Research trends in livestock facial identification – A review

This review examines and summarizes research related to video processing and convolutional neural network-based deep learning for animal face recognition, identification and re-identification, and assesses its applicability to precision livestock farming for improving animal welfare and production efficiency. To enhance efficiency in production and animal management, address concerns regarding animal welfare and health, and minimize environmental impact and resource usage, technologies for estimating growth, identifying individuals, and monitoring behavior are essential. As computer technology advances, livestock monitoring systems have also evolved. These systems are broadly categorized into sensor-based contact methods and video-based non-contact methods. Recently, the rapid advancement of computer technologies, including deep learning algorithms, has enabled the analysis of accumulated data to continuously monitor and analyze animal conditions without human intervention, resulting in efficient and automated monitoring systems. The integration of video processing and convolutional neural network-based deep learning for animal face recognition, identification, and re-identification holds significant promise for precision livestock farming. By leveraging these advanced technologies, it is possible to enhance production efficiency and animal management, address animal welfare and health concerns, and reduce environmental impact and resource usage. These systems represent a crucial step forward in the evolution of livestock management, offering precise and efficient tools for the estimation of growth, individual identification, and behavior monitoring, ultimately contributing to improved animal welfare and production outcomes.