Palm Tree Detection

Project overview

Images were annotated for a computer vision dataset focused on detecting and distinguishing palm trees from non-palm trees in outdoor and agricultural environments. The work involved identifying palm trees, creating accurate annotations, and labelling other trees under a separate non-palm class.

The main objective was to create high-quality training data for an AI model capable of automatically detecting palm trees and differentiating them from other types of trees.

Work performed

• Reviewed and organised the image dataset before annotation.
• Palm trees were identified across outdoor and agricultural environments.
• Accurate bounding boxes were created around visible palm trees.
• Other tree species were labelled using the Non_Palm_Tree class.
• Class labelling remained consistent throughout the dataset.
• Partially visible trees were annotated where applicable.
• Visual characteristics were used to separate palm trees from other species.
• Quality checks were performed, and inaccurate or incomplete annotations were corrected.
• Annotations were confirmed consistent and suitable for training an object detection model.

What made it hard

• Differentiating palm trees from visually similar non-palm trees.
• Trees appeared at different distances, sizes and growth stages.
• Partially visible or occluded trees had to be annotated carefully.
• Overlapping tree canopies made individual trees hard to separate.
• Complex backgrounds contained vegetation, buildings, roads and other objects.
• Varying camera angles and image resolutions affected tree visibility.
• Accurate bounding boxes around irregular tree shapes required close attention.

Result

A high-quality palm tree detection dataset was completed with consistent annotations for both palm and non-palm trees. The dataset was prepared for training a computer vision model capable of automatically detecting palm trees while distinguishing them from other tree species.

Final deliverable

Accurate and consistent annotations for the Palm_Tree and Non_Palm_Tree classes were delivered throughout the dataset. The final output was prepared according to project requirements and was ready for AI model training and automated palm tree detection applications.