Advanced intelligent monitoring technologies for animals: A survey
Pengfei Xu,Yuanyuan Zhang,Minghao Ji,Songtao Guo,Zhanyong Tang,Xiang Wang,Jing Guo,Junjie Zhang,Ziyu Guan
DOI: https://doi.org/10.1016/j.neucom.2024.127640
IF: 6
2024-04-06
Neurocomputing
Abstract:Effective animal intelligent monitoring is of great value in terms of ecological protection and endangered specie conservation. At present, computer vision technologies have shed light on animal intelligent monitoring. Especially, numerous deep learning-based models and methods have been developed to address various challenges, and have made substantial strides in this field. However, there are still several problems to be solved and related areas to be mined, such as exploring new strategies to enhance the robustness and generalization ability of models, designing novel models for complex environments, and establishing large-scale publicly available animal datasets for performance verification of models. Therefore, we comprehensively elaborated and analyzed existing works on animal intelligent monitoring based on advanced information science technologies, so as to provide useful information assistance for relevant researchers. In this paper, we focus on three primary task fields: precise animal localization, tracking and individual identification. Specifically, we elucidate the definition and significance of each monitoring task, and summarize the baseline models for addressing different problems. We provide a specific analysis of strategies and prototypes of the models and methods employed in each tasks following by the technical progression from traditional machine learning to deep learning. In addition, we make a comparison and analysis of the relevant methods, summarize their similarities and differences between them, and point out the advantages and disadvantages of these methods. Finally, we present several unresolved challenges and problems in animal intelligent monitoring and provide potential research directions in the future. We expect that our review can serve as reference and guidance for related research fields.
computer science, artificial intelligence