU counterpart algorithm,and show great potential for large-scale data processing in real-time processing. |
| |
| Keywords: | graphic processing unit (GPU) compute unified device architecture (CUDA) single instruction multiple thread(SIMT) corner detection Harris operator |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
| 点击此处可从《武汉大学学报(信息科学版)》浏览原始摘要信息 |
|
点击此处可从《武汉大学学报(信息科学版)》下载全文 |
|
