基于阵列传感器的金属分类及成像  

Metal Classification and Imaging Based on Array Sensors

作　　者：张裕宸 张志杰[1] 陈昊泽 王栋[1] 刘孝天 ZHANG Yuchen;ZHANG Zhijie;CHEN Haoze;WANG Dong;LIU Xiaotian(Key Laboratory of Instrumentation Science&Dynamic Measurement,Ministry of Education,North University of China,Taiyuan 030051,China)

机构地区：[1]中北大学仪器科学与动态测试教育部重点实验室,山西太原030051

出　　处：《测试技术学报》2023年第4期330-335,341,共7页Journal of Test and Measurement Technology

摘　　要：通过使用一种基于实测矩阵最值的灵敏度矩阵的修正方法,利用实测矩阵的最值对仿真所得的灵敏度进行修正,并将这3种矩阵分别进行电磁层析成像。其中,对2个磁导率不同的物体成像时,使用修正灵敏度矩阵所成的图像对于2种样品所处位置的中心点的灰度值不同,可以从图中明显看出2个样品的差异。在此基础上,通过对5种不同金属样品所成图像的最值进行分析,可以有效地分辨出不同金属样品。实验最终验证了修正灵敏度矩阵所成图像更加接近于实际电磁参数的分布,对于磁导率不同的物体更加有辨识性,可以有效对部分金属进行分类。In this paper,we use a method to modify the sensitivity matrix based on the maximum of the measured matrix,and use the maximum of the measured matrix to modify the sensitivity obtained by simulation,and use these three matrices to conduct electromagnetic tomography,respectively.Among them,when imaging two objects with different permeability,the gray value of the image formed by using the modified sensitivity matrix is different for the center point of the two samples,and the difference between the two samples can be clearly seen from the figure.On this basis,different metal samples can be effectively distinguished by analyzing the maximum values of images formed by several different metal samples.The experiment finally verifies that the image generated by the modified sensitivity matrix is closer to the distribution of actual electromagnetic parameters,and is more recognizable for objects with different permeability,which can effectively classify some metals.

关 键 词：电磁层析成像技术 总体最小二乘法 金属分类 图像重建 修正矩阵 

分 类 号：TP23[自动化与计算机技术—检测技术与自动化装置]