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

• URL: http://arxiv.org/abs/2511.07990v1
• Date: Wed, 12 Nov 2025 01:32:42 GMT
• Title: Hardware-Aware YOLO Compression for Low-Power Edge AI on STM32U5 for Weeds Detection in Digital Agriculture
• Authors: Charalampos S. Kouzinopoulos, Yuri Manna,
• Abstract summary: This work presents an optimized, low-power edge AI system for weeds detection based on the YOLOv8n object detector deployed on the STM32U575ZI microcontroller.<n>Weeds significantly reduce crop yields worldwide and pose major challenges to sustainable agriculture.
• Score: 0.0
• License: http://creativecommons.org/licenses/by/4.0/
• 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.

Related papers

• Investigating Target Class Influence on Neural Network Compressibility for Energy-Autonomous Avian Monitoring [36.85680419418593]
  We propose running machine learning models on inexpensive microcontroller units (MCUs) directly in the field.<n>We trained and compressed models for various numbers of target classes to assess the detection of multiple bird species.<n>We also provide benchmarking results for different hardware platforms and evaluate the feasibility of deploying energy-autonomous devices.
  arXiv  Detail & Related papers  (2026-02-19T16:24:33Z)
• Dynamic Meta-Ensemble Framework for Efficient and Accurate Deep Learning in Plant Leaf Disease Detection on Resource-Constrained Edge Devices [0.0]
  We introduce a novel Dynamic Meta-Enemble Framework (DMEF) for high-accuracy plant disease diagnosis under resource constraints.<n>DMEF employs an adaptive weighting mechanism that dynamically combines the predictions of three lightweight convolutional neural networks.<n>Experiments on benchmark datasets for potato and maize diseases demonstrate state-of-the-art classification accuracies of 99.53% and 96.61%, respectively.
  arXiv  Detail & Related papers  (2026-01-24T03:57:49Z)
• PlantDiseaseNet-RT50: A Fine-tuned ResNet50 Architecture for High-Accuracy Plant Disease Detection Beyond Standard CNNs [0.10874100424278171]
  Plant diseases pose a significant threat to agricultural productivity and global food security, accounting for 70-80% of crop losses worldwide.<n>Traditional detection methods rely heavily on expert visual inspection, which is time-consuming, labour-intensive, and often impractical for large-scale farming operations.<n>We present PlantDiseaseNet-RT50, a novel fine-tuned deep learning architecture based on ResNet50 for automated plant disease detection.<n>Our model features strategically unfrozen layers, a custom classification head with regularization mechanisms, and dynamic learning rate scheduling through cosine decay.
  arXiv  Detail & Related papers  (2025-12-20T20:36:20Z)
• A Hybrid CNN-ViT-GNN Framework with GAN-Based Augmentation for Intelligent Weed Detection in Precision Agriculture [0.0]
  This paper proposes a hybrid deep learning framework recipe for weed detection.<n>A Generative Adversarial Network (GAN)-based augmentation method was imposed to balance class robustness and better generalize the model.<n> Experimental results yield superior results with 99.33% accuracy, precision, recall, and F1-score on multi-benchmark datasets.
  arXiv  Detail & Related papers  (2025-11-19T15:32:08Z)
• DTGen: Generative Diffusion-Based Few-Shot Data Augmentation for Fine-Grained Dirty Tableware Recognition [37.947945188644624]
  We propose DTGen, a few-shot data augmentation scheme for fine-grained dirty tableware recognition.<n>Under extremely limited real few-shot conditions, DTGen can synthesize virtually unlimited high-quality samples.<n> DTGen provides a feasible deployment path for automated tableware cleaning and food safety monitoring.
  arXiv  Detail & Related papers  (2025-09-15T07:59:34Z)
• Involution-Infused DenseNet with Two-Step Compression for Resource-Efficient Plant Disease Classification [0.0]
  This study proposes a two-step model compression approach integrating Weight Pruning and Knowledge Distillation.<n>The results demonstrate ResNet50s superior performance post-compression, achieving 99.55% and 98.99% accuracy on the PlantVillage and PaddyLeaf datasets.
  arXiv  Detail & Related papers  (2025-05-31T22:43:23Z)
• Mapping of Weed Management Methods in Orchards using Sentinel-2 and PlanetScope Data [36.205487938326556]
  We leverage earth observation data and Machine Learning (ML) to classify weed management methods in orchards.<n>The findings demonstrate the potential of ML-driven remote sensing to enhance the efficiency and accuracy of weed management mapping in orchards.
  arXiv  Detail & Related papers  (2025-04-28T17:09:10Z)
• Towards Efficient and Intelligent Laser Weeding: Method and Dataset for Weed Stem Detection [51.65457287518379]
  This study is the first empirical investigation of weed recognition for laser weeding.<n>We integrate the detection of crop and weed with the localization of weed stem into one end-to-end system.<n>The proposed system improves weeding accuracy by 6.7% and reduces energy cost by 32.3% compared to existing weed recognition systems.
  arXiv  Detail & Related papers  (2025-02-10T08:42:46Z)
• EcoWeedNet: A Lightweight and Automated Weed Detection Method for Sustainable Next-Generation Agricultural Consumer Electronics [7.900421718750201]
  EcoWeedNet is a novel model that enhances weed detection performance without introducing significant computational complexity.<n>Results indicate EcoWeedNet's deployability on low-power consumer hardware, lower energy consumption, and hence reduced carbon footprint.
  arXiv  Detail & Related papers  (2025-01-31T22:46:20Z)
• Deep-CNN based Robotic Multi-Class Under-Canopy Weed Control in Precision Farming [2.6085535710135654]
  Real-time multi-class weed identification enables species-specific treatment of weeds and significantly reduces the amount of herbicide use. Here, we present a baseline for classification performance using five benchmark CNN models. We deploy MobileNetV2 onto our own compact autonomous robot textitSAMBot for real-time weed detection.
  arXiv  Detail & Related papers  (2021-12-28T03:51:55Z)
• Automated Pest Detection with DNN on the Edge for Precision Agriculture [0.0]
  This paper presents an embedded system enhanced with machine learning (ML) functionalities, ensuring continuous detection of pest infestation inside fruit orchards. Three different ML algorithms have been trained and deployed, highlighting the capabilities of the platform. Results show how it is possible to automate the task of pest infestation for unlimited time without the farmer's intervention.
  arXiv  Detail & Related papers  (2021-08-01T10:17:48Z)
• Potato Crop Stress Identification in Aerial Images using Deep Learning-based Object Detection [60.83360138070649]
  The paper presents an approach for analyzing aerial images of a potato crop using deep neural networks. The main objective is to demonstrate automated spatial recognition of a healthy versus stressed crop at a plant level. Experimental validation demonstrated the ability for distinguishing healthy and stressed plants in field images, achieving an average Dice coefficient of 0.74.
  arXiv  Detail & Related papers  (2021-06-14T21:57:40Z)
• Estimating Crop Primary Productivity with Sentinel-2 and Landsat 8 using Machine Learning Methods Trained with Radiative Transfer Simulations [58.17039841385472]
  We take advantage of all parallel developments in mechanistic modeling and satellite data availability for advanced monitoring of crop productivity. Our model successfully estimates gross primary productivity across a variety of C3 crop types and environmental conditions even though it does not use any local information from the corresponding sites. This highlights its potential to map crop productivity from new satellite sensors at a global scale with the help of current Earth observation cloud computing platforms.
  arXiv  Detail & Related papers  (2020-12-07T16:23:13Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.