Flora-YOLO: A lightweight framework for real-time rose grading on edge devices via structural optimization and feature refinement
- 1School of Airspace Science and Engineering, Shandong University, Weihai, Shandong, China
- 2School of Mathematics and Statistics, Shandong University, Weihai, Shandong, China
- 3Shandong Key Laboratory of Intelligent Electronic Packaging Testing and Application, Shandong University, Weihai, Shandong, China
Abstract
Precise flower grading in modern agriculture is critical for determining market value but remains hindered by inefficient manual labor and the computational constraints of edge devices. To address these challenges, this paper introduces Flora-YOLO, a lightweight and robust detection framework built upon the YOLO11n architecture. To overcome the inherent bottlenecks of agricultural vision, the framework incorporates four targeted optimizations, explicitly distinguishing novel components from adapted architectures: (1) an adapted C3k2-based Sandglass-Gated Bottleneck (C3k2_SGB) module to extract and amplify minute floral textures; (2) an integrated Adaptive Downsampling (ADown) module to preserve critical high-frequency spatial cues under severe foliage occlusion; (3) a newly proposed CSP-based Variance-Gated Attention (C2VGA) hybrid attention mechanism to amplify the weak semantic signals of early-stage buds while suppressing background noise; and (4) a novel Focused Shape-Intersection over Union (FSIoU) loss function to provide shape-aware gradient guidance for the precise bounding-box regression of irregular, non-rigid floral boundaries. Evaluated via 5-fold cross-validation on a custom dataset, the model achieves a stable 86.7% [email protected] and 50.4% [email protected]:0.95. This represents a 3.4% accuracy improvement over the baseline, achieved alongside a 28.6% reduction in computational load, requiring only 1.73 M parameters and 4.5 GFLOPs. Furthermore, out-of-domain testing and evaluation on public agricultural datasets confirm the model’s cross-species generalization. Finally, hardware deployment on both Android terminals and an NPU-accelerated NXP i.MX 8M Plus platform yields a real-time inference speed of 11.05 FPS with a 90.5 ms latency. These results establish Flora-YOLO as a highly efficient, accurate, and scalable solution for automated, edge-based sorting in precision floriculture.
Related
- FAnet: A lightweight end-to-end framework for Fructus Aurantii identification and quality grading
- Iterative Refinement for Structural Optimization of De Novo Protein Sequences
- AbRefine: Framework-Conditioned Refinement for Structure-Guided Antibody CDR Design
- TEFAR: a lightweight, configurable framework for semi-automatic artefact component classification in EEG and TMS-EEG
- X-SPEAR: An Extensible Framework for Spatial Panel Evaluation and Refinement
BibTeX
@article{liu2026florayolo,
title = {{Flora-YOLO: A lightweight framework for real-time rose grading on edge devices via structural optimization and feature refinement}},
author = {Wenwei Liu and Jinyu Xu and Guoao Wang and Shuo Ding and Yuanbo Zhang and Runze Tian and Zejiang Li and Yuanzhe Ji and Li Zhang},
journal = {PLOS One},
year = {2026},
month = sep,
doi = {10.1371/journal.pone.0358742},
url = {https://doi.org/10.1371/journal.pone.0358742},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - Flora-YOLO: A lightweight framework for real-time rose grading on edge devices via structural optimization and feature refinement
AU - Wenwei Liu
AU - Jinyu Xu
AU - Guoao Wang
AU - Shuo Ding
AU - Yuanbo Zhang
AU - Runze Tian
AU - Zejiang Li
AU - Yuanzhe Ji
AU - Li Zhang
PY - 2026
DA - 2026/09/21
JO - PLOS One
DO - 10.1371/journal.pone.0358742
UR - https://doi.org/10.1371/journal.pone.0358742
AB - Precise flower grading in modern agriculture is critical for determining market value but remains hindered by inefficient manual labor and the computational constraints of edge devices. To address these challenges, this paper introduces Flora-YOLO, a lightweight and robust detection framework built upon the YOLO11n architecture. To overcome the inherent bottlenecks of agricultural vision, the framework incorporates four targeted optimizations, explicitly distinguishing novel components from adapted architectures: (1) an adapted C3k2-based Sandglass-Gated Bottleneck (C3k2_SGB) module to extract and amplify minute floral textures; (2) an integrated Adaptive Downsampling (ADown) module to preserve critical high-frequency spatial cues under severe foliage occlusion; (3) a newly proposed CSP-based Variance-Gated Attention (C2VGA) hybrid attention mechanism to amplify the weak semantic signals of early-stage buds while suppressing background noise; and (4) a novel Focused Shape-Intersection over Union (FSIoU) loss function to provide shape-aware gradient guidance for the precise bounding-box regression of irregular, non-rigid floral boundaries. Evaluated via 5-fold cross-validation on a custom dataset, the model achieves a stable 86.7% [email protected] and 50.4% [email protected]:0.95. This represents a 3.4% accuracy improvement over the baseline, achieved alongside a 28.6% reduction in computational load, requiring only 1.73 M parameters and 4.5 GFLOPs. Furthermore, out-of-domain testing and evaluation on public agricultural datasets confirm the model’s cross-species generalization. Finally, hardware deployment on both Android terminals and an NPU-accelerated NXP i.MX 8M Plus platform yields a real-time inference speed of 11.05 FPS with a 90.5 ms latency. These results establish Flora-YOLO as a highly efficient, accurate, and scalable solution for automated, edge-based sorting in precision floriculture.
SN - 1932-6203
ER -