Carrot Segmentation

Project overview

Agricultural images containing carrot crops were annotated for a computer vision dataset focused on detecting and segmenting carrots in field environments. The work involved creating precise polygon annotations around visible carrot plants and crop regions to generate high-quality training data for an AI-based segmentation model.

Work performed

• Reviewed and organised the agricultural image dataset before annotation began.
• Identified visible carrot plants and crop regions in the field images.
• Created detailed polygon annotations following the boundaries of the carrot vegetation.
• Maintained consistent labelling using the Carrot class.
• Carrot rows and visible crop areas were annotated accurately.
• Quality checks were carried out and inaccurate or incomplete polygons were corrected.
• Annotations were confirmed as suitable for segmentation model training.

What made it hard

• Annotating irregular and dense carrot vegetation.
• Crop boundaries had to be followed across long field rows.
• Partially visible plants and overlapping vegetation complicated the outlines.
• Carrot plants were easily confused with soil, weeds and surrounding vegetation.
• Polygon boundaries had to stay accurate under changing lighting and field conditions.

Result

A high-quality carrot segmentation dataset was completed with precise polygon annotations for the visible carrot crop areas. The dataset was prepared for training a computer vision model capable of detecting and segmenting carrot crops automatically in agricultural field images.

Final deliverable

Accurate polygon annotations for carrot crop regions were delivered with consistent labelling and annotation quality maintained throughout the dataset. The final output met the project requirements and was ready for AI model training and agricultural computer vision applications.