This is an early access version, the complete PDF, HTML, and XML versions will be available soon.
Recognition of Foal Nursing Behavior Based on an Improved RT-DETR Model
by
Yanhong Liu
1,2,3,4,
Fang Zhou
5,
Wenxin Zheng
1,
Tao Bai
1,3,4,
Xinwen Chen
6,7,* and
Leifeng Guo
1,2,7,* 1
College of Computer and Information Engineering, Xinjiang Agricultural University, Urumqi 830052, China
2
Agricultural Information Institute, Chinese Academy of Agricultural Sciences, Beijing 100080, China
3
Xinjiang Agricultural Informatization Engineering Technology Research Center, Urumqi 830052, China
4
Ministry of Education Engineering Research Centre for Intelligent Agriculture, Urumqi 830052, China
5
College of Information Science and Technology, Shihezi University, Shihezi 832000, China
6
Institute of Animal Husbandry Quality Standards, Xinjiang Academy of Animal Science, Urumqi 830011, China
7
Xinjiang Intelligent Livestock Key Laboratory, Urumqi 830052, China
*
Authors to whom correspondence should be addressed.
Animals 2025, 15(3), 340; https://doi.org/10.3390/ani15030340
Submission received: 27 December 2024
/
Revised: 17 January 2025
/
Accepted: 23 January 2025
/
Published: 24 January 2025
(This article belongs to the Special Issue Equine Exercise Physiology and Feeding Management: How to Keep Energy Balance)
Simple Summary
Timely monitoring and analysis of foal suckling behavior can provide valuable insights into foals’ physiological condition. A foal’s suckling posture and the mare’s standing posture during the nursing period are important prerequisites for the foal’s suckling behavior. Unlike manual observation and wearable devices, this study proposes a non-contact method using artificial intelligence (AI) vision technology to monitor the mare’s standing posture and the foal’s suckling posture. This method enables accurate recognition of both the mare’s standing posture and the foal’s suckling posture. Additionally, this study also implements real-time statistical analysis of the time the foal spends in the suckling posture. The proposed method offers a new perspective on equine reproduction for equestrian clubs and horse breeding enterprises, while also providing a supplementary approach for veterinarians and horse managers to detect early abnormalities in foal development.
Abstract
Foal nursing behavior is a crucial indicator of healthy growth. The mare being in a standing posture and the foal being in a suckling posture are important markers for foal suckling behavior. To enable the recognition of a mare’s standing posture and its foal’s suckling posture in stalls, this paper proposes an RT-DETR-Foalnursing model based on RT-DETR. The model employs SACGNet as the backbone to enhance the efficiency of image feature extraction. Furthermore, by incorporating a multiscale multihead attention module and a channel attention module into the Adaptive Instance Feature Integration (AIFI), the model strengthens feature utilization and integration capabilities, thereby improving recognition accuracy. Experimental results demonstrate that the improved RT-DETR achieves a best mAP@50 of 98.5%, increasing by 1.8% compared to the RT-DETR. Additionally, this study achieves real-time statistical analysis of the duration of the foal in the suckling posture, which is one of the important indicators for determining whether the foal is suckling. This has significant implications for the healthy growth of foals.
Keywords:
feeding; artificial intelligence; foal suckling; behavior recognition
