Sugar Beet Size Classification

Project overview

Agricultural field images containing sugar beet plants were annotated for a computer vision dataset focused on detecting sugar beets and classifying them by size. Every visible sugar beet was annotated with a bounding box and assigned an appropriate size-based class.

The dataset was prepared as training data for an AI model capable of detecting sugar beets and distinguishing between small, medium and large examples.

Work performed

• Reviewed and organised the agricultural image dataset.
• Identified individual sugar beets in each image.
• Drew accurate bounding boxes around visible sugar beets.
• Classified every sugar beet as Small, Medium or Large.
• Applied class labels consistently across the dataset.
• Annotated partially visible sugar beets where applicable.
• Performed quality checks and corrected inaccurate bounding boxes or class labels.
• Ensured the annotations were suitable for object detection and size classification model training.

What made it hard

• Distinguishing between sugar beets of similar size.
• Size classification had to stay consistent from image to image.
• Overlapping and partially visible sugar beets complicated placement.
• Camera distances, viewing angles and lighting conditions varied widely.
• Creating tight and accurate bounding boxes around individual sugar beets required care.

Result

A high-quality sugar beet detection and size classification dataset was completed covering the Small, Medium and Large classes. Every visible sugar beet was given the appropriate size category and enclosed in an accurate bounding box, and the dataset was prepared for training an AI model capable of detecting and classifying sugar beets by size automatically.

Final deliverable

Accurately annotated sugar beet images were delivered with bounding boxes and size-based labels for the Small, Medium and Large classes, with consistent labelling and annotation quality throughout. The final output was prepared to the project requirements and was ready for AI model training and development.