Research on night vision monitoring and intelligent identification method of individual rodents
WANG Chi, LI Fudi, SUN Jianmei, LI Siyuan, SHEN Chen
Dept. of Precision Mechanical Engineering, Shanghai University, Shanghai 200444, China
Abstract: 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.
Keywords: computer vision;animal monitoring;low-light night vision;SSD network
2023, 49(9):40-45  收稿日期: 2022-1-20;收到修改稿日期: 2022-3-19
基金项目: 国家自然科学基金（62175144）
作者简介: 王驰（1982-），男，河南周口市人，教授，博士，研究方向为应用光学与融合传感技术。
参考文献
[1] RAHELINIRINA S, SCOBIE K, RAMASIND-RAZANA B, et al. Rodent control to fight plague: field assessment of methods based on rat density reduction[J]. Integrative Zoology, 2021, 16(6): 868-885
[2] PELEH T, LKE K, FRENTZ I, et al. Cross-site reproducibility of social deficits in group-housed BTBR mice using automated longitudinal behavioural monitoring[J]. Neuroscience, 2020, 445: 95-108
[3] ALMEIDA K A D, GARCIA E M, PENTEADO J O, et al. Multimarker approach to assess the exposure of the wild rodent Calomys lauchato a simulated crude oil spill[J]. Environmental Science and Pollution Research, 2020, 28(2): 2236-2244
[4] ROUT T M, BAKER C M, STEWART H, et al. Monitoring, imperfect detection, and risk optimization of a Tasmanian devil insurance population[J]. Conservation Biolofy, 2018, 32(2): 267-275
[5] 李晟, 王大军, 肖治术, 等. 红外相机技术在我国野生动物研究与保护中的应用与前景[J]. 生物多样性, 2014, 22(6): 685-695
[6] 吴厚德, 侯昱辰, 许文海, 等. 水下自主航行器微光照相机驱动系统设计[J]. 光学精密工程, 2018, 26(10): 2605-2613
[7] XUE F D, HE W T, XU F, et al. Hessian single-molecule localization microscopy using sCMOS camera[J]. Biophysics Reports, 2018, 4(4): 215-221
[8] 刘超, 张晓晖. 超低照度下微光图像的深度卷积自编码网络复原[J]. 光学精密工程, 2018, 26(4): 951-961
[9] 荆卫国, 孙明昭, 李永涛. 大动态范围微光夜视系统强光适应性试验方法研究[J]. 红外技术, 2019, 41(7): 689-692
[10] 王驰, 于明坤, 杨辰烨, 等. 抛洒地雷的夜间智能探测方法研究[J]. 中国光学, 2021, 14(5): 1202-1211
[11] CAO J W, SONG C X, PENG S L, et al. Front vehicle detection algorithm for smart car based on improved SSD model[J]. Sensors, 2021, 20(16): 1728-1746
[12] 郭斌, 谢康康, 胡晓峰, 等. 基于机器视觉的四回路保护阀开启压力自动调节系统[J]. 中国测试, 2018, 44(4): 80-85
[13] WEI Y X, LIU X P. Dangerous goods detection based on transfer learning in X-ray images[J]. Neural Computing & Applications, 2020, 32(12): 8711-8724
[14] 陈欣佳, 刘艳霞, 洪晓斌, 等. 基于SSD-CF的无人艇目标检测跟踪方法[J]. 中国测试, 2019, 45(2): 145-150
[15] LI G M, XU Y, ZHAO Y, et al. Evaluating convolutional neural networks for cage-free floor egg detection[J]. Sensors, 2020, 20(3): 332-349