PDF(1629 KB)
超声本征标签识别技术研究
贺升平, 张永龙, 张伟, 刘昱, 贺西平
PDF(1629 KB)
超声本征标签识别技术研究
Study on ultrasonic intrinsic tag identification technique
为保护贵重物项不被篡改替换,提出了识别物项特征信息的新型超声本征标签技术。该技术以超声相控阵系统为回波采集处理平台,以信号频谱为识别的特征量;设计了置信度判定阈值的物项识别分类器,编制了相应的识别程序。实验测试了异质金属、同质金属、陶瓷器件等多类物项。结果表明:在135次识别实验中,准确识别133次,识别正确率为98.52%,误报率为0.74%,漏报率为0.74%;平均单次物项采样、识别时间约为135 s。该技术满足工程需求,具有广阔的应用前景。
To protect valuable items from tampering and replacement, a novel ultrasonic intrinsic tag technology for identifying important item feature information is proposed. The ultrasonic phased array system is established as the echo acquisition and processing platform. An identification algorithm is proposed, and the ratio of the power spectrum of sample signals of two items is calculated as the identification feature. An item identification system is designed based on the confidence determination threshold, and the corresponding identification program is compiled. A prototype is built and tested by use of various types of items such as ceramic items and the metals with the same and different materials. The results show that in 135 times of the identification experiments, 133 times are accurately identified with the accurate rate of identification of 98.52%, the misreport rate of 0.74%, and the underreporting rate of 0.74%. The average time of signal acquisition and identification for an item is respectively about 135 seconds. The proposed ultrasonic intrinsic tag identification technique satisfies the needs of specific occasions and has a broad application prospect.
超声识别 / 超声背向散射 / 相控阵 / 特征提取 {{custom_keyword}} /
ultrasonic identification / ultrasonic backscattering / phased array system / feature extraction {{custom_keyword}} /
图1 两种不锈钢常温(25 ℃)下的显微组织构造图Fig.1 Microstructure of two stainless steels at room temperature(25 ℃) |
表1 A组物项的超声本征标签Tab.1 Ultrasonic intrinsic labels of items of group A |
| 物项 | 阵元1 | 阵元2 | 阵元3 | 阵元4 | 阵元5 | 阵元6 | 阵元7 | 阵元8 |
|---|---|---|---|---|---|---|---|---|
| 阈值 | 20.130 4 | 20.108 3 | 21.081 0 | 19.513 5 | 22.604 9 | 17.726 9 | 18.072 3 | 19.994 0 |
| 1# | 13.272 6 | 13.645 8 | 13.694 4 | 14.865 9 | 14.854 2 | 19.830 4 | 13.634 3 | 17.067 8 |
| 2# | 39.410 7 | 40.429 9 | 44.777 7 | 42.440 3 | 47.767 5 | 43.023 3 | 43.853 8 | 49.630 7 |
| 3# | 98.096 8 | 101.484 5 | 113.141 1 | 115.277 3 | 119.910 5 | 119.395 2 | 113.633 6 | 113.662 6 |
表2 B组物项的超声本征标签Tab.2 Ultrasonic intrinsic label of items of group B |
| 物项 | 阵元1 | 阵元2 | 阵元3 | 阵元4 | 阵元5 | 阵元6 | 阵元7 | 阵元8 |
|---|---|---|---|---|---|---|---|---|
| 阈值 | 26.634 3 | 30.067 0 | 31.440 1 | 34.477 9 | 46.710 8 | 38.175 2 | 56.982 9 | 64.300 4 |
| 4# | 9.756 8 | 11.027 9 | 10.832 3 | 10.494 2 | 21.395 0 | 12.345 6 | 14.670 2 | 29.889 6 |
| 5# | 111.421 8 | 110.375 5 | 114.861 9 | 117.362 8 | 122.798 4 | 120.646 8 | 126.477 0 | 130.494 1 |
| 6# | 125.195 1 | 134.176 4 | 141.380 9 | 141.112 5 | 152.336 1 | 150.024 0 | 153.902 9 | 165.574 5 |
| [1] |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
| [2] |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
| [3] |
贺西平, 刘小荣, 张宏普, 等. 超声背向散射衰减系数谱的金属防伪辨识[J]. 陕西师范大学学报(自然科学版), 2014, 42(6):40-44.
以3种性能相近的金属材料为例，提取10 MHz的高频超声波在材料内部的背向散射信号，平滑处理并截取材料一定深度处相邻两段间的时域信号，傅里叶变换后得到相应的幅度谱和材料的衰减系数谱，与事先留存的标准试样的衰减系数谱进行相关性计算.实验证明， 利用不同材料衰减系数谱的相关系数可以进行材料辨识.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
| [4] |
刘小荣, 贺西平, 张宏普, 等. 金属材料的超声衰减特征及辨识的新方法[J]. 科学通报, 2016, 61(8):844-854.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
| [5] |
刘昱, 贺升平, 贺西平, 等. 对陶瓷器件超声识别的算法设计及虚拟仪器实现[J]. 振动与冲击, 2022, 41(17):254-261.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
| [6] |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
| [7] |
刘昱, 贺升平, 贺西平, 等. 基于超声相控阵平面波检测的金属指纹标签识别系统[J]. 云南大学学报(自然科学版), 2023, 45(1):135-141.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
| [8] |
刘昱, 贺升平, 贺西平. 基于线性预测分析的金属材料超声识别方法[J]. 兵器装备工程学报, 2023, 44(10):307-312.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
| [9] |
卢康, 贺西平, 安笑笑, 等. 基于小波包变换的金属材料超声防伪识别[J]. 无损检测, 2017, 39(7):23-27.
以三种成分相异和三种成分相近的金属材料为试样，提取了10 MHz的高频超声脉冲在其内部传播时的散射信号，通过小波包变换得到散射信号在尺度空间上的能量分布，并将其作为信息防伪识别特征，再采用遗传算法优化后的BP神经网络作为分类器。结果表明，提出的方法可以成功识别成分相异和成分相近的金属材料，相比于成分相近未经热处理的金属材料，同种金属经高温热处理后更容易识别，该方法亦可用于对未知金属的防伪识别。
{{custom_citation.content}}
{{custom_citation.annotation}}
|
| [10] |
安笑笑, 贺西平, 卢康. 基于加权欧氏距离的陶瓷器超声波辨识方法[J]. 电子学报, 2018, 46(7):1737-1741.
为克服陶瓷器鉴定传统方法的局限性，提出一种基于加权欧氏距离的陶瓷器超声波辨识方法.以三个完全相同的方形陶瓷盒子作为实验样品，根据样品的断面扫描电镜照片，分析超声波波长与其内部颗粒尺寸之间的关系；提取10MHz超声波在样品内部的背向散射信号，计算各样品与标准样品的背向散射信号间的加权欧式距离，与事先留存的标准样品自身的加权欧氏距离进行比较.计算结果表明，不同样品之间的加权欧式距离不同.利用背向散射信号的加权欧氏距离，可以快速、准确地辨识陶瓷器样品.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
| [11] |
An ultrasonic backscattering model is developed for textured polycrystalline materials with orthotropic or trigonal grains of ellipsoidal shape. The model allows us to simulate realistic microstructures and orthotropic macroscopic material textures resulting from thermomechanical processing for a broad variety of material symmetries. The 3-D texture is described by a modified Gaussian orientation distribution function (ODF) of the crystallographic orientation of the grains along the macroscopic texture direction. The preferred texture directions are arbitrary relative to the axes of the ellipsoidal grains. The averaged elastic covariance and the directional anisotropy of the backscattering coefficient are obtained for a wave propagation direction arbitrary relative to the texture and grain elongation directions. One particular application of this analysis is the backscattering solution for cubic crystallites with common textures such as Cube, Goss, Brass and Copper. In our analysis, in the texture-defined coordinates the matrix of elastic constants for cubic crystallites takes the form of orthotropic or trigonal symmetry. Numerical results are presented, discussed and compared to the experimental data available in the literature illustrating the dependence of the backscattering coefficient on texture and grain shape. Copyright © 2014 Elsevier B.V. All rights reserved.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
| [12] |
An analytical solution for a three dimensional integral representation of the backscattering (BS) coefficient in polycrystals with elongated (generally ellipsoidal) grains is obtained; it is a natural generalization of the known explicit result for the BS coefficient in polycrystals with spherical grains. New insights into the dependence of the BS signal on frequency and averaged ellipsoidal grain radii are obtained. In particular it has been shown that the dominant factor for the backscattering is the averaged interaction length of the ellipsoidal grain in the direction of wave propagation, instead of the ellipsoidal cross-section. The theory was applied to a simplified model of Ti alloy duplex microstructure and was compared with experiment. For the experimental data analysis directional backscattering ratios are introduced and shown to be advantageous for characterization of duplex elongated microstructures/microtextures. In addition to the geometrical parameters of the elongated microtextures, the BS directional ratios depend on the newly introduced nondimensional material parameter q. The parameter q exhibits the relative contribution of the second phase (crystallites) to the backscattering signal, the effect of which is measurable and important. Comparison of the model with experiment shows there is a significant advantage in using the directional ratios of backscattering coefficients for data analysis.Copyright © 2011 Elsevier B.V. All rights reserved.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
| [13] |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
| [14] |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
| [15] |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
| [16] |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
| [17] |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
| [18] |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
| [19] |
杨辰龙, 冯玮, 边成亮, 等. 基于超声背散射信号递归分析的金属材料微缺陷识别[J]. 光学精密工程, 2019, 27(4):932-944.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
| [20] |
李宇勐. 随机Volterra方程的中心极限定理[J]. 应用概率统计, 2020, 36(2):173-180.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
| [21] |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
| [22] |
卢康, 贺西平, 安笑笑, 等. 基于TSVM的多特征融合超声金属防伪识别[J]. 云南大学学报(自然科学版), 2017, 39(4):584-589.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
| {{custom_ref.label}} |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
PDF(1629 KB)
/
| 〈 |
|
〉 |
