基于GPU的图像特征并行计算方法  

Parallel Computation Method of Image Features Based on GPU

作　　者：张杰[1] 柴志雷[1] 喻津 

机构地区：[1]江南大学物联网工程学院,无锡214122

出　　处：《计算机科学》2015年第10期297-300,324,共5页Computer Science

基　　金：国家自然科学基金资助项目:高可靠实时系统的计算平台(SoPC)研究(60703106);模糊支撑函数及其在图像特征表示中的应用(61170121)资助

摘　　要：特征提取与描述是众多计算机视觉应用的基础。局部特征提取与描述因像素级处理产生的高维计算而导致其计算复杂、实时性差,影响了算法在实际系统中的应用。研究了局部特征提取与描述中的关键共性计算模块——图像金字塔机制及图像梯度计算。基于NVIDIA GPU/CUDA架构设计并实现了共性模块的并行计算,并通过优化全局存储、纹理存储及共享存储的访问方式进一步实现了其高效计算。实验结果表明,基于GPU的图像金字塔和图像梯度计算比CPU获得了30倍左右的加速,将实现的图像金字塔和图像梯度计算应用于HOG特征提取与描述算法,相比CPU获得了40倍左右的加速。该研究对于基于GPU实现局部特征的高速提取与描述具有现实意义。Feature extraction and description are the foundation for many computer vision applications. Due to its high dimensional computation of pixel-wise processing, feature extraction and description are computationally intensive with poor real-time performance. Thus it is hard to be used in real-world applications. In this paper, the common computa- tional modules used in feature extraction and description, pyramidal scheme and gradient computation were studied. The method used to compute these modules in parallel based on NVIDIA GPU/CUDA was introduced. Furthermore, compu- tational efficiency was improved by optimizing memory accessing mechanism for global, texture and shared memory. Ex- perimental results show that a 30x speed-up is obtained by GPU-based pyramidal scheme and gradient computation against that of CPU. By employing these GPU-based optimization techniques into HOG （ Histogram of Gradient） imple- mentation based on GPU, it obtains a 40x speed-up against that of CPU. The method proposed in this paper is of signifi- cance for implementing fast feature extraction and description based on GPU.

关 键 词：图像金字塔机制 图像梯度计算 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]