Citrus Crop Databases from Misantla, Veracruz, Mexico

This dataset provides a curated collection of citrus plant images designed for the identification of common citrus diseases and pests. It was developed as part of a *Master’s thesis* and is intended to support research in computer vision, machine learning, and intelligent agricultural systems. The dataset follows a standard train/test split and includes nine (9) distinct classes , each corresponding to a specific citrus condition or pest. All images have a uniform resolution of 450 × 450 pixels , facilitating consistent model training and evaluation. Each image is paired with its corresponding annotation file, enabling supervised learning tasks such as image classification and object detection. Classes included The dataset comprises the following classes: - AranaRoja- Flor- FlorDanada - InsectoMinador - Minador - Pulgon - RonaDeLaHoja - RonaDelLimon - Trip Dataset structureThe directory structure is organized as follows: - train/ - 9 classes - 50 images per class - Total: 450 images - test/ - 9 classes - 25 images per class - Total: 225 images A classes.txt file is included to define class names and indices. Each image maintains a one-to-one correspondence with its annotation file, ensuring consistency between visual data and labels. Data characteristics - Number of classes: 9 - Total images: 675 - Image resolution: 450 × 450 pixels - Data split: Training and testing - Application domain: Citrus plant health and pest detection Intended useThis dataset is suitable for:- Training and evaluation of machine learning models for citrus disease and pest recognition- Image classification and object detection tasks- Research in precision agriculture and plant pathology - Educational use in computer vision and artificial intelligence Geographic and academic contextAll images were collected from citrus crops in the Misantla region, Veracruz, Mexico , reflecting real agricultural conditions. The dataset was developed under academic supervision as part of a *Master’s thesis* and is shared to promote reproducibility and further research.