Evaluating of Tree Branch Recognition Algorithm in Pruning Robots under Augmented Environmental Conditions.

Integrating service robots has revolutionized several sectors, by enhancing accuracy, efficiency, and scalability.Those robots are crucial in automating labor-intensive processes such as tree pruning, where accurate branch detection is vital.This research evaluates the performance of YOLOv8-seg model for recognizing tree branches as a step towards fully autonomous pruning.To address the challenges posed by diverse and complex real-world conditions, video sequences are augmented using techniques that simulate environmental variations, such as changes in brightness, contrast, and Gaussian noise.The evaluation metrics including the number of true detections, number of false detections, and precision, demonstrate robust and accurate branch perception under real-world conditions.These results highlight the potential of YOLOv8-seg to improve pruning systems, paving the way for scalable, efficient, and accurate robotic solutions in tree maintenance.