鼠类个体夜视监测与智能识别方法研究  

Research on night vision monitoring and intelligent identification method of individual rodents

作　　者：王驰[1] 李富迪 孙建美[1] 李思远 沈晨 WANG Chi;LI Fudi;SUN Jianmei;LI Siyuan;SHEN Chen(Dept.of Precision Mechanical Engineering,Shanghai University,Shanghai 200444,China)

机构地区：[1]上海大学精密机械工程系,上海200444

出　　处：《中国测试》2023年第10期40-45,共6页China Measurement & Test

基　　金：国家自然科学基金(62175144)。

摘　　要：基于机器学习算法,研究一种在夜间环境下鼠类个体身份识别的智能方法及监测系统。根据SSD目标检测算法,设计并优化鼠类个体身份识别模型;搭建微光夜视监测系统,进行微弱光环境下鼠类个体的成像监测实验。实验结果显示,在给定条件下对于不同鼠类个体目标,优化后的鼠类个体智能识别模型平均准确率可达98.14%,平均召回率为98.62%。表明所研究的监测系统和智能识别方法,可用于夜视条件下鼠类个体识别和行为监测方法的进一步研究。Based on machine learning algorithm,an intelligent method and monitoring system for individual identification of rodents in the nighttime environment are studied.According to the SSD target detection algorithm,the rat individual identification model was designed and optimized;the low-light night vision monitoring system was built to carry out the imaging monitoring experiment of the rat individual in the weak light environment.The experimental results show that under the given conditions for different mouse individual targets,the average accuracy rate of the optimized mouse individual intelligent recognition model can reach 98.14%,and the average recall rate is 98.62%.It shows that the researched intelligent identification method and monitoring system can be used for further research on rodent individual identification and behavior monitoring methods under night vision conditions.

关 键 词：机器视觉 动物监测 微光夜视 SSD网络 

分 类 号：TP79[自动化与计算机技术—检测技术与自动化装置] TB9[自动化与计算机技术—控制科学与工程]