INSECT204-A Novel Large-Scale Benchmark Dataset for Insect Pest Recognition
Description
INSECT204 is a novel, large-scale, and high-quality benchmark dataset specifically designed for insect pest recognition in agricultural and natural settings. It comprises 108,999 images across 204 distinct insect species, significantly advancing the scale and diversity of existing public datasets like IP102. To ensure ecological realism, the collection features images captured in complex real-world field environments, includes various life stages such as eggs, larvae, pupae, and adults, and incorporates both harmful pests and beneficial natural enemies. The dataset's integrity is maintained through a rigorous expert-based validation process where five entomologists reviewed each image, and it is partitioned into training, validation, and testing subsets with a 6:1:3 ratio to facilitate robust deep learning research. Currently available at https://agriinsect360.com/, INSECT204 serves as a comprehensive resource to address challenges like data imbalance and to support the development of accurate automated identification systems for precision agriculture.