Abstract: This research report details the development of a machine learning model for identifying, localizing, and counting crocodiles in drone images. It was primarily motivated by the need for effective crocodile population management in commercial crocodile farms, with potential applications in wildlife management. The model extracts physiological metrics such as head and body size, providing additional insights into the crocodile population. Challenges included limited annotated data. The model shows promise in accurately identifying and localizing crocodiles and extracting physiological metrics. However, it is sensitive to the apparent scale of crocodiles influenced by drone flight altitude. Careful consideration of flight altitude is necessary for accurate estimation of crocodile abundance and size. Overall, this model represents a significant advancement in crocodile population management and wildlife monitoring, with future research focusing on addressing limitations and expanding capabilities through larger, annotated datasets.