TOM2024

Description

The TOM2024 dataset is a valuable resource for agricultural research, comprising 25,844 raw images and 12,227 labeled images of tomato, onion, and maize crops. These images are categorized into 30 classes, facilitating precise identification of pests and diseases, which is crucial for improving crop management and food security. The dataset supports sustainable agriculture by promoting early and accurate pest and disease detection, reducing reliance on pesticides. Its accessibility allows researchers and institutions to develop advanced digital solutions, enhancing the effectiveness of pest and disease management. Additionally, the dataset's versatility—searchable by region, crop type, and other criteria—makes it suitable for model development, education, and agricultural extension services. With high-resolution images captured under diverse environmental conditions, TOM2024 offers a robust foundation for training AI models, ultimately contributing to the advancement of precision agriculture.

Steps to reproduce

The dataset was compiled through a systematic and structured data collection process to capture high-quality images of plant pests and diseases across key crops: maize, tomato, and onion. The steps employed for the curation involved the site selection, image acquisition, grouping of images into crop types and issue types (pest or disease), identification of issues by experts, cropping and resizing of images, validation of cropped images by experts, and saving of images.

Institutions

Categories

Funding

Licence