基于OpenCL图像模糊化算法优化与研究.pdfVIP

基于OpenCL 的图像模糊化算法优化研究? 1,2 1 1 张樱 张云泉 龙国平 中国科学院软件研究所并行软件与计算科学实验室 北京100190 1 中国科学院研究生院 北京100190 2 摘 要 现代 GPU 一般都提供特定硬件 如纹理部件、光栅化部件及各种片上缓存 以加速二维图像的处理和 显示过程,相应的编程模型 CUDA、OpenCL 都定义了特定程序设计接口 CUDA的纹理内存,OpenCL 的图像对象 便于图像应用能利用相关硬件支持。以典型图像模糊化处理算法在AMD 平台GPU 的优化为例,探讨OpenCL 的 图像对象在图像算法优化上的适用范围,尤其是其相对于更通用的基于全局内存加片上局部存储进行性能优 化方法的优劣。实验结果表明图像对象只有在图像为四通道且计算过程中需要缓存的数据量较小时能带来较 好的性能改善,其余情况采用全局内存加局部存储能获得更好性能。优化后的算法性能相对于精心实现的CPU 版加速比为200-1000;相对于NVIDIA NPP 库相应函数的性能加速比为 1.3-5。 关键词 AMD GPU，blur，OpenCL，图像对象 中图法分类号 TP302 文献标识码 A Research on image blur algorithm optimization using OpenCL 1,2 1 1 ZHANG Ying , ZHANG Yun-Quan , LONG Guo-Ping Laboratory of Parallel Software and Computational Science, Institute of Software, 1 the Chinese Academy of Sciences, Beijing 100190, China 2 Graduate University, the Chinese Academy of Science s, Beijing 100190, China Abstract Modern GPUs generally provide specific hardware such as texture, grating components and various on-chip cache to accelerate the 2D image processing and displaying process. Programming model defines specific APIs to facilitate image applications taking advantage of image-related GPU hardware, such as CUDA’s texture memory and OpenCL’s Images Object. Take the optimization of image blur algorithm on AMD GPU as an example, in order to get a deep insight into the using of OpenCL’s image object on image applications, especially its advantage and disadvantage compared to the more general optimization method based on global memory and the on-chip local memory. The experimental results demonstrate that the image object can provide better performance only when the processing image is four-channel and the amount of data to be cached is small. For other cases, optimizing with global memory and local memory can get better performance. After optimization, the speedup reaches 200x to 1000x