FieldVitis

This dataset is designed for deep learning applications in grape leaf disease identification. It consists of real-world images of grape leaves captured in loco, representing natural lighting conditions, variable backgrounds, and diverse leaf orientations. The images encompass both healthy specimens and leaves affected by common grapevine diseases, being classified into four classes: Black_rot, Esca, Leaf_blight, and Healthy. The class distribution is as follows: Healthy: 258 images Black_rot: 39 images Esca: 201 images Leaf_blight: 158 images This dataset was constructed by integrating samples from the following publicly available sources: ESCA-dataset— License: CC BY 4.0 PlantDoc-Dataset — License: CC BY 4.0 Grape Disease — License: CC BY 4.0 Images from these sources were manually selected and, in some cases, cropped to better focus on the grape leaves. The resulting dataset maintains the integrity of the original data while providing a more standardized format for deep learning experiments. We extend our sincere thanks to the authors and contributors of these datasets for their efforts in collecting and sharing valuable open-access resources.