A multi-source feature stable learning method for rapid identification of cork spot disorder in ‘Akizuki’ pear

IF 6.4 1区 农林科学 Q1 AGRONOMY Postharvest Biology and Technology Pub Date : 2024-10-30 DOI:10.1016/j.postharvbio.2024.113285
Jianghui Xiong , Shangfeng Gu , Yuan Rao , Li Liu , Xiaodan Zhang , Yuting Wu , Xiu Jin
{"title":"A multi-source feature stable learning method for rapid identification of cork spot disorder in ‘Akizuki’ pear","authors":"Jianghui Xiong ,&nbsp;Shangfeng Gu ,&nbsp;Yuan Rao ,&nbsp;Li Liu ,&nbsp;Xiaodan Zhang ,&nbsp;Yuting Wu ,&nbsp;Xiu Jin","doi":"10.1016/j.postharvbio.2024.113285","DOIUrl":null,"url":null,"abstract":"<div><div>The quality grading and sorting of ‘Akizuki’ pear, a high-quality fruit, is a vital part of the supply chain. Meanwhile, cork spot disorder, a common physiological issue that affects the healthy development of pear fruit, requires rapid and accurate identification. To further enhance the identification precision of cork spot disorder in ‘Akizuki’ pear, this study proposes a multi-source feature stable learning (MFSL) method based on neural architecture search and sample reweighting techniques, building upon the fusion of near-infrared spectrum and visual image. This method employs secondary reweighted optimization training on a multi-source fusion model, enabling it to fully learn label-related features and thereby enhance generalization. Experimental results show that the optimal modelling performance of the multi-source fusion feature has increased by 26.89 % in accuracy compared to the single spectrum and by 11.19 % compared to the single image. After optimization training, the testing accuracy of the model improved by 1.31 %, reaching 89.47 %, and the F1-score increased by 1.47 %, reaching 89.83 %. The results validate the effectiveness of the method in enhancing the model’s generalization performance. The proposed MFSL method for the precise identification of cork spot disorder in ‘Akizuki’ pear in this study significantly improves the accuracy of the multi-source fusion model in symptom recognition. The research results have important reference value for the efficient grading and sorting of pear fruit quality.</div></div>","PeriodicalId":20328,"journal":{"name":"Postharvest Biology and Technology","volume":"219 ","pages":"Article 113285"},"PeriodicalIF":6.4000,"publicationDate":"2024-10-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Postharvest Biology and Technology","FirstCategoryId":"97","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S0925521424005301","RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"AGRONOMY","Score":null,"Total":0}
引用次数: 0

Abstract

The quality grading and sorting of ‘Akizuki’ pear, a high-quality fruit, is a vital part of the supply chain. Meanwhile, cork spot disorder, a common physiological issue that affects the healthy development of pear fruit, requires rapid and accurate identification. To further enhance the identification precision of cork spot disorder in ‘Akizuki’ pear, this study proposes a multi-source feature stable learning (MFSL) method based on neural architecture search and sample reweighting techniques, building upon the fusion of near-infrared spectrum and visual image. This method employs secondary reweighted optimization training on a multi-source fusion model, enabling it to fully learn label-related features and thereby enhance generalization. Experimental results show that the optimal modelling performance of the multi-source fusion feature has increased by 26.89 % in accuracy compared to the single spectrum and by 11.19 % compared to the single image. After optimization training, the testing accuracy of the model improved by 1.31 %, reaching 89.47 %, and the F1-score increased by 1.47 %, reaching 89.83 %. The results validate the effectiveness of the method in enhancing the model’s generalization performance. The proposed MFSL method for the precise identification of cork spot disorder in ‘Akizuki’ pear in this study significantly improves the accuracy of the multi-source fusion model in symptom recognition. The research results have important reference value for the efficient grading and sorting of pear fruit quality.
查看原文
分享 分享
微信好友 朋友圈 QQ好友 复制链接
本刊更多论文
一种多源特征稳定学习法,用于快速识别 "秋月 "梨的软木塞斑病
优质水果'秋月'梨的质量分级和分类是供应链的重要组成部分。同时,木栓斑病是影响梨果健康发育的常见生理问题,需要快速准确地识别。为了进一步提高'秋月'梨软木塞斑病的识别精度,本研究在融合近红外光谱和视觉图像的基础上,提出了一种基于神经架构搜索和样本重权技术的多源特征稳定学习(MFSL)方法。该方法对多源融合模型进行二次重权优化训练,使其能够充分学习与标签相关的特征,从而增强泛化能力。实验结果表明,与单一光谱相比,多源融合特征的最佳建模性能提高了 26.89%,与单一图像相比,提高了 11.19%。经过优化训练后,模型的测试准确率提高了 1.31%,达到 89.47%,F1 分数提高了 1.47%,达到 89.83%。结果验证了该方法在提高模型泛化性能方面的有效性。本研究提出的用于精确识别秋月梨软木斑病的 MFSL 方法显著提高了多源融合模型在症状识别中的准确性。研究成果对梨果实质量的有效分级和分类具有重要的参考价值。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 去求助
来源期刊
Postharvest Biology and Technology
Postharvest Biology and Technology 农林科学-农艺学
CiteScore
12.00
自引率
11.40%
发文量
309
审稿时长
38 days
期刊介绍: The journal is devoted exclusively to the publication of original papers, review articles and frontiers articles on biological and technological postharvest research. This includes the areas of postharvest storage, treatments and underpinning mechanisms, quality evaluation, packaging, handling and distribution of fresh horticultural crops including fruit, vegetables, flowers and nuts, but excluding grains, seeds and forages. Papers reporting novel insights from fundamental and interdisciplinary research will be particularly encouraged. These disciplines include systems biology, bioinformatics, entomology, plant physiology, plant pathology, (bio)chemistry, engineering, modelling, and technologies for nondestructive testing. Manuscripts on fresh food crops that will be further processed after postharvest storage, or on food processes beyond refrigeration, packaging and minimal processing will not be considered.
期刊最新文献
Flower senescence: A comprehensive update on hormonal regulation and molecular aspects of petal death Phenylpropanoid pathway mediated the defense response of ‘Korla’ fragrant pear against Alternaria alternata infection Hydrogen sulfide enhances the disease resistance of ginger to rhizome rot during postharvest storage through modulation of antioxidant response and nitric oxide-mediated S-nitrosylaion Gypenoside GP5 effectively controls Colletotrichum gloeosporioides, an anthracnose fungus, by activating autophagy A comprehensive transcriptomic and metabolomic map reveals the molecular mechanism of persimmon fruit deastringency upon 40 °C warm water treatment
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
现在去查看 取消
×
提示
确定
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
已复制链接
已复制链接
快去分享给好友吧!
我知道了
×
扫码分享
扫码分享
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术
文献互助 智能选刊 最新文献 互助须知 联系我们:info@booksci.cn
Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。
Copyright © 2023 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1