RiceTagosodes-DB - Dataset of Aerial Images of Rice Crops for Detection of Tagosodes orizicolus

This dataset contains images and annotations used in the study “Detection of Tagosodes orizicolus in Aerial Images of Rice Crops Using Machine Learning”
When using this dataset, please cite:
Rivera-Cartagena, A.; Mejia-Cabrera, H.I.; Arcila-Diaz, J. Detection of Tagosodes orizicolus in Aerial Images of Rice Crops Using Machine Learning. AgriEngineering 2025, 7, 147. https://doi.org/10.3390/agriengineering7050147

The RiceTagosodes-DB dataset is composed of a total of 1,227 original images, of which 825 were captured in the city of Lambayeque, Peru, and 402 in Pacanguilla, Peru. These images were acquired using a drone over rice crops. Each image has dimensions of 5472 × 3648 pixels in JPEG format, with a resolution of 72 pixels per inch both horizontally and vertically, and a color depth of 24 bits. The final database consists of 1500 images, distributed into three subfolders: empty, rice, and sogata, with 500 images in each category. All images were standardized to dimensions of 256 × 256 pixels in compressed JPEG format, with a resolution of 96 pixels per inch both horizontally and vertically, and a color depth of 24 bits.