基于深度学习的剪切散斑干涉条纹图滤波方法  

Phase Fringe Pattern Filtering Method for Shearography Using Deep Learning

作　　者：林薇 崔海华[1] 郑炜 周新房 徐振龙 田威[1] Lin Wei;Cui Haihua;Zheng Wei;Zhou Xinfang;Xu Zhenlong;Tian Wei(College of Mechanical&Electrical Engineering,Nanjing University of Aeronautics and Astronautics,Nanjing 211106,Jiangsu,China;AVIC Xi’an Aircraft Industry Group Co.,Ltd.,Xi’an 710089,Shaanxi,China)

机构地区：[1]南京航空航天大学机电学院,江苏南京211106 [2]中航西安飞机工业集团股份有限公司,陕西西安710089

出　　处：《激光与光电子学进展》2022年第22期147-156,共10页Laser & Optoelectronics Progress

基　　金：国家重点研发计划(2019YFB2006100,2019YFB1707501);江苏省自然科学基金(BK20191280);南京航空航天大学研究生开放基金(kfjj20200520)。

摘　　要：剪切散斑干涉技术作为一种非接触式的高精度光学全场测量方法,可以对复合材料内部缺陷进行无损检测,但所得的相位条纹图中包含大量散斑噪声,会对检测结果和精度产生严重影响。为此,提出了一种基于无监督图像风格转换模型(CycleGAN)的相位条纹图滤波方法。该方法将剪切散斑干涉技术获取的原始噪声相位条纹图通过网络训练转换为理想无噪声条纹图,从而实现对相位条纹图中噪声的滤除。实验结果表明,所提方法能够实现对噪声的高效滤除,滤波图像边界清晰、对比显著,且运行时间明显优于其他方法,仅需30 ms左右便能实现条纹图的高质量滤波,符合动态无损检测的发展需求,为相位条纹图的噪声滤除提供了新的思路。As a noncontact high-precision optical full-field measurement method,shearography can be used for the nondestructive detection of internal defects in composite materials.However,the obtained phase fringe pattern contains a high amount of speckle noise that seriously affects the detection results and accuracy.Therefore,we propose a phase fringe-filtering method using an unsupervised image style conversion model(CycleGAN).Furthermore,the original noise phase fringe image obtained using shearography is converted into an ideal noiseless fringe image via network training to achieve noise filtering in the phase fringe pattern.The experimental results show that the proposed method achieves highefficiency filtering for noise in areas where the stripe distribution is relatively sparse,with clear boundaries and significant contrast in filtered images.Additionally,the running time of the proposed method is better than that of the other methods(by approximately 30 ms),achieves high-quality filtering,meets the development demand of dynamic nondestructive testing,and provides a new idea for the noise filtering of phase fringe pattern.

关 键 词：剪切散斑干涉技术 相位图 深度学习 噪声 图像处理 无损检测 

分 类 号：O439[机械工程—光学工程]