基于神经网络的电磁感应探雷目标识别方法  

Target Identification Method of Electromagnetic Induction Mine Detection Based on Neural Network

作　　者：王宏伟 杨力 张方雨 WANG Hongwei;YANG Li;ZHANG Fangyu(Field Engineering College,Army University of Engineering,Nanjing 210007,China)

机构地区：[1]陆军工程大学野战工程学院,江苏南京210007

出　　处：《探测与控制学报》2022年第5期90-96,共7页Journal of Detection & Control

基　　金：军队科研基金项目(KYGYSYLY2001)。

摘　　要：针对战后扫雷作业中的地雷探测识别问题,结合时域多时宽双极性脉冲电磁感应探测原理和方法,提出一种基于神经网络的地雷目标识别方法。根据电磁感应探测的输出信号特征,利用神经网络模型,实现对地雷目标和干扰物的区分和识别。试验结果表明,该方法对于低金属含量的杀伤人员地雷和典型干扰物目标,具有较好的识别能力,为地雷探测技术研究和探雷器材研制提供了理论依据和技术支持。Aiming at the problem of mine detection and identification in post-war mine clearance operation,combined with the principle and method of time-domain multi time width bipolar pulse electromagnetic induction detection,a mine target identification method based on neural network was proposed.According to the output signal characteristics of electromagnetic induction detection,the neural network model was used to distinguish and recognize mine targets and interfering objects.The results showed that this method had good identification ability for anti-personnel mines with low metal content and typical interfering objects and the feasibility of the method had been verified.This method also could provide theoretical basis and technical support for the research of mine detection technology and the development of mine detection equipment.

关 键 词：电磁感应探雷 多时宽双极性脉冲 BP神经网络 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]