Sweet Potato Size Estimation Dataset

The current agricultural sector faces challenges in crop growth monitoring and yield prediction, especially in sweet potato size estimation. Traditional methods often rely on manual measurement, which is inefficient and prone to errors. Existing solutions mostly involve image-based analysis tools, but they still have shortcomings in accuracy and real-time performance. This dataset aims to provide a high-quality image dataset of sweet potatoes to support the training of object detection models, addressing the technical challenges of sweet potato size estimation. The dataset includes images of sweet potatoes from real-growing environments, captured by high-resolution cameras and annotated in multiple rounds to ensure consistency and accuracy. Data collection used professional photographic equipment, ensuring image quality under different lighting and environmental conditions. Quality control measures include expert review and multiple rounds of annotation to ensure data annotation accuracy. The data will be stored in JPG format and organized into folders, with each folder corresponding to a planting area. The dataset structure is clear and user-friendly.