Automatic Classification of Body Condition Score of Cows: A Systematic Literature Review

With technological advances in recent decades, cattle farms are constantly striving to optimize local milk production. Several tools can help manage milking waves and the milking process, such as the Body Condition Score (BCS), an important tool for analyzing cattle health and making intelligent decisions to optimize milk production on a farm. Although the BCS is a useful tool, it is not widely used due to the difficulty of performing its analysis, as it has been found that manual assessments can be very time-consuming on large farms with hundreds of cows. Therefore, it would be useful to automate this process. In this work, we investigated the feasibility of automatic classification of BCS of cattle using computer vision techniques based primarily on deep learning systems, machine learning techniques or linear regression, and classical image processing. This systematic literature review (SLR) was conducted using standard methods for documentary research. The SLR included a search for related research and the current state of the art for cattle body condition classification. By exploring the potential of modern systems, this work is expected to provide a new perspective on the automatic classification of BCS and enable a more accurate and reliable evaluation of cattle body condition.