Onion Head Detection

Project overview

Onion images were annotated for a computer vision dataset focused on detecting onion heads in agricultural environments. Visible onion heads were identified and accurate annotations created to generate high-quality training data for an AI-based object detection model.

Work performed

• Reviewed and organised the onion image dataset before annotation.
• Visible onion heads were identified in each image.
• Accurate bounding boxes were created around individual onion heads.
• Labelling remained consistent throughout using the Onion_Head class.
• Partially visible onion heads were annotated where applicable.
• Onion heads were carefully distinguished from leaves, soil and surrounding objects.
• Quality checks were performed, and inaccurate or incomplete annotations were corrected.
• Annotations were confirmed suitable for computer vision model training.

What made it hard

• Detecting onion heads in dense agricultural environments.
• Partially visible or occluded onion heads had to be handled carefully.
• Onion heads needed separating from surrounding soil and plant leaves.
• Small onion heads were difficult to annotate accurately.
• Varying onion sizes, shapes, lighting conditions and backgrounds reduced clarity.
• Tight and consistent bounding boxes had to be maintained around each onion head.

Result

A high-quality onion head detection dataset was completed with accurate bounding box annotations. The dataset was prepared for training a computer vision model capable of automatically detecting onion heads in agricultural images.

Final deliverable

Accurate and consistent bounding box annotations for onion heads were delivered throughout the dataset. The final output was prepared according to project requirements and was ready for AI model training and agricultural computer vision applications.