金属板材近表面缺陷远聚焦像素成像方法研究  

Research on Far-Focused Pixel-Based Imaging Method of Near-Surface Defects in Metal Plate

作　　者：杜文杰 郑慧峰 沈超 段雪亮 曹永刚 DU Wenjie;ZHENG Huifeng;SHEN Chao;DUAN Xueliang;CAO Yonggang(College of Metrology and Measurement Engineering,China Jiliang University,Hangzhou Zhejiang 310018,China;BOE Technology Group Co.,Ltd.,Beijing 100176,China)

机构地区：[1]中国计量大学计量测试工程学院,浙江杭州310018 [2]京东方科技集团股份有限公司,北京100176

出　　处：《传感技术学报》2024年第5期862-868,共7页Chinese Journal of Sensors and Actuators

基　　金：浙江省重点研发计划资助项目(2022C01002);国家市场监管重点实验室(声学振动精密测量技术)开放基金项目(AVL202301);浙江省属高校基本科研业务费专项资金项目(2023YW89,2023YW112);湖南省自然科学基金项目(2023JJ50237)。

摘　　要：针对超声成像方法在金属板材近表面缺陷检测中存在的伪像过多、缺陷定量不准确的问题,开展基于远聚焦像素成像算法的金属板材近表面缺陷成像研究。首先分析了远聚焦算法在近表面缺陷成像中的局限性,然后采用自适应噪声抵消的方法提取近表面缺陷波。对于噪声抵消过程中的参考信号获取问题,结合远聚焦算法的采集特征,提出一种以平均回波信号计算参考信号的方法,信号处理的结果表明,该方法能够正确提取淹没于盲区中的近表面缺陷波。最后,采用优化后的远聚焦算法对铝合金板进行缺陷检测,成像结果表明,自适应噪声抵消后图像的信噪比提升了4.31 dB,缺陷定量的最大误差为0.51 mm^(2),以半峰值宽度为指标对成像分辨率进行测试,成像分辨率可达0.66 mm。Targeting at the problems of excessive artifacts and inaccurate defect quantification in the detection of near-surface defects of metal plate by ultrasonic imaging method,the near-surface defect imaging of metal plate based on far-focused pixel-based imaging algorithm is studied.Firstly,the limitations of far-focused pixel-based imaging algorithm in near-surface defect imaging are analyzed,and then adaptive noise cancellation method is used to extract near-surface defect waves.For the problem of reference signal acquisition in the process of noise cancellation,combining with the acquisition characteristics of far-focused pixel-based imaging algorithm,a method of averaging echo signal to calculate the reference signal is proposed.The signal processing results show that the method can correctly extract the near-surface defect waves submerged in the blind area.Finally,the optimized far-focused pixel-based imaging algorithm is used to detect the defects of the aluminum alloy plate.The imaging results show that the SNR of the image is increased by 4.31 dB after the adaptive noise cancellation,and the maximum error of defect quantification is 0.51 mm^(2).The imaging resolution can reach 0.66 mm when the full-width at half-maximum is taken as the index.

关 键 词：超声成像 金属板材 近表面检测 自适应噪声抵消 成像分辨率 

分 类 号：TB553[理学—物理]