基于全卷积神经网络的荔枝表皮缺陷提取

S24%TP391.41; [目的]增强荔枝表皮缺陷提取效果,满足其品质检测分级准确性要求.[方法]采用Tensorflow框架构建基于AlexNet的全卷积神经网络AlexNet-FCN,以ReLU为激活函数,Max-pooling为下采样方法,Softmax回归分类器的损失函数作为优化目标,建立荔枝表皮缺陷提取的全卷积神经网络模型,并用批量随机梯度下降法对模型进行优化.[结果]模型收敛后在验证集上裂果交并比(IoUd)为0.83,褐变交并比(IoUb)为0.60,褐变与裂果的总体交并比(IoUa)为0.68;与利用线性SVM、朴素贝叶斯分类器缺陷提取效果相比,该模型的特征提取能力显著提高....

Full description

Saved in:
Bibliographic Details
Published in华南农业大学学报 Vol. 39; no. 6; pp. 104 - 110
Main Authors 王佳盛, 陈燕, 曾泽钦, 李嘉威, 刘威威, 邹湘军
Format Journal Article
LanguageChinese
Published 华南农业大学 工程学院/南方农业机械与装备关键技术教育部重点实验室,广东 广州,510642 01.11.2018
Subjects
荔枝
全卷积神经网络
缺陷提取
图像处理
深度学习
品质检测
Online AccessGet full text
ISSN1001-411X
DOI10.7671/j.issn.1001-411X.2018.06.016

Cover

Abstract S24%TP391.41; [目的]增强荔枝表皮缺陷提取效果,满足其品质检测分级准确性要求.[方法]采用Tensorflow框架构建基于AlexNet的全卷积神经网络AlexNet-FCN,以ReLU为激活函数,Max-pooling为下采样方法,Softmax回归分类器的损失函数作为优化目标,建立荔枝表皮缺陷提取的全卷积神经网络模型,并用批量随机梯度下降法对模型进行优化.[结果]模型收敛后在验证集上裂果交并比(IoUd)为0.83,褐变交并比(IoUb)为0.60,褐变与裂果的总体交并比(IoUa)为0.68;与利用线性SVM、朴素贝叶斯分类器缺陷提取效果相比,该模型的特征提取能力显著提高.[结论]全卷积神经网络在水果表面缺陷提取中具有良好的应用前景.
AbstractList S24%TP391.41; [目的]增强荔枝表皮缺陷提取效果,满足其品质检测分级准确性要求.[方法]采用Tensorflow框架构建基于AlexNet的全卷积神经网络AlexNet-FCN,以ReLU为激活函数,Max-pooling为下采样方法,Softmax回归分类器的损失函数作为优化目标,建立荔枝表皮缺陷提取的全卷积神经网络模型,并用批量随机梯度下降法对模型进行优化.[结果]模型收敛后在验证集上裂果交并比(IoUd)为0.83,褐变交并比(IoUb)为0.60,褐变与裂果的总体交并比(IoUa)为0.68;与利用线性SVM、朴素贝叶斯分类器缺陷提取效果相比,该模型的特征提取能力显著提高.[结论]全卷积神经网络在水果表面缺陷提取中具有良好的应用前景.
Author 邹湘军
王佳盛
李嘉威
曾泽钦
刘威威
陈燕
AuthorAffiliation 华南农业大学 工程学院/南方农业机械与装备关键技术教育部重点实验室,广东 广州,510642
AuthorAffiliation_xml – name: 华南农业大学 工程学院/南方农业机械与装备关键技术教育部重点实验室,广东 广州,510642
Author_FL ZOU Xiangjun
WANG Jiasheng
LIU Weiwei
CHEN Yan
LI Jiawei
ZENG Zeqin
Author_FL_xml – sequence: 1
  fullname: WANG Jiasheng
– sequence: 2
  fullname: CHEN Yan
– sequence: 3
  fullname: ZENG Zeqin
– sequence: 4
  fullname: LI Jiawei
– sequence: 5
  fullname: LIU Weiwei
– sequence: 6
  fullname: ZOU Xiangjun
Author_xml – sequence: 1
  fullname: 王佳盛
– sequence: 2
  fullname: 陈燕
– sequence: 3
  fullname: 曾泽钦
– sequence: 4
  fullname: 李嘉威
– sequence: 5
  fullname: 刘威威
– sequence: 6
  fullname: 邹湘军
BookMark eNo9jz9LAzEchjNUsNZ-CxeHi79czuQySvEfFFwU3Moll2iLpGAQ665jPQURLFpxEDvpWLjqt_Ga-i08UZxeeOF9Xp4FVLFdqxFaIoA542Slg9vOWUwASBARso9DIDEGhoGwCqr-9_Oo7lxbAoQQQTmsIlY85p_5ZXExKvpj__Lmn4d-kvmPaz-594PzWf9mOnyYPY384NW_519342l2VWS3i2jOJEdO1_-yhvY21ncbW0FzZ3O7sdYMHIGIBSrlnAmRAldKKAUJ5SEzUQqxMbBKgUotiUyUloprQymAiBVjTDIlUg2C1tDyL_c0sSaxB61O9-TYlo-tQ2vP0l5P_pgCKz3pN3NFYUc
ClassificationCodes S24%TP391.41
ContentType Journal Article
Copyright Copyright © Wanfang Data Co. Ltd. All Rights Reserved.
Copyright_xml – notice: Copyright © Wanfang Data Co. Ltd. All Rights Reserved.
DBID 2B.
4A8
92I
93N
PSX
TCJ
DOI 10.7671/j.issn.1001-411X.2018.06.016
DatabaseName Wanfang Data Journals - Hong Kong
WANFANG Data Centre
Wanfang Data Journals
万方数据期刊 - 香港版
China Online Journals (COJ)
China Online Journals (COJ)
DatabaseTitleList
DeliveryMethod fulltext_linktorsrc
Discipline Agriculture
DocumentTitle_FL Extraction of litchi fruit pericarp defect based on a fully convolutional neural network
EndPage 110
ExternalDocumentID hnnydxxb201806016
GrantInformation_xml – fundername: 国家重点研发计划; 国家自然科学基金; 广东省科技计划
  funderid: (2018YFD0101001); (31571568); (2015A020209120, 2015A020209111)
GroupedDBID -04
2B.
4A8
5XA
5XE
92G
92I
93N
ABJNI
ACGFS
ALMA_UNASSIGNED_HOLDINGS
CCEZO
CHDYS
CW9
DYU
ECGQY
OZF
PSX
TCJ
TGD
U1G
U5N
ID FETCH-LOGICAL-s1046-cd77699d07cc9cc0a3726f4d08ff05303beb1bacebc7ef330098c666b6c9de093
ISSN 1001-411X
IngestDate Thu May 29 04:04:40 EDT 2025
IsPeerReviewed false
IsScholarly false
Issue 6
Keywords 荔枝
全卷积神经网络
缺陷提取
图像处理
深度学习
品质检测
Language Chinese
LinkModel OpenURL
MergedId FETCHMERGED-LOGICAL-s1046-cd77699d07cc9cc0a3726f4d08ff05303beb1bacebc7ef330098c666b6c9de093
PageCount 7
ParticipantIDs wanfang_journals_hnnydxxb201806016
PublicationCentury 2000
PublicationDate 2018-11-01
PublicationDateYYYYMMDD 2018-11-01
PublicationDate_xml – month: 11
  year: 2018
  text: 2018-11-01
  day: 01
PublicationDecade 2010
PublicationTitle 华南农业大学学报
PublicationTitle_FL Journal of South China Agricultural University
PublicationYear 2018
Publisher 华南农业大学 工程学院/南方农业机械与装备关键技术教育部重点实验室,广东 广州,510642
Publisher_xml – name: 华南农业大学 工程学院/南方农业机械与装备关键技术教育部重点实验室,广东 广州,510642
SSID ssib002040671
ssib051370473
ssib002806938
ssib036437894
ssib005319396
ssj0048704
Score 1.8151202
Snippet S24%TP391.41; [目的]增强荔枝表皮缺陷提取效果,满足其品质检测分级准确性要求.[方法]采用Tensorflow框架构建基于AlexNet的全卷积神经网络AlexNet-FCN,以ReLU为激活函...
SourceID wanfang
SourceType Aggregation Database
StartPage 104
Title 基于全卷积神经网络的荔枝表皮缺陷提取
URI https://d.wanfangdata.com.cn/periodical/hnnydxxb201806016
Volume 39
hasFullText 1
inHoldings 1
isFullTextHit
isPrint
journalDatabaseRights – providerCode: PRVCAB
  databaseName: Nutrition and Food Sciences Database
  issn: 1001-411X
  databaseCode: DYU
  dateStart: 20100101
  customDbUrl:
  isFulltext: true
  dateEnd: 99991231
  titleUrlDefault: https://www.cabidigitallibrary.org/product/zd
  omitProxy: true
  ssIdentifier: ssj0048704
  providerName: CAB International
– providerCode: PRVAON
  databaseName: DOAJ Directory of Open Access Journals
  issn: 1001-411X
  databaseCode: DOA
  dateStart: 20160101
  customDbUrl:
  isFulltext: true
  dateEnd: 99991231
  titleUrlDefault: https://www.doaj.org/
  omitProxy: true
  ssIdentifier: ssj0048704
  providerName: Directory of Open Access Journals
link http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwnV3Na9RAFB9qBdGD-ImfpUhzKqlJJpmPY7KbpXjw1IV6Kslk055W6AfUnvVYqyCCRSsexJ70WGj1v3Gb-l_43tvZ3ViL1MISXt--ee_35k3nTbKTN4xNaZjw_CDI3VIY7oZFmbkasoZrorBUXGSQ02iX72Mx2w4fzUfzY-Pt2q6ltdV8xmyc-F7JWaIKPIgrviX7H5EdKgUG0BBfuEKE4XqqGDtp5OiWk8ROGuJVpchRkRMrIppOIp1UOjF8WkSAfIpEkjiKOEnT0b7l6AYSGvSETqqwuQZCYBPdRE7sk2aSift6GmRdO1qTLYFqtUfWgRD1ta-FZEGCckmEILuAX6Fa4MQhAkai6cSiRoBysBsNxgjaB2UqodbgKydo4EcyEiFkikAr-LLWWKBokiIBLaE9ygZg6A-RAV4N_aEJCxB-_VmJr-xLg4PRfTY_p5FKKEQYKEV-DV0nN2KsYltTChoBuz5JO_2ZEAGtEs9-hZggjBzHSB-BknbIYO9pDHic2p5WntWDYweICDHguACFAQorTsNBoxKAhF0cWDygBAcaYAY9DcsBcwHRIJO0Bmgb0zUWdYHGHbowi4uw9kwaq3i5oU8nHw2Tar9ClZ086hnSnjZtF1t-f0_z8TwuhezncTQwMzSAOzH75Xb9Y-XTaUG21O0-K9bXc5SiMkPn2PlACoEHmzSftEerUkhc9aqOuLlA19IQpiU-qkLJ8bdtNUpbkc-lF8rhgg9u-emU0iHMC2zK-vDwXx7Qa4LdMusu1la0c1fYZXsrOhn355WrbGxj6Rq7FC8u23I8netM9D7u_9x_2Xux29vcq758qz7vVAdb1Y_X1cH7avv50eabw50PR592q-2v1ff9X-_2Drde9bbe3mDtVjrXmHXtSSvuCu7xcE0hpdC68KQx2hgv4zIQZVh4qiyhOzyew5Iuz0wnN7JTco5ViI0QIhdGFx1P85tsvPu027nFJmHmz0vuZTLH8xJkmUnf-KVnTFSoQHj8NntgvV6wM-nKwl-Bu3Maobvs4uh__B4bX11e69yHO4TVfIKerE1Q1H8D5DnDEg
linkProvider CAB International
openUrl ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=%E5%9F%BA%E4%BA%8E%E5%85%A8%E5%8D%B7%E7%A7%AF%E7%A5%9E%E7%BB%8F%E7%BD%91%E7%BB%9C%E7%9A%84%E8%8D%94%E6%9E%9D%E8%A1%A8%E7%9A%AE%E7%BC%BA%E9%99%B7%E6%8F%90%E5%8F%96&rft.jtitle=%E5%8D%8E%E5%8D%97%E5%86%9C%E4%B8%9A%E5%A4%A7%E5%AD%A6%E5%AD%A6%E6%8A%A5&rft.au=%E7%8E%8B%E4%BD%B3%E7%9B%9B&rft.au=%E9%99%88%E7%87%95&rft.au=%E6%9B%BE%E6%B3%BD%E9%92%A6&rft.au=%E6%9D%8E%E5%98%89%E5%A8%81&rft.date=2018-11-01&rft.pub=%E5%8D%8E%E5%8D%97%E5%86%9C%E4%B8%9A%E5%A4%A7%E5%AD%A6+%E5%B7%A5%E7%A8%8B%E5%AD%A6%E9%99%A2%2F%E5%8D%97%E6%96%B9%E5%86%9C%E4%B8%9A%E6%9C%BA%E6%A2%B0%E4%B8%8E%E8%A3%85%E5%A4%87%E5%85%B3%E9%94%AE%E6%8A%80%E6%9C%AF%E6%95%99%E8%82%B2%E9%83%A8%E9%87%8D%E7%82%B9%E5%AE%9E%E9%AA%8C%E5%AE%A4%2C%E5%B9%BF%E4%B8%9C+%E5%B9%BF%E5%B7%9E%2C510642&rft.issn=1001-411X&rft.volume=39&rft.issue=6&rft.spage=104&rft.epage=110&rft_id=info:doi/10.7671%2Fj.issn.1001-411X.2018.06.016&rft.externalDocID=hnnydxxb201806016
thumbnail_s http://utb.summon.serialssolutions.com/2.0.0/image/custom?url=http%3A%2F%2Fwww.wanfangdata.com.cn%2Fimages%2FPeriodicalImages%2Fhnnydxxb%2Fhnnydxxb.jpg