Active Learning-Driven Lightweight YOLOv9: Enhancing Efficiency in Smart Agriculture

TL;DR

This paper introduces an active learning-driven lightweight object detection framework tailored for real-time agricultural robot applications, significantly improving detection accuracy of tomatoes and flowers on edge devices under practical conditions.

Contribution

It presents a novel integration of data analysis, model design, and active learning to enhance detection performance while maintaining low computational costs for edge deployment.

Findings

Achieved 67.8% mAP in detection accuracy under limited annotations.

Reduced computational cost with a lightweight model suitable for edge devices.

Improved detection of small and occluded objects in complex agricultural environments.

Abstract

This study addresses the demand for real-time detection of tomatoes and tomato flowers by agricultural robots deployed on edge devices in greenhouse environments. Under practical imaging conditions, object detection systems often face challenges such as large scale variations caused by varying camera distances, severe occlusion from plant structures, and highly imbalanced class distributions. These factors make conventional object detection approaches that rely on fully annotated datasets difficult to simultaneously achieve high detection accuracy and deployment efficiency. To overcome these limitations, this research proposes an active learning driven lightweight object detection framework, integrating data analysis, model design, and training strategy. First, the size distribution of objects in raw agricultural images is analyzed to redefine an operational target range, thereby improving learning stability under real-world conditions. Second, an efficient feature extraction module is incorporated to reduce computational cost, while a lightweight attention mechanism is introduced to enhance feature representation under multi-scale and occluded scenarios. Finally, an active learning strategy is employed to iteratively select high-information samples for annotation and training under a limited labeling budget, effectively improving the recognition performance of minority and small-object categories. Experimental results demonstrate that, while maintaining a low parameter count and inference cost suitable for edge-device deployment, the proposed method effectively improves the detection performance of tomatoes and tomato flowers in raw images. Under limited annotation conditions, the framework achieves an overall detection accuracy of 67.8% mAP, validating its practicality and feasibility for intelligent agricultural applications.