Rice Field Insect Detection Dataset with Bounding Box Annotations for 21 Pest Categories
Year: 2026
Type: dataset
Abstract: This dataset contains rice field images with bounding-box annotations for 21 pest categories. It supports the training and evaluation of object detection models for automated insect detection in agricultural field conditions. The dataset.zip archive includes training and validation splits, corresponding YOLO-format annotation files, and a data.yaml file that defines the class labels and dataset paths. The accompanying software-code.zip archive contains the improved YOLOv9 implementation, scripts for training, validation, and inference, and a trained model checkpoint (best.pt). The README.md explains the dataset structure, annotation format, and software setup. The dataset was derived from Rice Field Insect Detection, version 2, hosted in the Jiangsu University workspace on Roboflow.
Author: li, Zehan
Language: English
Topic: Smart Agriculture and AI
Subfield: Plant Science
Domain: Life Sciences
SDG: Zero hunger
Open Access status: green
OpenAlex ID: W7225856995
