Improving Lightweight Weed Detection via Knowledge Distillation

TL;DR

This paper introduces knowledge distillation techniques to improve lightweight weed detection models, enabling accurate real-time identification of weed species on resource-constrained devices for precision agriculture.

Contribution

It proposes Channel-wise and Masked Generative Distillation methods to enhance lightweight object detection models for weed identification, demonstrating improved accuracy and real-time deployment feasibility.

Findings

CWD and MGD improve AP50 by 2.5% and 1.9% respectively over baseline.

Distilled models perform effectively on embedded devices like Jetson Orin Nano and Raspberry Pi 5.

The methods are practical for real-time weed detection in resource-limited agricultural settings.

Abstract

Weed detection is a critical component of precision agriculture, facilitating targeted herbicide application and reducing environmental impact. However, deploying accurate object detection models on resource-limited platforms remains challenging, particularly when differentiating visually similar weed species commonly encountered in plant phenotyping applications. In this work, we investigate Channel-wise Knowledge Distillation (CWD) and Masked Generative Distillation (MGD) to enhance the performance of lightweight models for real-time smart spraying systems. Utilizing YOLO11x as the teacher model and YOLO11n as both reference and student, both CWD and MGD effectively transfer knowledge from the teacher to the student model. Our experiments, conducted on a real-world dataset comprising sugar beet crops and four weed types (Cirsium, Convolvulus, Fallopia, and Echinochloa), consistently show increased AP50 across all classes. The distilled CWD student model achieves a notable improvement of 2.5% and MGD achieves 1.9% in mAP50 over the baseline without increasing model complexity. Additionally, we validate real-time deployment feasibility by evaluating the student YOLO11n model on Jetson Orin Nano and Raspberry Pi 5 embedded devices, performing five independent runs to evaluate performance stability across random seeds. These findings confirm CWD and MGD as an effective, efficient, and practical approach for improving deep learning-based weed detection accuracy in precision agriculture and plant phenotyping scenarios.