Image-Text Multi-Modal Dataset of Corn Leaf Diseases based on Manual Annotation and Contrast Generation Model

The dataset consists of two parts: leaf disease image data and corresponding text description data, totaling 1653 sets. Among them, the image modal data is sourced from open source datasets such as AI Challenger, Plant Village, OpenDataLab (CD&S), as well as existing self built, jointly constructed, and purchased sources. High definition images of nine typical leaf diseases, including large spot disease, small spot disease, brown spot disease, curved mold leaf spot disease, common rust disease, southern rust disease, gray spot disease, round spot disease, and dwarf mosaic disease, have been collected and organized; The text modality involves manually annotating diagnostic text descriptions of images based on prior knowledge such as literature, professional books, and scientific data. The content of the text modality covers key information such as disease types, pathological features, and severity.