基于几何纹理与Anscombe变换的蜂窝材料太赫兹图像降噪模型

doi:10.3901/JME.2021.22.096

机械工程学报 ›› 2021, Vol. 57 ›› Issue (22): 96-105.doi: 10.3901/JME.2021.22.096

扫码分享

基于几何纹理与Anscombe变换的蜂窝材料太赫兹图像降噪模型

孙凤山¹, 范孟豹¹, 曹丙花², 叶波^3,4, 刘林⁵

1. 中国矿业大学机电工程学院徐州 221116;
2. 中国矿业大学信息与控制工程学院徐州 221116;
3. 昆明理工大学信息工程与自动化学院昆明 650500;
4. 昆明理工大学云南省人工智能重点实验室昆明 650500;
5. 北京航天计量测试技术研究所北京 100076

收稿日期:2020-11-20 修回日期:2021-07-13 出版日期:2021-11-20 发布日期:2022-02-28
通讯作者: 范孟豹(通信作者),男,1981年出生,博士,教授,博士研究生导师。主要研究方向为涡流/太赫兹无损检测理论及应用。E-mail:wuzhi3495@cumt.edu.cn
作者简介:孙凤山,男,1994年出生,博士研究生。主要研究方向为太赫兹无损检测成像方法与图像处理。E-mail:TB20050013B4@cumt.edu.cn;叶波,男,1978年出生,博士,教授,博士研究生导师。主要研究方向为结构健康监测、电磁无损检测与评估、信号处理和反演问题、成像及图像处理、智能电网。E-mail:yeripple@hotmail.com
基金资助:
国家自然科学基金（62071471）和江苏高校优势学科建设工程资助项目。

Denoising Model of Terahertz Imaging of Honeycomb Material Based on Geometric Texture and Anscombe Transformation

SUN Fengshan¹, FAN Mengbao¹, CAO Binghua², YE Bo^3,4, LIU Lin⁵

1. School of Mechatronic Engineering, China University of Mining and Technology, Xuzhou 221116;
2. School of Information and Control Engineering, China University of Mining and Technology, Xuzhou 221116;
3. Faculty of Information Engineering and Automation, Kunming University of Science and Technology, Kunming 650500;
4. Yunnan Key Laboratory of Artificial Intelligence, Kunming University of Science and Technology, Kunming 650500;
5. Beijing Aerospace Institute for Metrology and Measurement Technology, Beijing 100076

Received:2020-11-20 Revised:2021-07-13 Online:2021-11-20 Published:2022-02-28

摘要/Abstract

摘要： 为解决太赫兹（Terahertz，THz）图像内泊松高斯混合噪声导致芳纶纤维蜂窝材料脱粘缺陷轮廓检测精度低的问题，基于Anscombe变换与小波阈值法构建了THz图像降噪模型。高斯噪声方差为降噪模型的必要参数，但实际THz图像噪声分布未知，且噪声与纹理在高频混叠，给方差准确估计提出了挑战。为此，首先以样件纹理几何形状为先验信息，构造Benzene-ring算子去除THz图像纹理，使其小波域高频分量中仅含有噪声；然后提出改进的Logistic混沌映射提高样本集的多样性，以训练Elman神经网络准确建立高频分量与高斯噪声方差间映射关系；最后依据噪声方差估计值，基于Anscombe变换将泊松高斯混合噪声转化为高斯噪声，并利用小波阈值法与Anscombe逆变换得到了最终THz降噪图像。仿真与试验结果表明，所提出的方法降噪效果最佳并有效提高缺陷轮廓检测精度，相比于高斯滤波、小波阈值以及非局部均值法，平均梯度指标分别提升12%、33%、9%，缺陷面积绝对误差分别降低234 mm²、304 mm²、263 mm²。

关键词: 太赫兹无损检测, 泊松高斯混合噪声, Benzene-ring算子, 噪声估计, 图像降噪

Abstract: In order to solve the problem of low accuracy caused by mixed Poisson-Gaussian noise in Terahertz (THz) images of aramid fiber honeycomb material in contour detection of defects, a THz image denoising model is constructed based on the Anscombe transformation and wavelet threshold method. The variance of Gaussian noise is a necessary parameter of denoising model, but the noise distribution of THz images is unknown. Meanwhile, the texture and noise are mixed in high frequency, which gives challenges to accurate variance estimation. Firstly, taking the texture geometry of the sample as the prior information, the Benzene-ring operator is constructed to remove the texture of the THz image, so that the high-frequency components only contain noise. Secondly, an improved Logistic chaotic mapping is proposed to improve the diversity of dataset to train Elman neural network for building the mapping relationship between the high-frequency component and the variance of Gaussian noise. Finally, the Anscombe transformation is performed to transform the mixed Poisson-Gaussian noise to gaussian noise. The THz denoised image is obtained by using wavelet threshold method and Anscombe inverse transformation. The experiment and simulation results show that the proposed model can eliminate the mixed Poisson-Gaussian noise better than other three methods and improves the accuracy of defect contour detections. Compared with the Gaussian filtering, wavelet threshold and non-local mean value methods, the average gradient index is increased by 12%, 33%, and 9%, and the absolute error of defect area is reduced by 234 mm², 304 mm², and 263 mm², respectively.

Key words: Terahertz nondestructive testing, mixed Poisson-Gaussian noise, Benzene-ring operator, noise estimation, image denoising

中图分类号:

TH744
TP391

孙凤山, 范孟豹, 曹丙花, 叶波, 刘林. 基于几何纹理与Anscombe变换的蜂窝材料太赫兹图像降噪模型[J]. 机械工程学报, 2021, 57(22): 96-105.

SUN Fengshan, FAN Mengbao, CAO Binghua, YE Bo, LIU Lin. Denoising Model of Terahertz Imaging of Honeycomb Material Based on Geometric Texture and Anscombe Transformation[J]. Journal of Mechanical Engineering, 2021, 57(22): 96-105.

参考文献

[1] YIN Liangpeng, ZHOU Zongtao, ZHU Luo, et al. Reinforcing effect of aramid fibers on fatigue behavior of SBENZENE-RING/aramid fiber composites[J]. Polymer Testing, 2019, 80(4):92-99.
[2] 刘宇峰, 张中伟, 许正辉, 等. 空间高稳定碳/碳蜂窝夹层结构制备及性能[J]. 宇航学报, 2020, 41(8):1067-1075. LIU Yufeng, ZHANG Zhongwei, XU Zhenghui, et al. Preparation and properties of carbon/carbon honeycomb sandwich panels for space ultra stable structures[J]. Journal of Astronautics, 2020, 41(8):1067-1075.
[3] LI Hongfu, XU Yue, ZHANG Tao, et al. Interfacial adhesion and shear behaviors of aramid fiber/polyamide 6composites under different thermal treatments[J]. Polymer Testing, 2020, 81(5):1-10.
[4] 邹田春, 秦嘉徐, 李龙辉, 等. 钛合金-芳纶纤维复合材料单搭接接头渐进失效分析[J]. 材料导报, 2020, 34(20):20143-20146. ZOU Tianchun, QIN Jiaxu, LI Longhui, et al. Progressive failure analysis of Titanium alloy-aramid fiber composites single lap joints[J]. Meterial Reports, 2020, 34(20):20143-20146.
[5] 沈功田. 承压设备无损检测与评价技术发展现状[J]. 机械工程学报, 2017, 53(12):1-12. SHEN Gongtian. Development status of nondestructive testing and evaluation technique for pressure equipment[J]. Journal of Mechanical Engineering, 2017, 53(12):1-12.
[6] CAO Binghua, WANG Mengyun, LI Xiaohan, et al. Accurate thickness measurement of multilayer coatings on metallic substrate using pulsed terahertz technology[J]. Journal of Mechanical Science and Technology, 2007, 21(6):946-949.
[7] FAN Mengbao, CAO Binghua, TIAN Guiyun. Enhanced measurement of paper basis weight using phase shift in terahertz time-domain spectroscopy[J]. Journal of Sensors, 2017, 10(1):1-14.
[8] YE Dongdong, WANG Weize, Huang Jibo, et al. Nondestructive interface morphology characterization of thermal barrier coatings using terahertz time-domain spectroscopy[J]. Coatings, 2019, 9(2):80-89.
[9] 王月, 李琦. 基于噪声估计的太赫兹图像自适应迭代去噪[J]. 中国激光, 2020, 47(8):111-121. WANG Yue, LI Qi. Adaptive iterative denoising of Terahertz image based on noise estimation[J]. Chinese Journal of Lasers, 2020, 47(8):111-121.
[10] MA Wenshu, LI Qi, LU Jiaye. De-noising research on terahertz holographic reconstructed image based on weighted nuclear norm minimization method[J]. Frontiers of Optoelectronics, 2018, 11(3):267-274.
[11] CHEN Guanghao, LI Qi. Markov chain Monte Carlo sampling based Terahertz holography image denoising[J]. Applied Optics, 2015, 54(14):4345-4351.
[12] 崔珊珊, 李琦. 基于小波变换的太赫兹数字全息再现图像去噪研究[J]. 红外与激光工程, 2015, 44(6):1836-1840. CUI Shanshan, LI Qi. De-noising research on Terahertz digital holography based on wavelet transform[J]. Infrared and Laser Engineering, 2015, 44(6):1836-1840.
[13] MAKITALO M, FOI A. Optimal inversion of the generalized Anscombe transformation for Poisson-Gaussian noise[J]. IEEE Transactions on Image Processing, 2013, 22(1):91-10.
[14] SHEN Yan, CHEN Ying, LIU Qing, et al. Improved Anscombe transformation and total variation for denoising of lowlight infrared images[J]. Infrared Physics and Technology, 2018, 93(9):192-198.
[15] 沈帆, 李翰林, 孙斌, 等. 基于Anscombe变换的X射线图像序列盲源分离降噪[J]. 光学精密工程, 2020, 28(1):244-250. SHEN Fan, LI Hanlin, SUN Bin, et al. X-ray image denoising using blind source separation in Anscombe domain[J]. Optics and Precision Engineering, 2020, 28(1):244-250.
[16] LIU Xinhao, TANAKA M, OKUTOMI M. Single-image noise level estimation for blind denoising[J]. IEEETransactions on Image Processing, 2013, 22(12):5226-5237.
[17] JIANG Ping, ZHANG Jianzhou. Fast and reliable noise level estimation based on local statistic[J]. Pattern Recognition Letters, 2016, 78(3):8-13.
[18] DONOHO D L. De-noising by soft-thresholding[J]. IEEETransactions on Information Theory, 1995, 41(3):613-627.
[19] 张承志, 冯华君, 徐之海, 等. 图像噪声方差分段估计法[J]. 浙江大学学报, 2018, 52(9):1804-1810. ZHANG Chengzhi, FENG Huajun, XU Zhihai, et al. Piecewise noise variance estimation of images based on wavelet transform[J]. Journal of Zhejiang University, 2018, 52(9):1804-1810.
[20] LUCA C, GIUSEPPE V, ANTONIO C, et al. A cellular automata based approach to track salient objects in videos[J]. Natural Computing, 2019, 18(4):341-352.
[21] LIANG Yudong, WANG Jinjun, ZHOU Sanping, et al. Incorporating image priors with deep convolutional neural networks for image super-resolution[J]. Neurocomputing, 2016, 94(6):340-347.
[22] ZHOU Rigui, LIU Daqian. Image edge extraction based on improved sobel operator[J]. International Journal of Theoretical Physics, 2019, 58(9):11-17.
[23] GAO Yaohui, CHEN Mengkun, YANG Chaien, et al. Image edge extraction based on improved sobel operator[J]. Journal of Applied Physics, 2019, 58(9):11-17.
[24] WANG Yingmin, ZHANG Fujun, CUI Tao, et al. Fault diagnosis for manifold absolute pressure sensor (map) of diesel engine based on elman neural network observer[J]. Chinese Journal of Mechanical Engineering, 2016, 29(2):386-395.
[25] 姜兴宇, 王世杰, 赵凯, 等. 面向网络化制造的智能工序质量控制系统[J]. 机械工程学报, 2010, 46(4):186-194. JIANG Xingyu, WANG Shijie, ZHAO Kai, et al. Intelligent process quality control system for networked manufacturing[J]. Journal of Mechanical Engineering, 2010, 46(4):186-194.
[26] TSAI H C. Confined teaching-learning-based optimization with variable search strategies for continuous optimization[J]. Information Sciences, 2019, 500(10):34-47.
[27] 李鹏越, 田建东, 王国霖, 等. 面向机器人环境共融的图像去雪算法[J]. 机械工程学报, 2019, 55(11):98-104. LI Pengyue, TIAN Jiandong, WANG Guolin, et al. Image snow removal methods for robotic environment fusion[J]. Journal of Mechanical Engineering, 2019, 55(11):98-104.
[28] GRANGETTO M, MAGLI E, OLMO G. Ensuring quality of service for image transmission:hybrid loss protection[J]. IEEE Transactions on Image Processing, 2004, 13(6):751-757.

基于几何纹理与Anscombe变换的蜂窝材料太赫兹图像降噪模型

Denoising Model of Terahertz Imaging of Honeycomb Material Based on Geometric Texture and Anscombe Transformation

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 3

编辑推荐

Metrics

本文评价

[1]	孙梦楠, 董祉序, 徐威, 孙兴伟, 刘伟军. 基于改进低秩稀疏分解的光斑图像降噪方法研究[J]. 机械工程学报, 2024, 60(2): 17-26.
[2]	朵文博, 李宏伟, 李帅兵, 杨栋, 卢保朋, 康永强, 曹炳磊. 基于太赫兹成像的层状复合绝缘结构内部分层缺陷SSA-CNN定量表征^*[J]. 电气工程学报, 2023, 18(3): 63-72.
[3]	孙凤山, 范孟豹, 曹丙花, 刘林. 基于时频关键信息融合的热障涂层太赫兹准确测厚方法[J]. 机械工程学报, 2023, 59(14): 10-22.