YOLO-Based Crop Object Detection and Counting System

Intelligent crop detection and counting systems are critical for enhancing agricultural productivity and reducing labor costs. This paper proposes a high-precision system combining an optimized YOLOv3 model with an adaptive counting algorithm to address the limitations of traditional methods, such as low efficiency and environmental sensitivity. The YOLOv3 architecture is enhanced through multi-scale prediction and feature pyramid networks (FPN), enabling robust detection of small crop targets in complex backgrounds. Experimental results demonstrate a detection accuracy exceeding 95% and a counting error rate below 5%, outperforming conventional approaches. This system provides a scalable solution for precision agriculture, with significant theoretical and practical implications.