TITLE:
Research on Sheep Face Recognition Based on Deep Learning and YOLOv3
AUTHORS:
Xu Zhen, Guoqing Chen, Lerong Wen, Haifeng Jia
KEYWORDS:
YOLOv3, Deep Learning, Biometric Technology, Grassland Livestock Husbandry, Sustainable Development
JOURNAL NAME:
Open Journal of Applied Sciences,
Vol.16 No.4,
April
30,
2026
ABSTRACT: To address the issues of low management efficiency, imprecise data tracking, and cumbersome rapid decision-making in traditional grassland animal husbandry, this study proposes a livestock monitoring model based on biometric technology. The study integrates sheep face recognition and in-depth data analysis, and combines them with deep learning algorithms to achieve non-contact, high-precision individual identification of livestock. A total of 400 sheep were tested in typical grassland areas of Inner Mongolia. The results indicated that the system achieved a comprehensive recognition accuracy of 94.1% and a recall rate of 81%, which effectively resolved the problems of traditional ear tags being prone to loss and damage, while also facilitating the monitoring of livestock diseases. This technology enables the full-lifecycle tracking of livestock, encompassing the intelligent monitoring of health status, the precise management of pastures, and the scientific prediction of breeding cycles. The research provides technical support for the digital transformation of grassland animal husbandry, and particularly offers practical assistance for the sustainable development of ecological animal husbandry in degraded grassland regions.