Precision livestock farming (PLF) is the development and utilization of technologies to assist animal production and address challenges of farm stewardship. Supportive technologies for animal farming are therefore needed for data collection and analysis and decision-making. Consequently, productivity and health of animals and economic benefits of farmers may be compromised. Without quality care, some early signs of animal abnormalities may not be detected timely. This will become even more challenging with predicted labor shortages for farm jobs in the future. As production is intensified to meet the increased demands, producers are confronted with increasing pressure to provide quality care for increasing number of animals per management unit. Based on the prediction by Yitbarek, 2.6 billion cattle, 2.9 billion sheep and goats, 1.1 billion pigs, and 37.0 billion poultry will be produced in 2050. As livestock and poultry contribute to a large proportion (~30% globally) of daily protein intake through products like meat, milk, egg, and offal, animal production is predicted to increase accordingly to feed the growing human population. The world population is projected to be 9.2 billion in 2050. Finally, recommendations on future research were provided to develop and improve CNN-based computer vision systems for improved welfare, environment, engineering, genetics, and management of farm animals.įood security is one of the biggest challenges in the world. Besides practices in optimizing CNN-based computer vision systems, system applications were also organized based on year, country, animal species, and purposes. Judgment of model performance and performance based on architectures were discussed.

Strategies of algorithm development included distribution of development data, data augmentation, hyperparameter tuning, and selection of evaluation metrics. CNN architectures were reviewed based on the computer vision tasks in animal farming. In this research, preparations for system development, including camera settings, inclusion of variations for data recordings, choices of graphics processing units, image preprocessing, and data labeling were summarized. Cattle, sheep/goats, pigs, and poultry were the major farm animal species of concern. The objective of this study is to systematically review applications of CNN-based computer vision systems on animal farming in terms of the five deep learning computer vision tasks: image classification, object detection, semantic/instance segmentation, pose estimation, and tracking. Convolutional neural network (CNN)-based computer vision systems have been increasingly applied in animal farming to improve animal management, but current knowledge, practices, limitations, and solutions of the applications remain to be expanded and explored.

Animal farm pd