Combined query embroidery image retrieval based on enhanced CNN and blend transformer

Abstract Embroidery images carry rich historical information and are an important form of embroidery art. In the field of combination query image retrieval, how to efficiently retrieve the embroidery image information required by users has become a current research challenge. In recent years, convol...

Full description

Saved in:
Bibliographic Details
Main Authors: Xinzhen Zhuo, Donghai Huang, Yang Lin, Ziyang Huang
Format: Article
Language:English
Published: Nature Portfolio 2024-11-01
Series:Scientific Reports
Subjects:
Online Access:https://doi.org/10.1038/s41598-024-79012-y
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1846165407783714816
author Xinzhen Zhuo
Donghai Huang
Yang Lin
Ziyang Huang
author_facet Xinzhen Zhuo
Donghai Huang
Yang Lin
Ziyang Huang
author_sort Xinzhen Zhuo
collection DOAJ
description Abstract Embroidery images carry rich historical information and are an important form of embroidery art. In the field of combination query image retrieval, how to efficiently retrieve the embroidery image information required by users has become a current research challenge. In recent years, convolutional neural networks (CNNs) have achieved significant success in image feature extraction, but they tend to focus on local information, making it easy to ignore global context information when processing such textured embroidery images. Therefore, we propose a combination query retrieval method for embroidery images. First, we propose Blend-Transformer, which introduces Group External Attention (GEA). GEA can integrate feature information from three different dimensions, effectively capturing the local and global context information of embroidery images. Second, we propose Enhanced CNN, which introduces Shuffle Attention (SA), regrouping the reference image features extracted by CNN and reaggregating them by channel to enhance the richness of embroidery image feature information. Through experiments on the TCE-S and ICR2020 standard datasets, we verify the excellent performance of the proposed algorithm in embroidery image retrieval. Our method fills the gap in embroidery image retrieval research and provides a new perspective for the protection of embroidery art.
format Article
id doaj-art-7f7d0acfc4254b8ca2952cc945dd475b
institution Kabale University
issn 2045-2322
language English
publishDate 2024-11-01
publisher Nature Portfolio
record_format Article
series Scientific Reports
spelling doaj-art-7f7d0acfc4254b8ca2952cc945dd475b2024-11-17T12:18:41ZengNature PortfolioScientific Reports2045-23222024-11-0114111410.1038/s41598-024-79012-yCombined query embroidery image retrieval based on enhanced CNN and blend transformerXinzhen Zhuo0Donghai Huang1Yang Lin2Ziyang Huang3School of Design, Fujian University of TechnologySchool of Design, Fujian University of TechnologySchool of Design, Fujian University of TechnologyFaculty of Social Sciences, The University of SheffieldAbstract Embroidery images carry rich historical information and are an important form of embroidery art. In the field of combination query image retrieval, how to efficiently retrieve the embroidery image information required by users has become a current research challenge. In recent years, convolutional neural networks (CNNs) have achieved significant success in image feature extraction, but they tend to focus on local information, making it easy to ignore global context information when processing such textured embroidery images. Therefore, we propose a combination query retrieval method for embroidery images. First, we propose Blend-Transformer, which introduces Group External Attention (GEA). GEA can integrate feature information from three different dimensions, effectively capturing the local and global context information of embroidery images. Second, we propose Enhanced CNN, which introduces Shuffle Attention (SA), regrouping the reference image features extracted by CNN and reaggregating them by channel to enhance the richness of embroidery image feature information. Through experiments on the TCE-S and ICR2020 standard datasets, we verify the excellent performance of the proposed algorithm in embroidery image retrieval. Our method fills the gap in embroidery image retrieval research and provides a new perspective for the protection of embroidery art.https://doi.org/10.1038/s41598-024-79012-yEmbroidery imagesConvolutional neural networkImage retrievalTransformer
spellingShingle Xinzhen Zhuo
Donghai Huang
Yang Lin
Ziyang Huang
Combined query embroidery image retrieval based on enhanced CNN and blend transformer
Scientific Reports
Embroidery images
Convolutional neural network
Image retrieval
Transformer
title Combined query embroidery image retrieval based on enhanced CNN and blend transformer
title_full Combined query embroidery image retrieval based on enhanced CNN and blend transformer
title_fullStr Combined query embroidery image retrieval based on enhanced CNN and blend transformer
title_full_unstemmed Combined query embroidery image retrieval based on enhanced CNN and blend transformer
title_short Combined query embroidery image retrieval based on enhanced CNN and blend transformer
title_sort combined query embroidery image retrieval based on enhanced cnn and blend transformer
topic Embroidery images
Convolutional neural network
Image retrieval
Transformer
url https://doi.org/10.1038/s41598-024-79012-y
work_keys_str_mv AT xinzhenzhuo combinedqueryembroideryimageretrievalbasedonenhancedcnnandblendtransformer
AT donghaihuang combinedqueryembroideryimageretrievalbasedonenhancedcnnandblendtransformer
AT yanglin combinedqueryembroideryimageretrievalbasedonenhancedcnnandblendtransformer
AT ziyanghuang combinedqueryembroideryimageretrievalbasedonenhancedcnnandblendtransformer