Australian Animal Species Selection and Image Data Collection

Using computer image-processing technology for animal identification has improved labour-intensive agriculture and animal husbandry efficiency and accuracy. Additionally, camera traps have provided researchers with abundant biodiversity and wildlife habitat conservation data. However, despite these advancements, there is still a need for taxonomic rank-based animal classification studies, and fewer datasets are available to train AI models based on taxonomic ranks. The importance of taxonomy was ignored by researchers and dataset creators in previous animal classification studies, which is the basis and prerequisite of classifying wild animal images. To this end, this paper presents a dataset labelled with the taxonomic ranking. The dataset is focused on Australian animals and includes 47 taxonomically significant animal species, each with a maximum of 600 manually labelled images. This dataset will contribute significantly to the field of animal identification, as it provides researchers and practitioners with a valuable tool to improve the efficiency and accuracy of animal identification, which can have significant implications for agriculture, animal husbandry, biodiversity research, and wildlife conservation.