{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,24]],"date-time":"2026-07-24T18:02:27Z","timestamp":1784916147078,"version":"3.55.0"},"reference-count":30,"publisher":"Public Library of Science (PLoS)","issue":"7","license":[{"start":{"date-parts":[[2026,7,24]],"date-time":"2026-07-24T00:00:00Z","timestamp":1784851200000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["www.plosone.org"],"crossmark-restriction":false},"short-container-title":["PLoS One"],"abstract":"<jats:p>To tackle the challenges in fine-grained retrieval stemming from noise, official seal occlusions, small text blocks, and other issues prevalent in archival text images, and to fulfill the requirements of integrating both textual and visual dual features while enhancing retrieval accuracy and efficiency, this study has devised a five-tier architectural model. This model comprises an input layer, a preprocessing layer, a scene graph generation layer, an attention fusion layer, and a retrieval matching layer. The model incorporates a dedicated scene graph generation module tailored for archival data, aiming to enhance element detection. Additionally, it features a three-tier attention fusion module that integrates scene graph, text, and cross-modal features to ensure precise feature alignment. Training is carried out using a multi-task loss function, and an index is created to streamline retrieval and matching processes. Experimental results show that the proposed model achieves a Top-1 accuracy of 83.7% and an average precision of 88.3% on the test set, representing a 25.1% improvement over the Top-1 accuracy of an optical character recognition (OCR) combined with word frequency and inverse document frequency model. The proposed model achieves a Top-1 accuracy of 6.1% higher than the archival retrieval network model for examples with official seal occlusion and a Top-1 accuracy of 76.8% for small text blocks. The response time for a single retrieval is 52.6ms. Research provides technical support for efficient retrieval of large-scale archives in archives, effectively solving the problem of archive retrieval in complex scenarios, and significantly improving the efficiency of archive management and utilization.<\/jats:p>","DOI":"10.1371\/journal.pone.0353505","type":"journal-article","created":{"date-parts":[[2026,7,24]],"date-time":"2026-07-24T17:35:51Z","timestamp":1784914551000},"page":"e0353505","update-policy":"https:\/\/doi.org\/10.1371\/journal.pone.corrections_policy","source":"Crossref","is-referenced-by-count":0,"title":["Construction of a fine-grained retrieval model for archival text-image based on the integration of scene graph generation and attention mechanism"],"prefix":"10.1371","volume":"21","author":[{"ORCID":"https:\/\/orcid.org\/0009-0006-6204-8693","authenticated-orcid":true,"given":"Mengyuan","family":"Zhang","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"340","published-online":{"date-parts":[[2026,7,24]]},"reference":[{"key":"pone.0353505.ref001","doi-asserted-by":"crossref","first-page":"1414","DOI":"10.1108\/GKMC-10-2022-0240","article-title":"Assessmenat of university archives management practices","volume":"74","author":"MS Bajwa","year":"2025","journal-title":"Glob Knowl Mem Commun"},{"issue":"1","key":"pone.0353505.ref002","doi-asserted-by":"crossref","first-page":"31","DOI":"10.59373\/kharisma.v4i1.64","article-title":"Optimizing digital archive management to improve the quality of integrated public services","volume":"4","author":"Mardhiyah","year":"2025","journal-title":"Kharisma"},{"issue":"1","key":"pone.0353505.ref003","doi-asserted-by":"crossref","first-page":"327","DOI":"10.1007\/s10278-024-01186-8","article-title":"Fine-tuned large language model for extracting patients on pretreatment for lung cancer from a picture archiving and communication system based on radiological reports","author":"K Yasaka","year":"2025","journal-title":"J Imaging Inform Med"},{"issue":"6","key":"pone.0353505.ref004","doi-asserted-by":"crossref","first-page":"4886","DOI":"10.1109\/LRA.2024.3384084","article-title":"Indoor and outdoor 3D scene graph generation via language-enabled spatial ontologies","author":"J Strader","year":"2024","journal-title":"IEEE Robot Autom Let"},{"issue":"1","key":"pone.0353505.ref005","doi-asserted-by":"crossref","first-page":"29","DOI":"10.1515\/pdtc-2023-0050","article-title":"Artificial intelligence\u2019s role in digitally preserving historic archives","volume":"53","author":"Z Teel","year":"2024","journal-title":"Preserv Digit Technol Cult"},{"issue":"5","key":"pone.0353505.ref006","doi-asserted-by":"crossref","first-page":"4035","DOI":"10.1609\/aaai.v38i5.28197","article-title":"SGFormer: semantic graph transformer for point cloud-based 3D scene graph generation","volume":"38","author":"C Lv","year":"2024","journal-title":"AAAI"},{"issue":"4","key":"pone.0353505.ref007","article-title":"Interactive content-based image retrieval with deep learning for CT abdominal organ recognition","volume":"69","author":"C-M Lo","year":"2024","journal-title":"Phys Med Biol"},{"issue":"3","key":"pone.0353505.ref008","doi-asserted-by":"crossref","first-page":"2607","DOI":"10.1007\/s11760-023-02934-z","article-title":"Improve the efficiency of handcrafted features in image retrieval by adding selected feature generating layers of deep convolutional neural networks","volume":"18","author":"G Shamsipour","year":"2024","journal-title":"Signal Image Video Process"},{"key":"pone.0353505.ref009","doi-asserted-by":"crossref","first-page":"17","DOI":"10.58496\/ADSA\/2024\/003","article-title":"Semantic image retrieval analysis based on deep learning and singular value decomposition","volume":"2024","author":"MH Hadid","year":"2024","journal-title":"Appl Data Sci Anal"},{"issue":"4","key":"pone.0353505.ref010","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1145\/3631129","article-title":"Probing historical image contexts: enhancing visual archive retrieval through computer vision","volume":"16","author":"L Du","year":"2023","journal-title":"J Comput Cult Herit"},{"issue":"5","key":"pone.0353505.ref011","first-page":"12230","article-title":"Cross-modality interaction reasoning for enhancing vision-language pre-training in image-text retrieval","author":"T Yao","journal-title":"Appl Intell"},{"issue":"3","key":"pone.0353505.ref012","doi-asserted-by":"crossref","first-page":"2337","DOI":"10.1109\/TMC.2023.3257268","article-title":"EDIR: Efficient distributed image retrieval of novel objects in mobile networks","author":"N Felemban","year":"2024","journal-title":"IEEE T Mobile Comput"},{"issue":"8","key":"pone.0353505.ref013","doi-asserted-by":"crossref","first-page":"567","DOI":"10.1080\/01431161.2023.2168136","article-title":"Cross-domain remote sensing image retrieval with Gabor-based CNN","author":"Z Zhuo","year":"2023","journal-title":"Int J Remote Sens"},{"issue":"1","key":"pone.0353505.ref014","first-page":"1","article-title":"Informative scene graph generation via debiasing","author":"L Gao","year":"2025","journal-title":"Int J Comput Vis"},{"issue":"4","key":"pone.0353505.ref015","first-page":"3657","article-title":"R3CD: scene graph to image generation with relation-aware compositional contrastive control diffusion","volume":"38","author":"J Liu","year":"2024","journal-title":"Proc AAAI Conf Artif Intell"},{"issue":"3","key":"pone.0353505.ref016","doi-asserted-by":"crossref","first-page":"626","DOI":"10.1177\/02666669241286224","article-title":"Developing smart archives in society 5.0: leveraging artificial intelligence for managing audiovisual archives in Africa","volume":"41","author":"I Schellnack-Kelly","year":"2024","journal-title":"Inform Dev"},{"issue":"3","key":"pone.0353505.ref017","article-title":"Modified LSHADE-SPACMA with new mutation strategy and external archive mechanism for numerical optimization and point cloud registration","volume":"58","author":"S Fu","year":"2025","journal-title":"Artif Intell Rev"},{"issue":"1","key":"pone.0353505.ref018","doi-asserted-by":"crossref","first-page":"403","DOI":"10.1007\/s41870-023-01552-9","article-title":"An efficient dimensionality reduction based on adaptive-GSM and transformer assisted classification for high dimensional data","volume":"16","author":"N Rajender","year":"2023","journal-title":"Int J Inf Tecnol"},{"issue":"2","key":"pone.0353505.ref019","doi-asserted-by":"crossref","first-page":"853","DOI":"10.1007\/s41870-023-01529-8","article-title":"A deep learning approach for image and text classification using neutrosophy","volume":"16","author":"MA Wajid","year":"2023","journal-title":"Int J Inf Tecnol"},{"issue":"3","key":"pone.0353505.ref020","doi-asserted-by":"crossref","first-page":"64","DOI":"10.1109\/MGRS.2024.3415391","article-title":"Federated learning across decentralized and unshared archives for remote sensing image classification: a review","volume":"12","author":"B B\u00fcy\u00fckta\u015f","year":"2024","journal-title":"IEEE Geosci Remote Sens Mag"},{"issue":"2","key":"pone.0353505.ref021","doi-asserted-by":"crossref","first-page":"4309","DOI":"10.1007\/s11042-023-15215-1","article-title":"Content-based image retrieval with fuzzy clustering for feature vector normalization","volume":"83","author":"V-H Vu","year":"2023","journal-title":"Multimed Tools Appl"},{"issue":"2","key":"pone.0353505.ref022","doi-asserted-by":"crossref","first-page":"605","DOI":"10.1007\/s40745-024-00525-4","article-title":"An improved boosting bald eagle search algorithm with improved African vultures optimization algorithm for data clustering","volume":"12","author":"FS Gharehchopogh","year":"2024","journal-title":"Ann Data Sci"},{"issue":"6","key":"pone.0353505.ref023","doi-asserted-by":"crossref","first-page":"3647","DOI":"10.1007\/s11831-024-10093-8","article-title":"A Comprehensive survey of multi-level thresholding segmentation methods for image processing","volume":"31","author":"M Amiriebrahimabadi","year":"2024","journal-title":"Arch Computat Methods Eng"},{"issue":"3","key":"pone.0353505.ref024","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1145\/3708350","article-title":"Scene adaptive context modeling and balanced relation prediction for scene graph generation","volume":"21","author":"K Xu","year":"2025","journal-title":"ACM Trans Multimedia Comput Commun Appl"},{"issue":"1","key":"pone.0353505.ref025","first-page":"1","article-title":"Dual-branch networks for privacy-preserving cross-modal retrieval in cloud computing","author":"J Peng","year":"2025","journal-title":"J Supercomput"},{"issue":"4","key":"pone.0353505.ref026","first-page":"3495","article-title":"TD\u00b2-Net: toward denoising and debiasing for video scene graph generation","volume":"38","author":"X Lin","year":"2024","journal-title":"Proc AAAI Conf Artif Intell"},{"issue":"4","key":"pone.0353505.ref027","doi-asserted-by":"crossref","first-page":"1405","DOI":"10.1109\/TC.2025.3525614","article-title":"Secure and efficient cross-modal retrieval over encrypted multimodal data","volume":"74","author":"L Yang","year":"2025","journal-title":"IEEE Trans Comput"},{"issue":"8","key":"pone.0353505.ref028","article-title":"Multi-dimensional feature collaborative fusion networks for multimodal fake news detection","author":"L Ji","year":"2025","journal-title":"J Supercomput"},{"issue":"1","key":"pone.0353505.ref029","first-page":"1","article-title":"Reliability-aware label distribution learning with attention-rectified for facial expression recognition","author":"L Peng","year":"2025","journal-title":"Appl Intell"},{"issue":"5","key":"pone.0353505.ref030","doi-asserted-by":"crossref","first-page":"3469","DOI":"10.1007\/s11276-020-02382-4","article-title":"Representation separation adversarial networks for cross-modal retrieval","volume":"30","author":"J Deng","year":"2020","journal-title":"Wireless Netw"}],"container-title":["PLOS One"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0353505","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,7,24]],"date-time":"2026-07-24T17:35:57Z","timestamp":1784914557000},"score":1,"resource":{"primary":{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0353505"}},"subtitle":[],"editor":[{"given":"Rehan","family":"Ashraf","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"editor"}]}],"short-title":[],"issued":{"date-parts":[[2026,7,24]]},"references-count":30,"journal-issue":{"issue":"7","published-online":{"date-parts":[[2026,7,24]]}},"URL":"https:\/\/doi.org\/10.1371\/journal.pone.0353505","relation":{},"ISSN":["1932-6203"],"issn-type":[{"value":"1932-6203","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,7,24]]}}}