正态随机仿射变换的图像数据增强方法
TP391; 针对现有图像数据增强方法会生成大量无效冗余数据,导致训练数据质量降低和网络泛化性能减弱的问题,提出一种基于正态分布的随机仿射变换(random affine transformation based on normal distribution,NRAff)图像数据增强方法.NRAff的核心是设计一个正态随机仿射变换模块,在随机仿射变换中引入正态分布,使图像随机仿射变换幅度以原图像为中心呈正态分布形式输出,通过限制变换图像输出的分布范围,去除无效数据,获取更有效且具有正态分布特性的图像数据.NRAff方法仿照生物视觉感知系统的正态分布采样机制,使生成的图像分布接近生物视觉主观感知...
Saved in:
| Published in | 计算机工程与应用 Vol. 60; no. 23; pp. 176 - 186 |
|---|---|
| Main Authors | , , |
| Format | Journal Article |
| Language | Chinese |
| Published |
辽宁工程技术大学 软件学院,辽宁 葫芦岛 125105%光电信息控制和安全技术重点实验室,天津 300308
01.12.2024
|
| Subjects | |
| Online Access | Get full text |
| ISSN | 1002-8331 |
| DOI | 10.3778/j.issn.1002-8331.2307-0327 |
Cover
| Abstract | TP391; 针对现有图像数据增强方法会生成大量无效冗余数据,导致训练数据质量降低和网络泛化性能减弱的问题,提出一种基于正态分布的随机仿射变换(random affine transformation based on normal distribution,NRAff)图像数据增强方法.NRAff的核心是设计一个正态随机仿射变换模块,在随机仿射变换中引入正态分布,使图像随机仿射变换幅度以原图像为中心呈正态分布形式输出,通过限制变换图像输出的分布范围,去除无效数据,获取更有效且具有正态分布特性的图像数据.NRAff方法仿照生物视觉感知系统的正态分布采样机制,使生成的图像分布接近生物视觉主观感知效果,突出目标感知的正态分布特性,使网络在变换的特征中学习不变的特征.该方法能够提高图像数据分布的一致性,使网络学习到更多有效的、潜在的仿射变换不变特征,提高网络抗过拟合能力.在图像分类数据集CIFAR10,CIFAR100,SVHN,Fashion-MNIST和Imagenette上,与当前先进的数据增强方法进行实验和对比分析,实验结果表明,提出的图像增强方法在分类准确率上均有不同程度的提升,验证了NRAff方法的有效性和普适性. |
|---|---|
| AbstractList | TP391; 针对现有图像数据增强方法会生成大量无效冗余数据,导致训练数据质量降低和网络泛化性能减弱的问题,提出一种基于正态分布的随机仿射变换(random affine transformation based on normal distribution,NRAff)图像数据增强方法.NRAff的核心是设计一个正态随机仿射变换模块,在随机仿射变换中引入正态分布,使图像随机仿射变换幅度以原图像为中心呈正态分布形式输出,通过限制变换图像输出的分布范围,去除无效数据,获取更有效且具有正态分布特性的图像数据.NRAff方法仿照生物视觉感知系统的正态分布采样机制,使生成的图像分布接近生物视觉主观感知效果,突出目标感知的正态分布特性,使网络在变换的特征中学习不变的特征.该方法能够提高图像数据分布的一致性,使网络学习到更多有效的、潜在的仿射变换不变特征,提高网络抗过拟合能力.在图像分类数据集CIFAR10,CIFAR100,SVHN,Fashion-MNIST和Imagenette上,与当前先进的数据增强方法进行实验和对比分析,实验结果表明,提出的图像增强方法在分类准确率上均有不同程度的提升,验证了NRAff方法的有效性和普适性. |
| Abstract_FL | In view of the fact that the existing image data augmentation method generates a large amount of invalid and redundant data,resulting in reduced training data quality and weakened network generalization performance,a random affine transformation based on normal distribution(NRAff)image data enhancement method based on normal distribution is proposed.The core of NRAff is to design a normal random affine transformation module,which introduces normal distribution in the random affine transformation,so that the amplitude of the random affine transformation of the image is output in the form of normal distribution centered on the original image,and the image data with normal distribution characteristics is obtained by limiting the distribution range of the transformed image output,removing invalid data,and obtaining more efficient image data with normal distribution characteristics.The NRAff method imitates the normal distri-bution sampling mechanism of the biological visual perception system,so that the generated image distribution is close to the subjective perception effect of biological vision,highlights the normal distribution characteristics of target perception,and enables the network to learn unchanged features in the transformed features.This method can improve the consistency of image data distribution,enable the network to learn more effective and potential affine transformation invariant features,and improve the network resistance to overfitting.Experiments and comparative analysis are carried out on the image clas-sification dataset CIFAR10,CIFAR100,SVHN,Fashion-MNIST and Imagenette,and the experimental results show that the proposed image enhancement method has different degrees of improvement in classification accuracy,which verifies the effectiveness and universality of the NRAff method. |
| Author | 张晟翀 姜文涛 陈霖霖 |
| AuthorAffiliation | 辽宁工程技术大学 软件学院,辽宁 葫芦岛 125105%光电信息控制和安全技术重点实验室,天津 300308 |
| AuthorAffiliation_xml | – name: 辽宁工程技术大学 软件学院,辽宁 葫芦岛 125105%光电信息控制和安全技术重点实验室,天津 300308 |
| Author_FL | JIANG Wentao CHEN Linlin ZHANG Shengchong |
| Author_FL_xml | – sequence: 1 fullname: JIANG Wentao – sequence: 2 fullname: CHEN Linlin – sequence: 3 fullname: ZHANG Shengchong |
| Author_xml | – sequence: 1 fullname: 姜文涛 – sequence: 2 fullname: 陈霖霖 – sequence: 3 fullname: 张晟翀 |
| BookMark | eNo9jz9Lw0AYh2-oYK39Em4OiXf3Xu69jFL8BwUXncsluZQGuYKHSDbFriVTRao41FkpOFlFv4w58VtYUJx-8AzPw2-NNOzQGkI2GA0BUW0V4cA5GzJKeaAAWMiBYkCBY4M0_-kqaTs3SGjEACOEuEnAPz74i8vvaeXvFp-vH_V8VFc3fjz7mo7q2_f6qvKTuR8_1bP7-m3hr1_882SdrOT6xJn237bI8e7OUWc_6B7uHXS2u4FjVIrAJFJJbkSUyURJneQS0xSB8dhwCVEmOOeCKsmMVDlqzDnNME4pMoVKYwQtsvnrPdc217bfK4Znp3ZZ7BWu6KdlWXLKxfIoE_ADU5dcrw |
| ClassificationCodes | TP391 |
| 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 | 10.3778/j.issn.1002-8331.2307-0327 |
| 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 | Engineering |
| DocumentTitle_FL | Image Data Augmentation Method for Normal Random Affine Transformation |
| EndPage | 186 |
| ExternalDocumentID | jsjgcyyy202423014 |
| GrantInformation_xml | – fundername: 国家部委预研基金; (辽宁省自然科学基金); (辽宁省教育厅重点基金) funderid: 国家部委预研基金; (辽宁省自然科学基金); (辽宁省教育厅重点基金) |
| GroupedDBID | -0Y 2B. 4A8 5XA 5XJ 92H 92I 93N ABJNI ACGFS ALMA_UNASSIGNED_HOLDINGS CCEZO CUBFJ CW9 PSX TCJ TGT U1G U5S |
| ID | FETCH-LOGICAL-s1064-eb6862e45d6b86abf67cc73129e2635d422240861e68f7a7f20d79c071878a753 |
| ISSN | 1002-8331 |
| IngestDate | Thu May 29 04:10:55 EDT 2025 |
| IsPeerReviewed | false |
| IsScholarly | false |
| Issue | 23 |
| Keywords | 仿射变换 图像分类 data augmentation image classification 数据增强 正态分布 normal distribution affine transformation |
| Language | Chinese |
| LinkModel | OpenURL |
| MergedId | FETCHMERGED-LOGICAL-s1064-eb6862e45d6b86abf67cc73129e2635d422240861e68f7a7f20d79c071878a753 |
| PageCount | 11 |
| ParticipantIDs | wanfang_journals_jsjgcyyy202423014 |
| PublicationCentury | 2000 |
| PublicationDate | 2024-12-01 |
| PublicationDateYYYYMMDD | 2024-12-01 |
| PublicationDate_xml | – month: 12 year: 2024 text: 2024-12-01 day: 01 |
| PublicationDecade | 2020 |
| PublicationTitle | 计算机工程与应用 |
| PublicationTitle_FL | Computer Engineering and Applications |
| PublicationYear | 2024 |
| Publisher | 辽宁工程技术大学 软件学院,辽宁 葫芦岛 125105%光电信息控制和安全技术重点实验室,天津 300308 |
| Publisher_xml | – name: 辽宁工程技术大学 软件学院,辽宁 葫芦岛 125105%光电信息控制和安全技术重点实验室,天津 300308 |
| SSID | ssib051375739 ssib001102935 ssj0000561668 ssib023646291 ssib057620132 |
| Score | 2.002351 |
| Snippet | TP391; 针对现有图像数据增强方法会生成大量无效冗余数据,导致训练数据质量降低和网络泛化性能减弱的问题,提出一种基于正态分布的随机仿射变换(random affine transformation based on normal... |
| SourceID | wanfang |
| SourceType | Aggregation Database |
| StartPage | 176 |
| Title | 正态随机仿射变换的图像数据增强方法 |
| URI | https://d.wanfangdata.com.cn/periodical/jsjgcyyy202423014 |
| Volume | 60 |
| hasFullText | 1 |
| inHoldings | 1 |
| isFullTextHit | |
| isPrint | |
| link | http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwnR3LahRBcIibix7EJ74JYp_Cxp2e6dexe3eWIMaLCeQW5rVKDiuY5JCcFHMNOUUkiod4VgKejKI_4474F1ZX92SHJEIUlqHorldX9XZXD1PVQXAvSjsDyUTazgUdtGPBB-2MpgU8bPH1AtZNZhOF5x7x2YX4wSJbnGipxldLa6vZTL5xYl7J_3gV2sCvNkv2Hzx7yBQaAAb_whM8DM9T-ZgknOge0ZEFZAeiQpIoojSRfduiusRoksTEGGKghREDOLEFAEFJpAJyShKBVNilADlBnKjmg4QeGbuARCFgnAjA4cQoC5iIuPss65CXJNJS6dBKAUCJhm7AARoZdkkiDWoLgGOuiYpRtxh666mB8oVl4MRK5GcANmMUsIIiUiLQtVg10OQCymscF-CqvhUEVnI3PtUvQmh85KMSOxowj-mhGgma_PggQCttHeIGqtH2OrZqWwCsyKeRUQ_7nIP4YZ9XH4xMu8flIaUKiTYWADkaKQ24xEzbQLLDCGXoQPgpbz_DUE4f_cCt2hp9C5Z2WoGxnAqKEtmtpSnPR8uTBqWs8WFSgAhJ0f9IZacGqK_HfGDstFsbwU0TANh0hIWMGhsj7px1dp3fOd1NEH6FoFFjHwwFb4RUoat2fnS3joSQuFtbCTOHEjA3od2JXM2II9XQl1eWn-Tr6-t2AlD7OuBMMEkFhM2tYFL35h4-Hp8FIHRW47OAvSiB03FhKBZGgolxSVw4d9OOL2PqLwbgIfeptF4zV7TYqn3_70pj9t9wkA6fNALV-QvBeX_CnNJuubgYTGw8vRSca9QdvRxE1ccP1YuXv3e3q3cHP7_-GO1vjrbfVFt7v3Y3R2-_j15tVzv71dan0d770beD6vWX6vPOlWChn8x3Z9v-9pT2SgjnjHaZ2eSvMmYFzyRPswEXeS4iiO9LW4CqsO9-447kYcnlQKRiQDuFUDkcOaSQqWDR1aA1fDYsrwVTIhJZKXhWCJ7Hac6lLCkrRcFYFBcsldeDu37IS351XFk65qgbp0G6GZwd_7VvBa3V52vlbYj6V7M73r9_AAzZtKE |
| linkProvider | EBSCOhost |
| 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=%E6%AD%A3%E6%80%81%E9%9A%8F%E6%9C%BA%E4%BB%BF%E5%B0%84%E5%8F%98%E6%8D%A2%E7%9A%84%E5%9B%BE%E5%83%8F%E6%95%B0%E6%8D%AE%E5%A2%9E%E5%BC%BA%E6%96%B9%E6%B3%95&rft.jtitle=%E8%AE%A1%E7%AE%97%E6%9C%BA%E5%B7%A5%E7%A8%8B%E4%B8%8E%E5%BA%94%E7%94%A8&rft.au=%E5%A7%9C%E6%96%87%E6%B6%9B&rft.au=%E9%99%88%E9%9C%96%E9%9C%96&rft.au=%E5%BC%A0%E6%99%9F%E7%BF%80&rft.date=2024-12-01&rft.pub=%E8%BE%BD%E5%AE%81%E5%B7%A5%E7%A8%8B%E6%8A%80%E6%9C%AF%E5%A4%A7%E5%AD%A6+%E8%BD%AF%E4%BB%B6%E5%AD%A6%E9%99%A2%2C%E8%BE%BD%E5%AE%81+%E8%91%AB%E8%8A%A6%E5%B2%9B+125105%25%E5%85%89%E7%94%B5%E4%BF%A1%E6%81%AF%E6%8E%A7%E5%88%B6%E5%92%8C%E5%AE%89%E5%85%A8%E6%8A%80%E6%9C%AF%E9%87%8D%E7%82%B9%E5%AE%9E%E9%AA%8C%E5%AE%A4%2C%E5%A4%A9%E6%B4%A5+300308&rft.issn=1002-8331&rft.volume=60&rft.issue=23&rft.spage=176&rft.epage=186&rft_id=info:doi/10.3778%2Fj.issn.1002-8331.2307-0327&rft.externalDocID=jsjgcyyy202423014 |
| thumbnail_s | http://utb.summon.serialssolutions.com/2.0.0/image/custom?url=http%3A%2F%2Fwww.wanfangdata.com.cn%2Fimages%2FPeriodicalImages%2Fjsjgcyyy%2Fjsjgcyyy.jpg |