Article Details

Abstract

With the large-scale and intelligent development of animal husbandry in China, traditional cattle identification methods exhibit problems including easy detachment and stress induction and can hardly meet the requirements of precision livestock farming. Cattle face recognition technology based on Deep Learning provides a new approach for individual identification with its non-contact and high-precision characteristics. This paper systematically reviews the research progress of this technology. It illustrates the principles and advantages of Deep Learning in fine-grained recognition. It sorts out key technologies such as network structure optimization, few-shot learning and lightweight deployment. It summarizes the complete process from data collection to model evaluation. In view of current challenges including dependence on two-dimensional appearance features, large demand for annotated data and model complexity, this paper prospects future directions such as multi-modal fusion, self-supervised and few-shot learning and lightweight models to promote the in-depth application of this technology in smart animal husbandry.

Keywords

cattle face recognition; Deep Learning; precision livestock farming; individual identification; computer vision; Convolutional Neural Network

References

1. [1] Bing Shuying, Zhang Yuyu, Ji Yuanhao, etc Improved lightweight cow face recognition model for YOLOv8s [J]. China Agricultural Machinery Chemistry Journal, 2025, 46 (8): 128-134
2. [2] Tian Huijuan, Cao Mengqi, Ru Chunrui, etc Research on Cow Facial Recognition Method Based on Improved MobileNetV2 [J]. Modern Information Technology, 2025, 9 (11): 20-24
3. [3] Han Yu, Qi Kangkang, Zheng Jiye, etc A lightweight beef cattle facial recognition method based on improved YOLOv11 [J]. Smart Agriculture, 2025, 7 (3): 173-184
4. [4] Gaojie, Cao Hao Research on Cow Face Recognition Method Based on Improved YOLO v7 [J]. Journal of Xinyang Agriculture and Forestry College, 2024, 34 (3): 125-130
5. [5] Liu Jinwei, Li Fuzhong, Chen Xinwen, etc Cow face detection and recognition model based on twin networks and metric learning [J]. Computer Technology and Development, 2025, 35 (9): 175-181
6. [6] Luan Haotian, Qi Yongsheng, Liu Liqiang, etc VanillaFaceNet: A high-precision and fast inference based cow face recognition method [J]. Journal of Agricultural Engineering, 2024, 40 (18): 120-131
7. [7] Jiao Jie, Qi Yongsheng, Liu Liqiang, etc A scene adaptive dual branch cow face efficient recognition algorithm [J]. Acta Sinica, 2024, 52 (9): 3251-3261
8. [8] Liu Bo Research on Identification and Single Object Tracking Methods for Cows in Natural Scenes [D]. Baotou: Inner Mongolia University of Science and Technology, 2024
9. [9] Liu Jinwei, Guo Leifeng, Liu Donghao, etc Research progress and application scenario analysis of cow face recognition technology [J]. Northern Animal Husbandry, 2025 (1): 13
10. [10] KUMAR S，PANDEY A，SATWIK K S R，et al. Deep Learning Framework for Recognition of Cattle Using Muzzle Point Image Pattern [J].Measurement，2018，116：1-17.
11. [11] ANDREW W，GREATWOOD C，BURGHARDT T. Visual Localisation and Individual Identification of Holstein Friesian Cattle via Deep Learning [C]//Proceedings of the IEEE International Conference on Computer Vision Workshops.Venice：IEEE，2017：2850-2859.
12. [12] BELLO R W，OGUNDOKUN R O，OWOLAWI P A，et al. Application of Convolutional Neural Networks in Animal Husbandry: A Review [J].Mathematics，2025，13（1）：1-20.
13. [13] HANSEN M F，SMITH M L，SMITH L N，et al. Towards On-Farm Pig Face Recognition Using Convolutional Neural Networks [J].Computers in Industry，2018，98：145-152.
14. [14] SALAU J，HAAS J H，JUNGE W，et al. Automated Calculation of Udder Depth and Rear Leg Angle in Holstein-Friesian Cows Using a Multi-Kinect Cow Scanning System [J].Biosystems Engineering，2017，160：154-169.

Funding

2023 Wuhan University Student Research and Innovation Team (XST202310)

About the authors

Li Tianhe (2005-), female, Han ethnicity, from Jingzhou, Hubei Province, is currently pursuing a bachelor's degree with a research focus on deep learning.