TomatoLite Dataset for Object Detection

# Tomato Dataset ## Overview This repository provides the dataset used in our paper: **"Lightweight Tomato Disease Detection in Complex Field Environments based on Improved YOLOv8"**. The dataset contains field images of tomato plants collected under real-world conditions (varying illumination, occlusion, cluttered backgrounds). Annotations are provided in **YOLO** format and (optionally) **Pascal VOC** format for reproducibility. > IMPORTANT: Train/val splits are provided and should be used as-is to reproduce the paper results. ## Task - **Task type**: Object detection - **Target**: Tomato disease/pest symptoms on leaves (bounding boxes) ## Classes The class index-to-name mapping is provided in: - `yolo/Labels.txt` Please ensure your training code uses the exact same class ordering.