This research collaborates with the Department of Animal Science at Chung Hsing University to delve into the behavior of sheep during the birthing period using image recognition technology. Our objective is to establish an efficient image recognition system to gain a more profound understanding of various behavioral patterns exhibited by sheep during the production phase. The core purpose of this study is to enhance our understanding of sheep reproductive behavior, leading to the prediction of birthing times and promoting the intelligence and efficiency of farm management.
Through image recognition, we plan to capture various features of sheep during the birthing process, such as movement, eating, activity patterns, etc. We will utilize deep learning techniques to analyze and identify these behavior patterns, contributing to the creation of an accurate and reliable behavioral database, ultimately improving the precision of predicting birthing times.
The successful implementation of a sheep behavior image recognition system will provide agriculture with more reliable information. This not only assists farmers in managing sheep more effectively, enhancing productivity but also contributes to reducing labor costs. Furthermore, this research offers valuable resources for the field of animal behavior, fostering in-depth studies on animal behavior and providing additional avenues for future developments in related fields.
