Suppr超能文献

基于深度学习的蜂群蜂蜜储存区域检测,通过线性回归预测蜂蜜物理参数

Deep Learning-Based Detection of Honey Storage Areas in Colonies for Predicting Physical Parameters of Honey via Linear Regression.

作者信息

Khokthong Watit, Kritangkoon Panpakorn, Sinpoo Chainarong, Takioawong Phuwasit, Phokasem Patcharin, Disayathanoowat Terd

机构信息

Department of Biology, Faculty of Science, Chiang Mai University, Chiang Mai 50200, Thailand.

Research Center of Deep Technology in Beekeeping and Bee Products for Sustainable Development Goals (SMART BEE SDGs), Chiang Mai University, Chiang Mai 50200, Thailand.

出版信息

Insects. 2025 May 29;16(6):575. doi: 10.3390/insects16060575.

Abstract

Traditional methods for assessing honey storage in beehives predominantly rely on manual visual inspection, which often leads to inconsistencies and inefficiencies. This study presents an automated deep learning approach utilizing the YOLOv11 model to detect, classify, and quantify honey cells within frames across monthly sampling periods. The model's performance varied depending on image resolution and dataset partitioning. Using the free version of YOLOv11 with high-resolution images (960 × 960 resolution) and a dataset split of 90:5:5 for training, validating, and testing, the model achieved a mean average precision at IoU threshold of 0.5 (mAP@0.5) of 83.4% for uncapped honey cells and 80.5% for capped honey cells. A strong correlation (r = 0.94) was observed between the 90:5:5 and 80:10:10 dataset splits, indicating that increasing the volume of training data enhances classification accuracy. In parallel, the study investigated the relationship between the physical properties of honey and image-based honey storage detection. Of the four tested properties, electrical conductivity (R = 0.19) and color (R = 0.21) showed weak predictive power for honey storage area estimation, with even weaker associations found for pH and moisture content. The honey storage areas via 90:5:5 and 80:10:10 datasets moderately correlated (r = 0.44-0.46) with increasing electrical conductivity and color. Especially, electrical conductivity exhibited statistically significant correlations with dataset performance across different dataset splits ( < 0.05), suggesting some potential influence of chemical composition on model accuracy. Our findings demonstrate the viability of image-based honey classification as a reliable technique for monitoring beehive productivity. Additionally, the research on image-based honey detection can be a non-invasive solution for improved honey production, beehive productivity, and optimized beekeeping practices.

摘要

传统的评估蜂箱中蜂蜜储存情况的方法主要依赖人工目视检查,这常常导致结果不一致且效率低下。本研究提出了一种利用YOLOv11模型的自动化深度学习方法,用于在每月采样期间的帧内检测、分类和量化蜂蜜巢室。模型的性能因图像分辨率和数据集划分而异。使用高分辨率图像(960×960分辨率)的免费版YOLOv11,并将数据集按90:5:5的比例划分为训练集、验证集和测试集,该模型在交并比(IoU)阈值为0.5时,未封盖蜂蜜巢室的平均精度均值(mAP@0.5)达到83.4%,封盖蜂蜜巢室的为80.5%。在90:5:5和80:10:10数据集划分之间观察到强相关性(r = 0.94),表明增加训练数据量可提高分类准确率。同时,该研究调查了蜂蜜的物理特性与基于图像的蜂蜜储存检测之间的关系。在测试的四个特性中,电导率(R = 0.19)和颜色(R = 0.21)对蜂蜜储存面积估计的预测能力较弱,而pH值和水分含量的相关性更弱。通过90:5:5和80:10:10数据集得出的蜂蜜储存面积与电导率和颜色增加呈中等程度相关(r = 0.44 - 0.46)。特别是,电导率在不同数据集划分中与数据集性能呈现出统计学上的显著相关性(<0.05),这表明化学成分对模型准确性有一定潜在影响。我们的研究结果证明了基于图像的蜂蜜分类作为监测蜂箱生产力的可靠技术的可行性。此外,基于图像的蜂蜜检测研究可以成为提高蜂蜜产量、蜂箱生产力和优化养蜂实践的非侵入性解决方案。

https://cdn.ncbi.nlm.nih.gov/pmc/blobs/10c2/12193066/aaf213364b0d/insects-16-00575-g006.jpg

相似文献

4
Cost-effectiveness of using prognostic information to select women with breast cancer for adjuvant systemic therapy.
Health Technol Assess. 2006 Sep;10(34):iii-iv, ix-xi, 1-204. doi: 10.3310/hta10340.
5
Artificial intelligence for detecting keratoconus.
Cochrane Database Syst Rev. 2023 Nov 15;11(11):CD014911. doi: 10.1002/14651858.CD014911.pub2.
6
Antiseptics for burns.
Cochrane Database Syst Rev. 2017 Jul 12;7(7):CD011821. doi: 10.1002/14651858.CD011821.pub2.
7
Stabilizing machine learning for reproducible and explainable results: A novel validation approach to subject-specific insights.
Comput Methods Programs Biomed. 2025 Jun 21;269:108899. doi: 10.1016/j.cmpb.2025.108899.
8
Intraoperative frozen section analysis for the diagnosis of early stage ovarian cancer in suspicious pelvic masses.
Cochrane Database Syst Rev. 2016 Mar 1;3(3):CD010360. doi: 10.1002/14651858.CD010360.pub2.
9
Systemic pharmacological treatments for chronic plaque psoriasis: a network meta-analysis.
Cochrane Database Syst Rev. 2021 Apr 19;4(4):CD011535. doi: 10.1002/14651858.CD011535.pub4.
10
Signs and symptoms to determine if a patient presenting in primary care or hospital outpatient settings has COVID-19.
Cochrane Database Syst Rev. 2022 May 20;5(5):CD013665. doi: 10.1002/14651858.CD013665.pub3.

本文引用的文献

5
Data Augmentation in Classification and Segmentation: A Survey and New Strategies.
J Imaging. 2023 Feb 17;9(2):46. doi: 10.3390/jimaging9020046.
6
How deep learning is empowering semantic segmentation: Traditional and deep learning techniques for semantic segmentation: A comparison.
Multimed Tools Appl. 2022;81(21):30519-30544. doi: 10.1007/s11042-022-12821-3. Epub 2022 Apr 6.
7
Quality of honeys from different botanical origins.
J Food Sci Technol. 2021 Nov;58(11):4167-4177. doi: 10.1007/s13197-020-04884-7. Epub 2020 Nov 6.
8
Overview of Bee Pollination and Its Economic Value for Crop Production.
Insects. 2021 Jul 31;12(8):688. doi: 10.3390/insects12080688.

文献AI研究员

20分钟写一篇综述,助力文献阅读效率提升50倍。

立即体验

用中文搜PubMed

大模型驱动的PubMed中文搜索引擎

马上搜索

文档翻译

学术文献翻译模型,支持多种主流文档格式。

立即体验