Young Plant & Flower Detection

Project overview

Agricultural images were annotated for a computer vision dataset focused on detecting young flower plants in field environments. The work involved identifying visible young flower plants and creating accurate bounding box annotations to generate high-quality training data for an AI-based object detection model.

The main objective was to prepare a reliable dataset for automatically detecting young flower plants at early growth stages in real-world agricultural conditions.

Work performed

• Reviewed and organised the agricultural image dataset before annotation.
• Identified young flower plants in each field image.
• Created accurate bounding boxes around individual plants.
• Labelling was kept consistent under the Plant class.
• Partially visible and small plants were annotated where applicable.
• Young flower plants were carefully distinguished from weeds, soil, crop residue and surrounding vegetation.
• Handled plants at different sizes and early growth stages.
• Performed quality checks and corrected inaccurate, missing or poorly positioned bounding boxes.
• Ensured the annotations were consistent and suitable for object detection model training.

What made it hard

• Young plants were often very small and difficult to identify.
• Early-stage plants showed few visual features, making them hard to separate from weeds.
• Dense vegetation and complex field backgrounds made detection challenging.
• Partially visible plants required careful annotation.
• Similar-looking weeds were easily confused with young flower plants.
• Lighting conditions and shadows varied and affected plant visibility.
• Keeping bounding boxes tight and accurate around small plants demanded significant attention to detail.

Result

A high-quality young flower plant detection dataset was completed with accurate and consistent bounding box annotations under the Plant class. The dataset was prepared for training a computer vision model capable of automatically detecting young flower plants in agricultural environments.

Final deliverable

Accurate and consistent bounding box annotations for young flower plants were delivered across the dataset. The final output was prepared to the project requirements and was ready for AI model training and automated agricultural plant detection applications.