Automated Video-Based Analysis Framework for Behavior Monitoring of Individual Animals in Zoos Using Deep Learning-A Study on Polar Bears

被引:19
作者
Zuerl, Matthias [1 ]
Stoll, Philip [1 ]
Brehm, Ingrid [2 ]
Raab, Rene [1 ]
Zanca, Dario [1 ]
Kabri, Samira [1 ]
Happold, Johanna [1 ]
Nille, Heiko [1 ]
Prechtel, Katharina [2 ]
Wuensch, Sophie [2 ]
Krause, Marie [2 ]
Seegerer, Stefan [3 ]
von Fersen, Lorenzo [4 ]
Eskofier, Bjoern [1 ]
机构
[1] Friedrich Alexander Univ Erlangen Nurnberg, Dept Artificial Intelligence Biomed Engn, Machine Learning & Data Analyt Lab, D-91052 Erlangen, Germany
[2] Friedrich Alexander Univ Erlangen Nurnberg, Dept Biol, Anim Physiol, D-91058 Erlangen, Germany
[3] Free Univ Berlin, Dept Math & Comp Sci, Comp Educ Res Grp, D-14195 Berlin, Germany
[4] Nuremberg Zoo, D-90480 Nurnberg, Germany
来源
ANIMALS | 2022年 / 12卷 / 06期
关键词
animal welfare; animal behavior; deep learning; object detection; animal monitoring; behavior observation; Ursus maritimus; URSUS-MARITIMUS; WELFARE; IDENTIFICATION; LOCOMOTION;
D O I
10.3390/ani12060692
中图分类号
S8 [畜牧、 动物医学、狩猎、蚕、蜂];
学科分类号
0905 ;
摘要
The monitoring of animals under human care is a crucial tool for biologists and zookeepers to keep track of the animals' physical and psychological health. Additionally, it enables the analysis of observed behavioral changes and helps to unravel underlying reasons. Enhancing our understanding of animals ensures and improves ex situ animal welfare as well as in situ conservation. However, traditional observation methods are time- and labor-intensive, as they require experts to observe the animals on-site during long and repeated sessions and manually score their behavior. Therefore, the development of automated observation systems would greatly benefit researchers and practitioners in this domain. We propose an automated framework for basic behavior monitoring of individual animals under human care. Raw video data are processed to continuously determine the position of the individuals within the enclosure. The trajectories describing their travel patterns are presented, along with fundamental analysis, through a graphical user interface (GUI). We evaluate the performance of the framework on captive polar bears (Ursus maritimus). We show that the framework can localize and identify individual polar bears with an F1 score of 86.4%. The localization accuracy of the framework is 19.9 +/- 7.6 cm, outperforming current manual observation methods. Furthermore, we provide a bounding-box-labeled dataset of the two polar bears housed in Nuremberg Zoo.
引用
收藏
页数:16
相关论文
共 37 条
[21]   DeepLabCut: markerless pose estimation of user-defined body parts with deep learning [J].
Mathis, Alexander ;
Mamidanna, Pranav ;
Cury, Kevin M. ;
Abe, Taiga ;
Murthy, Venkatesh N. ;
Mathis, Mackenzie Weygandt ;
Bethge, Matthias .
NATURE NEUROSCIENCE, 2018, 21 (09) :1281-+
[22]   A Markerless 3D Computerized Motion Capture System Incorporating a Skeleton Model for Monkeys [J].
Nakamura, Tomoya ;
Matsumoto, Jumpei ;
Nishimaru, Hiroshi ;
Bretas, Rafael Vieira ;
Takamura, Yusaku ;
Hori, Etsuro ;
Ono, Taketoshi ;
Nishijo, Hisao .
PLOS ONE, 2016, 11 (11)
[23]   ChickTrack-A quantitative tracking tool for measuring chicken activity [J].
Neethirajan, Suresh .
MEASUREMENT, 2022, 191
[24]   Development of automatic surveillance of animal behaviour and welfare using image analysis and machine learned segmentation technique [J].
Nilsson, M. ;
Herlin, A. H. ;
Ardo, H. ;
Guzhva, O. ;
Astrom, K. ;
Bergsten, C. .
ANIMAL, 2015, 9 (11) :1859-1865
[25]  
Pérez-Escudero A, 2014, NAT METHODS, V11, P743, DOI [10.1038/NMETH.2994, 10.1038/nmeth.2994]
[26]   You Only Look Once: Unified, Real-Time Object Detection [J].
Redmon, Joseph ;
Divvala, Santosh ;
Girshick, Ross ;
Farhadi, Ali .
2016 IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR), 2016, :779-788
[27]   MobileNetV2: Inverted Residuals and Linear Bottlenecks [J].
Sandler, Mark ;
Howard, Andrew ;
Zhu, Menglong ;
Zhmoginov, Andrey ;
Chen, Liang-Chieh .
2018 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR), 2018, :4510-4520
[28]   Past, present and future approaches using computer vision for animal re-identification from camera trap data [J].
Schneider, Stefan ;
Taylor, Graham W. ;
Linquist, Stefan ;
Kremer, Stefan C. .
METHODS IN ECOLOGY AND EVOLUTION, 2019, 10 (04) :461-470
[29]   A critical review o f animal-based welfare indicators for polar bears (Ursus maritimus) in zoos: Identification and evidence of validity [J].
Skovlund, C. R. ;
Kirchner, M. K. ;
Moos, L. W. ;
Alsted, N. ;
Manteca, X. ;
Tallo-Parra, O. ;
Stelvig, M. ;
Forkman, B. .
ANIMAL WELFARE, 2021, 30 (01) :1-18
[30]   Giant panda behaviour recognition using images [J].
Swarup, Pranjal ;
Chen, Peng ;
Hou, Rong ;
Que, Pinjia ;
Liu, Peng ;
Kong, Adams Wai Kin .
GLOBAL ECOLOGY AND CONSERVATION, 2021, 26