Researchers have introduced AerialYield-B2D, a new dataset designed for computer vision tasks in controlled greenhouse environments. This dataset contains over 500 RGB images with more than 30,000 annotated blueberry instances, categorized into five distinct ripeness stages. It includes detailed annotations such as class-specific binary masks, overall berry masks, semantic label maps, and image-level berry counts, making it suitable for tasks like ripeness segmentation and fruit counting. AI
IMPACT Provides a specialized dataset for advancing AI models in agricultural computer vision, particularly for fruit ripeness and yield estimation.
RANK_REASON Release of a new academic dataset for computer vision tasks. [lever_c_demoted from research: ic=1 ai=1.0]
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
