Carrot Raw Material Recognition Dataset

The current agricultural sector faces challenges in crop recognition and monitoring due to insufficient data and inaccurate labeling, which affects the development of smart agriculture. Existing solutions often rely on small-scale datasets, leading to poor model generalization capabilities. The Carrot Raw Material Recognition Dataset aims to provide a large-scale, high-quality annotated dataset to support object detection tasks in agriculture. The dataset ensures data diversity and representativeness by using drones and high-definition cameras to capture images of carrots in various environments. Additionally, multiple rounds of labeling and expert review quality control measures ensure high data quality and consistency. Data will be stored in JPG format, with corresponding JSON files provided, organized by category and scenario.