首页 > 论文 > 光学学报 > 40卷 > 18期(pp:1810002--1)

基于卷积神经网络去噪正则化的条纹图修复

Fringe Pattern Inpainting Based on Convolutional Neural Network Denoising Regularization

  • 摘要
  • 论文信息
  • 参考文献
  • 被引情况
  • PDF全文
分享:

摘要

条纹投影轮廓术测量表面存在高动态范围反射率物体时,采集的条纹图中出现的强度饱和区域将导致对应区域的相位计算误差或缺失,最终影响三维形貌的恢复。为此,提出一种基于卷积神经网络(CNN)去噪正则化的条纹图高光区域修复算法。该方法仅需要在正常曝光和短曝光条件下获取两帧条纹图,快速实现条纹修复,步骤如下:利用Otsu方法对短曝光条纹的调制度图做二值化处理以确定反光区域位置;把短曝光条纹对应区域进行灰度调节后融入正常曝光条纹中,形成迭代修复算法的初值;通过CNN去噪正则化的修复算法,实现条纹图局部高光区域的快速修复,再利用修复后的条纹实现对高动态范围反射物体的三维面形重建。与其他几种常用方法对比,所提方法在条纹修复效果和修复时间上都具有较大优势。

Abstract

Intensity saturation zone in the fringe pattern will appear when fringe projection profilometry is used to measure objects with high dynamic range reflectivity, which will affect the phase reconstruction of the tested object. In this paper, we proposed a fringe pattern inpainting method based on convolutional neural network (CNN) denoising regularization. Two fringe patterns under normal and short exposure time are respectively captured to quickly build a fringe with good quality using following steps. Otsu threshold method is used to determine highlight region by treating the modulation information of short exposure fringe pattern. Set an initial value for iteration by fusing the normal exposure fringe pattern with gray-adjusted short exposure fringe pattern. Realize fast fringe pattern inpainting using CNN denoising regularization and finally obtain a fringe to realize the high dynamic range phase reconstruction. Compared with other methods, the proposed method has advantage in effect and time of fringe inpainting.

广告组1 - 空间光调制器+DMD
补充资料

中图分类号:TP391

DOI:10.3788/AOS202040.1810002

所属栏目:图像处理

基金项目:国家重大仪器设备开发专项;

收稿日期:2020-01-16

修改稿日期:2020-06-03

网络出版日期:2020-09-01

作者单位    点击查看

彭广泽:四川大学电子信息学院光电系, 四川 成都 610065
陈文静:四川大学电子信息学院光电系, 四川 成都 610065

联系人作者:陈文静(chenwj0409@scu.edu.cn)

备注:国家重大仪器设备开发专项;

【1】Su X Y, Zhang Q C, Chen W J. Three-dimensional imaging based on structured illumination [J]. Chinese Journal of Lasers. 2014, 41(2): 0209001.
苏显渝, 张启灿, 陈文静. 结构光三维成像技术 [J]. 中国激光. 2014, 41(2): 0209001.

【2】Chen Y T, Cao Y P, Chen C, et al. Phase measurement profilometry based on binary gratings with unequal duty cycle [J]. Acta Optica Sinica. 2018, 38(8): 0815021.
陈雨婷, 曹益平, 陈澄, 等. 采用不等占空比二元光栅的相位测量轮廓术 [J]. 光学学报. 2018, 38(8): 0815021.

【3】Salahieh B, Chen Z Y, Rodriguez J J, et al. Multi-polarization fringe projection imaging for high dynamic range objects [J]. Optics Express. 2014, 22(8): 10064-10071.

【4】Zhang S, Yau S T. High dynamic range scanning technique [J]. Optics Engineering. 2009, 48(3): 033604.

【5】Li D, Kofman J. Adaptive fringe-pattern projection for image saturation avoidance in 3D surface-shape measurement [J]. Optics Express. 2014, 22(8): 9887-9901.

【6】Zhao H J, Liang X Y, Diao X C, et al. Rapid in situ 3D measurement of shiny object based on fast and high dynamic range digital fringe projector [J]. Optics and Lasers in Engineering. 2014, 54: 170-174.

【7】Budianto B. Lun D P K. Inpainting for fringe projection profilometry based on iterative regularization[C]∥2014 19th International Conference on Digital Signal Processing. 20-23 Aug. 2014, Hong Kong, China. New York: , 2014, 668-672.

【8】Budianto B, Lun D. Inpainting for fringe projection profilometry based on geometrically guided iterative regularization [J]. IEEE Transactions on Image Processing. 2015, 24(12): 5531-5542.

【9】Zhang K, Zuo W M, Gu S H, et al. Learning deep CNN denoiser prior for image restoration[C]∥2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). 21-26 July 2017, Honolulu, HI, USA. New York: , 2017, 2808-2817.

【10】Meinhardt T, Moeller M, Hazirbas C, et al. Learning proximal operators: , 2017, 1799-1808.

【11】Su X Y, Su L K, Li W S, et al. A new 3D profilometry based on modulation measurement [J]. Proceedings of SPIE. 1998, 3558: 1-7.

【12】Su X Y, Chen W J. Fourier transform profilometry: a review [J]. Optics and Lasers in Engineering. 2001, 35(5): 263-284.

【13】Zoran D, Weiss Y. From learning models of natural image patches to whole image restoration[C]∥2011 International Conference on Computer Vision. 6-13 Nov. 2011, Barcelona, Spain. New York: , 2011, 479-486.

【14】Venkatakrishnan S V, Bouman C A, Wohlberg B. Plug-and-play priors for model based reconstruction[C]∥2013 IEEE Global Conference on Signal and Information Processing. 3-5 Dec. 2013, Austin, TX, USA. New York: , 2013, 945-948.

【15】Elad M, Aharon M. Image denoising via sparse and redundant representations over learned dictionaries [J]. IEEE Transactions on Image Processing. 2006, 15(12): 3736-3745.

【16】Dabov K, Foi A, Katkovnik V, et al. Image denoising by sparse 3D transform-domain collaborative filtering [J]. IEEE Transactions on Image Processing: a Publication of the IEEE Signal Processing Society. 2007, 16(8): 2080-2095.

【17】Zhang K, Zuo W M, Chen Y J, et al. Beyond a Gaussian denoiser: residual learning of deep CNN for image denoising [J]. IEEE Transactions on Image Processing: a Publication of the IEEE Signal Processing Society. 2017, 26(7): 3142-3155.

【18】Otsu N. A threshold selection method from gray-levelhistograms [J]. IEEE Transactions on Systems, Man, and Cybernetics. 1979, 9(1): 62-66.

【19】Bertozzi A, Sch?nlieb C B. Unconditionally stable schemes for higher order inpainting [J]. Communications in Mathematical Sciences. 2011, 9(2): 413-457.

【20】Bertalmio M, Sapiro G, Caselles V, et al. Image inpainting[C]∥Proceedings of the 27th Annual Conference on Computer Graphics and Interactive Techniques-SIGGRAPH''''00. New York: , 2000, 417-424.

【21】Vedaldi A, Lenc K. MatConvNet: , 2015, 689-692.

引用该论文

Peng Guangze,Chen Wenjing. Fringe Pattern Inpainting Based on Convolutional Neural Network Denoising Regularization[J]. Acta Optica Sinica, 2020, 40(18): 1810002

彭广泽,陈文静. 基于卷积神经网络去噪正则化的条纹图修复[J]. 光学学报, 2020, 40(18): 1810002

您的浏览器不支持PDF插件,请使用最新的(Chrome/Fire Fox等)浏览器.或者您还可以点击此处下载该论文PDF