Hardware-Aware YOLO Compression for Low-Power Edge AI on STM32U5 for Weeds Detection in Digital Agriculture

TL;DR

This paper develops a low-power, hardware-aware weed detection system using YOLOv8n on STM32U5 microcontrollers, applying compression techniques to enable real-time, energy-efficient precision agriculture.

Contribution

It introduces a novel combination of model compression methods tailored for low-power microcontrollers for real-time weed detection in agriculture.

Findings

Achieved 51.8mJ energy per inference.

Maintained balanced detection accuracy with model compression.

Enabled real-time weed detection on microcontroller hardware.

Abstract

Weeds significantly reduce crop yields worldwide and pose major challenges to sustainable agriculture. Traditional weed management methods, primarily relying on chemical herbicides, risk environmental contamination and lead to the emergence of herbicide-resistant species. Precision weeding, leveraging computer vision and machine learning methods, offers a promising eco-friendly alternative but is often limited by reliance on high-power computational platforms. This work presents an optimized, low-power edge AI system for weeds detection based on the YOLOv8n object detector deployed on the STM32U575ZI microcontroller. Several compression techniques are applied to the detection model, including structured pruning, integer quantization and input image resolution scaling in order to meet strict hardware constraints. The model is trained and evaluated on the CropAndWeed dataset with 74 plant species, achieving a balanced trade-off between detection accuracy and efficiency. Our system supports real-time, in-situ weeds detection with a minimal energy consumption of 51.8mJ per inference, enabling scalable deployment in power-constrained agricultural environments.