Page 29 - 《应用声学》2022年第4期
P. 29

第 41 卷 第 4 期                                                                       Vol. 41, No. 4
             2022 年 7 月                          Journal of Applied Acoustics                      July, 2022

             ⋄ 研究报告 ⋄



                    超声全聚焦成像中等声程线伪影剔除方法                                                             ∗




                                                  宋寿鹏      †    陈仪倩



                                               (江苏大学机械工程学院       镇江   203022)
                摘要:针对超声全聚焦成像算法中等声程线扩散产生的原理性伪影问题,提出了一种基于超声回波声场有效
                等声程线和图像强度分布特征相结合的伪影剔除方法。首先分析了全聚焦成像算法中等声程线产生伪影的机
                理,根据数据均方根误差自适应辨识有效等声程线;通过 Canny 算子获取全聚焦图像待处理候选区域,再根据
                候选区域中有效等声程线相交次数和图像强度的分布特征辨识缺陷图像和伪影,利用图像中强度最低像素值
                扩展填充剔除伪影后的区域,进而达到剔除伪影的目的。通过在不同直径通孔类和槽类标准缺陷试块上声场
                模拟实验,结果表明伪影剔除效果显著。同时,该方法对噪声也有一定的抑制能力,在不加滤波的条件下信号
                信噪比高于 15 dB 时就可获得较好的去伪影效果。
                关键词:超声;全聚焦成像;伪影;等声程线;缺陷检测
                中图法分类号: TH878           文献标识码: A          文章编号: 1000-310X(2022)04-0527-08
                DOI: 10.11684/j.issn.1000-310X.2022.04.004




                    The method of eliminating isoacoustic contour artifacts in ultrasonic

                                                  total focus method

                                              SONG Shoupeng CHEN Yiqian

                               (School of Mechanical Engineering, Jiangsu University, Zhenjiang 203022, China)

                 Abstract: Aiming at the problem of artifacts caused by the diffusion of isoacoustic contour of ultrasonic total
                 focus method, an artifact elimination method based on the combination of isoacoustic contour of the ultrasonic
                 echo field and the characteristic of image intensity distribution is proposed. Firstly, it analyzes the mechanism
                 of the artefacts caused by the diffusion of isoacoustic contour of ultrasonic total focus method, and adaptively
                 identifies the effective isoacoustic contour according to the root mean square error of the data; obtains the
                 candidate area of the ultrasonic total focus method images by the Canny operator which to be processed, and
                 then identifies defects and artifacts according to the times of effective isoacoustic contours intersections and the
                 characteristic of image intensity distribution in the candidate area, and uses the lowest intensity pixel value in
                 the image to expand and fill the areas which are eliminated artifacts, thereby achieving the purpose of removing
                 artifacts. Through ultrasonic field simulation experiments on different diameters holes and grooves standard
                 defect test blocks, the results show that the artifact eliminated effect is significant. At the same time, this
                 method has a certain ability to suppress noise. When the signal-to-noise ratio is higher than 15 dB under the
                 condition of no filtering, the better artifact eliminated effect can be obtained.
                 Keywords: Ultrasound; Total focus method; Artifacts; Isoacoustic contour; Defect detection


             2021-05-27 收稿; 2021-07-08 定稿
             国家自然科学基金项目 (51375217)
             ∗
             作者简介: 宋寿鹏 (1967– ), 男, 山西人, 博士, 教授, 博士研究生导师, 研究方向: 智能检测与信息处理。
             † 通信作者 E-mail: songshoupeng@126.com
   24   25   26   27   28   29   30   31   32   33   34