Research on silkworm disease identification based on an im-proved SAI-RTDETR model

This dataset is meticulously constructed for silkworm disease identification, covering images of various common silkworm diseases and healthy silkworms. Images are captured from different angles and under diverse conditions, reflecting the multifaceted states of silkworms across growth stages. Each image undergoes rigorous screening and preprocessing to ensure quality, with lesion areas precisely annotated to provide detailed information on disease locations and types, offering a robust foundation for model training and algorithm development.Emphasizing diversity and representativeness, the dataset includes images taken under different lighting, angles, and backgrounds to mirror the complexity of real-world silkworm farming. This enhances the generalization and robustness of models trained on this dataset, ensuring stable and accurate disease identification across varying scenarios. It strongly supports smart agriculture and precision farming, driving sustainable development in the sericulture industry. The dataset will continue to be refined as the paper is published.