基于图形处理器的实时二维并行窗口傅里叶变换用于条纹图案分析

Real-time 2D parallel windowed Fourier transform for fringe pattern analysis using Graphics Processing Unit.

作者信息

Gao Wenjing, Huyen Nguyen Thi Thanh, Loi Ho Sy, Kemao Qian

机构信息

School of Computer Engineering, Nanyang Technological University, Singapore.

出版信息

Opt Express. 2009 Dec 7;17(25):23147-52. doi: 10.1364/OE.17.023147.

DOI:10.1364/OE.17.023147

PMID:20052242

Abstract

In optical interferometers, fringe projection systems, and synthetic aperture radars, fringe patterns are common outcomes and usually degraded by unavoidable noises. The presence of noises makes the phase extraction and phase unwrapping challenging. Windowed Fourier transform (WFT) based algorithms have been proven to be effective for fringe pattern analysis to various applications. However, the WFT-based algorithms are computationally expensive, prohibiting them from real-time applications. In this paper, we propose a fast parallel WFT-based library using graphics processing units and computer unified device architecture. Real-time WFT-based algorithms are achieved with 4 frames per second in processing 256x256 fringe patterns. Up to 132x speedup is obtained for WFT-based algorithms using NVIDIA GTX295 graphics card than sequential C in quad-core 2.5GHz Intel(R)Xeon(R) CPU E5420.

摘要

在光学干涉仪、条纹投影系统和合成孔径雷达中，条纹图案是常见的结果，并且通常会因不可避免的噪声而退化。噪声的存在使得相位提取和相位解缠具有挑战性。基于窗口傅里叶变换（WFT）的算法已被证明在各种应用的条纹图案分析中是有效的。然而，基于WFT的算法计算成本高昂，这使得它们无法应用于实时场景。在本文中，我们提出了一种使用图形处理单元和计算机统一设备架构的基于WFT的快速并行库。基于实时WFT的算法在处理256x256条纹图案时达到了每秒4帧。与在四核2.5GHz英特尔至强（R）Xeon（R）CPU E5420上运行的顺序C代码相比，使用英伟达GTX295显卡运行基于WFT的算法时，加速比高达132倍。