智慧农业（中英文）2026，Vol.8Issue(4)：85-99,15.DOI:10.12133/j.smartag.SA202512007
基于Flor-YOLO的香石竹鲜切花分级轻量化检测方法
Lightweight Detection Method for Grading Fresh Cut Dianthus caryophyllus L.Based on Flor-YOLO
摘要
Abstract
[Objective]Carnation(Dianthus caryophyllus L.)is one of the most economically valuable cut flower crops worldwide.Post-harvest openness is a key quality indicator influencing pricing,logistics tolerance,and shelf life.However,manual grading is ineffi-cient and subjective due to dense petal overlap and complex edge structures.With the shift toward large-scale production and rising la-bor costs,accurate automated grading has become essential.Existing object detection models face a trade-off between computational efficiency and feature fidelity:High-precision architectures are computationally expensive for edge deployment,while lightweight models often lack sufficient feature representation.Additionally,conventional spatial downsampling introduces spectral aliasing,lead-ing to the loss of high-frequency petal texture information and limiting the separability of adjacent openness grades.Therefore,a light-weight yet detail-preserving detection framework is required.To address this need,Flor-YOLO(Flower openness recognition You On-ly Look Once)was proposed by integrating frequency-domain perception with structural re-parameterization for efficient and accurate carnation openness grading.[Methods]Based on the YOLO11n baseline,the Flor-YOLO architecture was proposed with targeted im-provements to the backbone,downsampling mechanism,and detection head.Backbone reconstruction:A lightweight LiteChimeraNet was constructed to enhance feature expression under limited computing power.A RepStem re-parameterization module was intro-duced at the input stage to establish an anti-aliasing mechanism via multi-branch training and single-path inference.Simultaneously,the C3k2_PConv module,utilizing partial convolution,was integrated to reduce memory access cost(MAC)and focus computation on petal foregrounds.Additionally,a RepNCSPELAN4_CAA module embedded with context anchor attention was incorporated in deep layers to capture long-range dependencies of the global flower topology.Frequency-domain downsampling:To mitigate texture aliasing and detail loss caused by spatial downsampling,a WaveletPool module was introduced.Utilizing the 2D discrete wavelet transform(2D-DWT),this module orthogonally decomposed feature maps into low-and high-frequency sub-bands,explicitly preserv-ing high-frequency information in horizontal,vertical,and diagonal directions to alleviate spectral aliasing.Detection head optimiza-tion:A lightweight shared detail-enhanced detection head(SDL-Head)was designed.It reduced parameter redundancy through cross-scale weight sharing and incorporated detail-enhanced convolution(DEConv),fusing central and angular difference operators,to boost sensitivity to the geometric morphology of petal edges.Furthermore,a scale-adaptive layer combined with Intersection over Union(IoU)-aware soft labels was applied to improve multi-scale feature alignment.A dataset comprising 1 748 original images of"Hongkang"carnations was collected and expanded to 6 580 samples via hybrid data augmentation.The model was trained on an NVIDIA RTX 4060 GPU for 250 epochs using SGD optimization,and comparative evaluations were conducted against the YOLO se-ries,NanoDet-m,and Hyper-YOLO-t.[Results and Discussions]Ablation studies and comparative experiments on the self-constructed dataset revealed significant performance gains.Ablation analysis:Reconstructing the backbone to LiteChimeraNet reduced floating point operations from 6.3 GFLOPs(baseline)to 1.5 GFLOPs,a decrease of 76.2%,while maintaining stable mean Average Precision(mAP@50),verifying its efficiency in removing background redundancy.Introducing WaveletPool significantly improved mAP@50 by 1.79 percentage points,confirming the critical role of explicitly preserving high-frequency components for serrated texture repre-sentation.Integrating SDL-Head further optimized feature alignment,increasing the recall rate to 94.47%.Overall performance:Flor-YOLO achieved a precision of 93.04%,recall of 94.47%,and mAP@50 of 96.10%.Compared to the YOLO11n baseline,these met-rics improved by 3.52,1.34,and 3.25 percentage points,respectively.Meanwhile,parameters and floating point operations were re-duced by 51.2%to 1.26 M and 1.1 GFLOPs(82.5%reduction).Flor-YOLO exhibited distinct advantages over YOLOv5n,YOLOv8n,YOLOv9t,YOLOv10n,and YOLOv12n in accuracy,mAP,and inference speed.Mechanism analysis:Spectral energy statistics showed that high-frequency energy intensified with increasing openness grades,aligning with the visual characteristics of petal expan-sion and wrinkle formation,thus validating the discriminative value of high-frequency information.Grad-CAM++visualizations fur-ther validated that the improved model stably focused on petal edges and flower centers,demonstrating superior robustness over the baseline in complex backgrounds.[Conclusions]By constructing the LiteChimeraNet backbone,incorporating frequency-domain downsampling,and designing a detail-enhanced head,the proposed model effectively enhances the representation of critical details such as petal edges and flower centers while maintaining extremely low computational costs.Comprehensively,Flor-YOLO achieves an optimal balance between accuracy,model size,and real-time performance,demonstrating strong potential for deployment on low-power mobile terminals and embedded sorting equipment.Furthermore,the proposed frequency-aware lightweight design paradigm provides a valuable reference for other agricultural vision tasks relying on subtle textural differences.关键词
香石竹分级/目标检测/YOLO11n/轻量化/小波池化/频域分析Key words
Dianthus caryophyllus L.grading/object detection/YOLO11n/lightweight/wavelet pooling/frequency-domain analysis分类
信息技术与安全科学引用本文复制引用
李传孟,杨洁,张晓宇..基于Flor-YOLO的香石竹鲜切花分级轻量化检测方法[J].智慧农业（中英文）,2026,8(4):85-99,15.基金项目
2025YS三区人才支持计划专项(09900/990025166) （09900/990025166）
云南省教育厅科学研究基金(0111723084,111724057) The 2025 YS Three-Area Talent Support Program(09900/990025166) （0111723084,111724057）
The Scientific Research Fund of Yunnan Pro-vincial Department of Education(0111723084,111724057) （0111723084,111724057）
