GPU accelerated image processing for electronic speckle pattern shearing interferometry

Heng SHAO; Yong ZHOU; Jun-feng QI; Zhong-yuan NIE

doi:10.3788/YJYXS20193410.1021

您当前的位置：

首页 >

文章列表页 >

GPU accelerated image processing for electronic speckle pattern shearing interferometry

Image Processing | 更新时间：2020-08-12

- GPU accelerated image processing for electronic speckle pattern shearing interferometry
- Chinese Journal of Liquid Crystals and Displays Vol. 34, Issue 10, Pages: 1021-1029(2019)
- 作者机构：
  
  北京卫星制造厂有限公司, 北京 100094
- 作者简介：
- 基金信息：
- DOI：10.3788/YJYXS20193410.1021
  CLC： TP312
- Received：22 April 2019，
  
  Accepted：10 June 2019，
  
  Published：05 October 2019
- 稿件说明：
移动端阅览
Heng SHAO, Yong ZHOU, Jun-feng QI, et al. GPU accelerated image processing for electronic speckle pattern shearing interferometry[J]. Chinese journal of liquid crystals and displays, 2019, 34(10): 1021-1029.
DOI：

Heng SHAO, Yong ZHOU, Jun-feng QI, et al. GPU accelerated image processing for electronic speckle pattern shearing interferometry[J]. Chinese journal of liquid crystals and displays, 2019, 34(10): 1021-1029. DOI： 10.3788/YJYXS20193410.1021.

摘要

电子剪切散斑干涉是在电子散斑干涉基础上发展起来的一种测量位移导数的新方法，尤其适合于无损检测实时测量。由于噪声干扰，测量得到的相位差图有大量噪点需要滤波去除。滤波处理计算量大，采用CPU计算需要很长时间完成。为缩短滤波时间，实现实时滤波处理，对四步相移法和Butterworth低通滤波、正余弦均值滤波和复数均值滤波算法，利用GPU并行计算能力强和存储带宽高的特性，开发了基于GPU加速计算的解相位差和滤波算法。对比基于CPU的解相位差和滤波计算，GPU加速计算将图像总体处理时间约1.9 s（i7 6500U）降低到了239~366 ms（GeForce 940MX）或86~116 ms（Quadro P3000），且正余弦均值滤波和复数均值滤波算法均可以通过降低滤波次数缩短滤波时间，从而保证在性能较弱的GPU上也能将计算时间压缩到0.2 s内，满足实时图像处理的需求。

Abstract

Electronic speckle pattern shearing interferometry (ESPSI) is a new method for measuring displacement derivatives based on electronic speckle pattern interferometry. It is especially suitable for real-time measurement of nondestructive testing. The acquired phase difference map in the ESPSI is usually seriously affected by noise. Thus filtering processing is needed to get a good phase difference map. Filtering by CPU takes too long time for real-time processing. To take full advantage of GPU's strong parallel computing capability and high storage bandwidth

GPU-accelerated phase difference calculation and filtering algorithm based on four-step phase-shift method and Butterworth low-pass filter

sine-cosine mean filter and complex mean filter methods were developed to accelerate image processing. GPU-accelerate processing shortened the processing time from about 1.9 s (i7 6500U) to 239~366 ms (GeForce 940MX) or 86~116 ms (Quadro P3000). The time cost by sine-cosine mean filter and complex mean filter methods can be further reduced to less than 0.2 s by filtering fewer times to meet the needs of real-time ESPSI measurement when the performance of the GPU is weak.

关键词

Keywords

references

RASTOGI P K. Speckle and speckle shearing interferometry—Ⅱ. At the threshold of a new era[J]. Optics and Lasers in Engineering , 1997, 26(4/5): 279-282.

洪友仁, 何浩培, 何小元.剪切散斑:一种光学测量技术及其应用[J].实验力学, 2006, 21(6):667-688.

HUNGY Y, HO H P, HE X Y. Shearography: an optical measurement technique and applications[J]. Journal of Experimental Mechanics , 2006, 21(6): 667-688. (in Chinese)

WANG Y H, THOMAS D, ZHANG P, et al . Whole field strain measurement on complex surfaces by digital speckle pattern interferometry[J]. Materials Evaluation , 2008, 66(5): 507-512.

李峰.电子散斑干涉的图像处理及涡旋光特性研究[D].济南: 山东师范大学, 2015.

LI F. Image processing of electronic speckle pattern interference and optical characteristics research of vortex beam[D]. Ji'nan: Shandong Normal University, 2015. (in Chinese)

宋凡峰, 王开福, 郑蓝.电子散斑相位检测的滤波方法研究[J].激光杂志, 2007, 28(3):50-51.

SONG F F, WANG K F, ZHENG L. Investigation of filtering method for phase measurement in electronic speckle pattern interferometry (ESPI)[J]. Laser Journal , 2008, 28(3): 50-51. (in Chinese)

KEMAO Q, NAM L T H, FENG L, et al . Comparative analysis on some filters for wrapped phase maps[J]. Applied Optics , 2007, 46(30): 7412-7418.

李灿灿.电子散斑干涉图像滤波方法研究及在动态热变形中的应用[D].天津: 天津大学, 2012.

LI C C.Study of electronic speckle pattern interferometry and its application on dynamic thermal deformation[D]. Tianjin: Tianjin University, 2012. (in Chinese)

顾国庆, 王开福, 燕新九.基于同态滤波的电子散斑干涉图像处理[J].激光技术, 2010, 34(6):750-752, 797.

GU G Q, WANG K F, YAN X J. Electronic speckle interferometry image processing based on homomorphic filtering[J]. Laser Technology , 2010, 34(6): 750-752, 797. (in Chinese)

HA S, PARK J, YOU D. A GPU-accelerated semi-implicit fractional-step method for numerical solutions of incompressible Navier-Stokes equations[J]. Journal of Computational Physics , 2018, 352: 246-264.

WINKLER D, MEISTER M, REZAVAND M, et al . gpuSPHASE—A shared memory caching implementation for 2D SPH using CUDA[J]. Computer Physics Communications , 2017, 213: 165-180.

SOLTANI H, LY D Q, AHMED W. Accelerating cell dynamics simulations of soft materials using CUDA-GPU[J]. Materials Today: Proceedings , 2017, 4(1): 75-86.

LIU W G, SCHMIDT B, VOSS G. Accelerating molecular dynamics simulations using graphics processing units with CUDA[J]. Computer Physics Communications , 2008, 179(9): 634-641.

GONG T, FAN T T, GUO J Z, et al . GPU-based parallel optimization of immune convolutional neural network and embedded system[J]. Engineering Applications of Artificial Intelligence , 2017, 62: 384-395.

FONSECA A, CABRAL B. Prototyping a GPGPU neural network for deep-learning big data analysis[J]. Big Data Research , 2017, 8: 50-56.

LITJENS G, KOOI T, BEJNORDI B E, et al . A survey on deep learning in medical image analysis[J]. Medical Image Analysis , 2017, 42: 60-88.

LOCANS U, ADELMANN A, SUTER A, et al . Real-time computation of parameter fitting and image reconstruction using graphical processing units[J]. Computer Physics Communications , 2017, 215: 71-80.

PINHEIRO A, DESTERRO F, SANTOS M C, et al . GPU-based implementation of a diagnostic wind field model used in real-time prediction of atmospheric dispersion of radionuclides[J]. Progress in Nuclear Energy , 2017, 100: 146-163.

MAAS A A M, SOMERS P A A M. Two-dimensional deconvolution applied to phase-stepped shearography[J]. Optics and Lasers in Engineering , 1997, 26(4/5): 351-360.

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Development of AOI inspection of Mura defects on TFT-LCD surface

Application of dual-wavelength fiber guided light based on OpenCV in high-precision assembly and alignment

Design and implementation of piecewise approximation bilateral filtering algorithm based on FPGA

Defect detection of small object solder joints based on improved YOLOv7

Human posture estimation and movement recognition in fitness behavior

Related Author

CHEN Zekang

SHEN Yi

ZHAI Chenyang

DONG Chenyao

WANG Shuangxi

LÜ Guanhui

LIU Xiaomei

ZHANG Zengbao

Related Institution

College of Engineering， Shantou University

Guangdong Provincial Key Laboratory of Automotive Display and Touch Technologies

College of Mechanical and Electrical Engineering， Hainan University

Key Laboratory of Chemical Lasers， Dalian Institute of Chemical Physics， Chinese Academy of Sciences

School of Microelectronics， Shanghai University

AI问答

Address：No.3888 Dong Nanhu Road, Changchun, Jilin, China 130033 Postal code：130033
Tel：0431-86176059 Email：yjxs@ciomp.ac.cn
Technical support is provided by Beijing Founder electronics co., LTD 吉ICP备11002662号-17 京公网安备11010802024621
It is recommended to read the content of this site in Chrome&IE9+. Please switch to extreme mode in browser 360.
Cookies We use cookies to help provide and enhance our service and tailor content. By continuing, you agree to the use of cookies.

⁰