作者：钟浩¹, 刘桂雄¹, 陈冬雪¹, 彭艳华¹, 曾启林²

作者单位：1． 华南理工大学机械与汽车工程学院, 广东 广州 510640;
2. 广州华工百川科技有限公司, 广东 广州 510640

关键词：轮胎;剪切散斑干涉;相位图;局域;多方向;频域滤波

摘要：

为解决激光数字剪切散斑干涉技术应用于轮胎缺陷检测中,轮胎缺陷相位条纹图中噪声对包裹相位图解包和轮胎缺陷尺寸测量产生严重影响的问题,该文通过对散斑干涉相位图的条纹特征分析,研究一种根据条纹密度将相位图划分成不同滤波区域进行多方向频域滤波的方法。实验结果表明:在两组模拟的相位图滤波实验中,局域多方向滤波方法的相位误差均值Pmean和相位均方根误差RMS都比正余弦滤波方法、多方向频域滤波方法要小,滤波结果更接近真值;在第3组滤波实验中,局域多方向频域滤波方法的残差点数分别是正余弦滤波方法、多方向频域滤波方法的11.39%、56.25%,体现出较好的滤波特性。

Local multidirectional frequency domain filtering for tire phase fringe patterns of laser speckle pattern interferometry

ZHONG Hao¹, LIU Guixiong¹, CHEN Dongxue¹, PENG Yanhua¹, ZENG Qilin²

1. School of Mechanical and Automotive Engineering, South China University of Technology, Guangzhou 510640, China;
2. Guangzhou SCUT Bestry Technology Co., Ltd., Guangzhou 510640, China

Abstract: Noises in the phase fringe pattern of a laser digital shearing speckle interferometer disturb significantly the phase unwrapping accuracy and defect size measurement of tires. In this paper, a method dividing the phase pattern into different filtering zones for multidirectional frequency domain filtering was introduced according to the fringe density. The method could filter out most of the noise and better preserve the edge information. Experimental results show that the mean phase error P_mean and the root mean square(RMS) of the local multidirectional low-pass frequency domain filtering are both smaller than these of the sin/cos method and the multidirectional low-pass frequency domain filtering, and much closer to the truth value. In the third set of filtering experiment, the residues of the local method were 11.39% and 56.25% respectively the residues of the other two methods.

Keywords: tire;shear speckle interferometer;phase patterns;local;multi direction;frequency domain filter

2016, 42(6): 100-103  收稿日期: 2015-09-21;收到修改稿日期: 2015-10-23

基金项目: 粤港共性技术招标项目(2013B010134008)

作者简介: 钟浩(1992-),男,江西赣州市人,硕士研究生,专业方向为先进传感器技术与应用。

参考文献

[1] PILEVAR A H, SAIEN S. A new filter to remove salt and pepper noise in color images[J]. Signal Image and Video Processing,2015,9(4):779-786.
[2] 王永红,李骏睿,等. 散斑干涉相位条纹图的频域滤波处理[J]. 中国光学,2014,7(3):389-295.
[3] 陈冬雪,刘桂雄,等. 轮胎激光散斑干涉相位条纹图多方向频域滤波方法[J]. 中国测试,2015,41(11):88-92.
[4] 王文平. 电子散斑干涉条纹图和相位图滤波方法的应用研究[D]. 天津:天津大学,2007.
[5] 蔡长青. 散斑干涉计量关键问题研究及其应用[D]. 广州:华南理工大学,2013.
[6] SONG R, GUO H D. Improved goldstein SAR interferogram filter basedon adaptive-neighborhood technique[J]. IEEE Geoscience and Remote Sensing Letters January,2015:140-144.
[7] SUN Q, LI Z W. Improved Goldstein filter for InSAR noise reduction based on local SNR[J]. IEEE Geoscience and Remote Sensing Letters,2013(20):1896-1903.
[8] LIN X, LI F F. Nonlocal SAR interferometric phase filtering through higher order singular value decomposition[J]. IEEE Geoscience and Remote Sensing Letters April,2015:806-810.