부산대학교 도서관

Simple Summary: Monitoring drinking behavior allows producers to assess the health and well-being of their beef cattle. Changes in regular drinking behavior can serve as an indicator of potential health issues. Detecting these issues early on enables the application of timely interventions, mitigating the likelihood of severe complications and enhancing the prospects for prompt and efficient treatments. In the current study, we used computer vision techniques to study and analyze the drinking behavior of beef cattle. Two different camera positions were used to identify the drinking behavior. Our proposed method was able to successfully identify both the drinking behavior and drinking time of beef cattle. Monitoring the drinking behavior of animals can provide important information for livestock farming, including the health and well-being of the animals. Measuring drinking time is labor-demanding and, thus, it is still a challenge in most livestock production systems. Computer vision technology using a low-cost camera system can be useful in overcoming this issue. The aim of this research was to develop a computer vision system for monitoring beef cattle drinking behavior. A data acquisition system, including an RGB camera and an ultrasonic sensor, was developed to record beef cattle drinking actions. We developed an algorithm for tracking the beef cattle's key body parts, such as head–ear–neck position, using a state-of-the-art deep learning architecture DeepLabCut. The extracted key points were analyzed using a long short-term memory (LSTM) model to classify drinking and non-drinking periods. A total of 70 videos were used to train and test the model and 8 videos were used for validation purposes. During the testing, the model achieved 97.35% accuracy. The results of this study will guide us to meet immediate needs and expand farmers' capability in monitoring animal health and well-being by identifying drinking behavior. [ABSTRACT FROM AUTHOR]