Amur tiger stripes:individual identification based on deep convolutional neural network

作     者：Chunmei SHI Dan LIU Yonglu CUI Jiajun XIE Nathan James ROBERTS Guangshun JIANG 

作者机构：Department of MathematicsSchool of ScienceNortheast Forestry UniversityHarbinChina Feline Research CenterNational Forestry and Grassland AdministrationCollege of Wildlife and Protected AreasNortheast Forestry UniversityHarbinChina Siberian Tiger ParkHarbinHeilongjiangChina 

出 版 物：《Integrative Zoology》 (整合动物学（英文版）)

年 卷 期：2020年第15卷第6期

页      面：461-470页

核心收录：

学科分类：0710[理学-生物学] 12[管理学] 1201[管理学-管理科学与工程(可授管理学、工学学位)] 07[理学] 0905[农学-畜牧学] 081104[工学-模式识别与智能系统] 08[工学] 0906[农学-兽医学] 0835[工学-软件工程] 071002[理学-动物学] 0811[工学-控制科学与工程] 0812[工学-计算机科学与技术（可授工学、理学学位）] 

基　　金：the Fundamental Research Funds for the Central Universities(2572018BC07,2572017PZ14) the Heilongjiang postdoctoral project fund project(LBH-Z18003) Biodiversity Survey,Monitoring and Assessment Project of Ministry of Ecology and Environment,China(2019HB2096001006) the National Natural Science Foundation of China(NSFC 31872241,31572285) the Individual Identification Technological Research on Camera-trapping images of Amur tigers(NFGA 2017) 

主　　题：Amur tiger deep convolutional neural network individual identification stripe feature 

摘      要：The automatic individual identification of Amur tigers(Panthera tigris altaica)is important for population monitoring and making effective conservation *** existing research primarily relies on manual identifi-cation,which does not scale well to large *** this paper,the deep convolution neural networks algorithm is constructed to implement the automatic individual identification for large numbers of Amur tiger *** experimental data were obtained from 40 Amur tigers in Tieling Guaipo Tiger Park,*** number of images collected from each tiger was approximately 200,and a total of 8277 images were *** experiments were carried out on both the left and right side of *** results suggested that the recognition accuracy rate of left and right sides are 90.48%and 93.5%,*** accuracy of our network has achieved the similar level compared to other state of the art networks like LeNet,ResNet34,and ZF_*** running time is much shorter than that of other ***,this study can provide a new approach on automatic individual identification technology in the case of the Amur tiger.