基于GA-BP神经网络的目标识别方法  

Target recognition method based on GA-BP neural network

作　　者：龙礼[1] 温秀兰[1] 林逸雪 LONG Li;WEN Xiulan;LIN Yixue(Departments of Automation ,University of Nanjing Institute of Technology ,Nanjing 211176, China;School of Mechanical Engineering,Nanjing University of Science and Technology, Nanjing 210094 , China)

机构地区：[1]南京工程学院自动化系,江苏南京211176 [2]南京理工大学机械工程学院,江苏南京210094

出　　处：《传感器与微系统》2019年第10期47-50,共4页Transducer and Microsystem Technologies

基　　金：江苏省高等学校自然科学研究面上资助项目(18KJB510019);国家自然科学基金资助项目(61104085);南京工程学院创新基金重大资助项目(CKJA201704)

摘　　要：针对单一特征识别算法泛化能力差、识别准确率低的问题,本文深入分析了典型目标地震波信号的时域特征,融合多个特征构建训练样本,采用BP神经网络算法对目标进行识别;针对BP神经网络收敛速度慢、泛化能力差的缺点,通过遗传算法对BP神经网络优化。最终形成基于GA-BP神经网络的目标识别方法,并进行实验验证。结果表明:基于GA-BP神经网络算法无论是在收敛性、准确性还是计算速度上都优于传统算法。Aiming at the problem of poor generalization ability and low recognition accuracy of single feature recognition algorithm,the time domain characteristics of typical target seismic wave signals is deeply analyzed,integrates multiple features to construct training samples,and uses BP neural network algorithm to identify targets.BP neural network is optimized by genetic algorithm aiming at shortcomings of slow convergence speed and poor generalization ability.Finally,the target recognition method based on GA-BP neural network is formed and verified by experiments.The results show that the GA-BP neural network algorithm is superior to the traditional algorithm in terms of convergence,accuracy and calculation speed.

关 键 词：无人值守 目标识别 地震波信号 识别算法 

分 类 号：TP212.9[自动化与计算机技术—检测技术与自动化装置]