{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,14]],"date-time":"2026-05-14T18:13:16Z","timestamp":1778782396232,"version":"3.51.4"},"reference-count":34,"publisher":"Public Library of Science (PLoS)","issue":"5","license":[{"start":{"date-parts":[[2026,5,14]],"date-time":"2026-05-14T00:00:00Z","timestamp":1778716800000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["62166035"],"award-info":[{"award-number":["62166035"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100012226","name":"Fundamental Research Funds for the Central Universities","doi-asserted-by":"publisher","award":["31920240091"],"award-info":[{"award-number":["31920240091"]}],"id":[{"id":"10.13039\/501100012226","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100004775","name":"Natural Science Foundation of Gansu Province","doi-asserted-by":"publisher","award":["25JRRA991"],"award-info":[{"award-number":["25JRRA991"]}],"id":[{"id":"10.13039\/501100004775","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["www.plosone.org"],"crossmark-restriction":false},"short-container-title":["PLoS One"],"abstract":"<jats:p>In the digital preservation, restoration, and research of Thangka paintings, real-time semantic segmentation plays a crucial role in rapid image analysis. However, Thangka images exhibit intricate compositions, where principal figures often blend with backgrounds, ritual objects, and intricate ornaments, leading to blurred boundaries and fine details that challenge conventional segmentation methods in balancing accuracy and efficiency. To address this, we propose an improved PIDNet-based model incorporating ECA-Pag (Efficient Channel Attention-Path Aggregation) and LGFM (Local-Global Feature Fusion Module) modules, along with PConv3 (Partial Convolution 3), enhancing feature extraction and segmentation precision. Experimental results demonstrate that our model achieves 73.28% mIoU and mB-Fscore (mean Boundary F-score) of 40.01% on a custom Thangka dataset while maintaining 109.03 FPS (Frames Per Second), ensuring both high accuracy and real-time performance. Furthermore, evaluations on the Cityscapes benchmark confirm the model\u2019s generalization capability, outperforming baseline methods. This work provides an efficient and reliable solution for Thangka image segmentation, with potential applications in cultural heritage preservation and broader computer vision tasks.<\/jats:p>","DOI":"10.1371\/journal.pone.0332164","type":"journal-article","created":{"date-parts":[[2026,5,14]],"date-time":"2026-05-14T17:40:33Z","timestamp":1778780433000},"page":"e0332164","update-policy":"https:\/\/doi.org\/10.1371\/journal.pone.corrections_policy","source":"Crossref","is-referenced-by-count":0,"title":["Research on the semantic segmentation of Thangka images via an improved PIDNet"],"prefix":"10.1371","volume":"21","author":[{"given":"Jiao","family":"Wu","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0009-0009-7113-179X","authenticated-orcid":true,"given":"Tiejun","family":"Wang","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Xiaoyan","family":"Hu","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"LingMei","family":"Tao","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Tianjiao","family":"Duan","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Yanjiao","family":"Wei","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"340","published-online":{"date-parts":[[2026,5,14]]},"reference":[{"key":"pone.0332164.ref001","volume-title":"Tibetan thangka painting: methods and materials","author":"DP Jackson","year":"1988"},{"issue":"31","key":"pone.0332164.ref002","doi-asserted-by":"crossref","first-page":"23103","DOI":"10.1007\/s00521-023-08957-4","article-title":"Deep learning: systematic review, models, challenges, and research directions","volume":"35","author":"T Talaei Khoei","year":"2023","journal-title":"Neural Comput & Applic"},{"issue":"3","key":"pone.0332164.ref003","doi-asserted-by":"crossref","first-page":"3448","DOI":"10.1109\/TITS.2022.3228042","article-title":"Deep Dual-Resolution Networks for Real-Time and Accurate Semantic Segmentation of Traffic Scenes","volume":"24","author":"H Pan","year":"2023","journal-title":"IEEE Trans Intell Transport Syst"},{"key":"pone.0332164.ref004","doi-asserted-by":"crossref","unstructured":"Fan M, Lai S, Huang J, Wei X, Chai Z, Luo J, et al. Rethinking BiSeNet For Real-time Semantic Segmentation. In: 2021 IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR). 2021; 9711\u201320. https:\/\/doi.org\/10.1109\/cvpr46437.2021.00959","DOI":"10.1109\/CVPR46437.2021.00959"},{"key":"pone.0332164.ref005","doi-asserted-by":"crossref","unstructured":"Xu J, Xiong Z, Bhattacharyya SP. PIDNet: A Real-time Semantic Segmentation Network Inspired by PID Controllers. In: 2023 IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2023. 19529\u201339. https:\/\/doi.org\/10.1109\/cvpr52729.2023.01871","DOI":"10.1109\/CVPR52729.2023.01871"},{"issue":"13","key":"pone.0332164.ref006","doi-asserted-by":"crossref","first-page":"2533","DOI":"10.3390\/electronics14132533","article-title":"Thangka Element Semantic Segmentation with an Integrated Multi-Scale Attention Mechanism","volume":"14","author":"T Wang","year":"2025","journal-title":"Electronics"},{"issue":"1","key":"pone.0332164.ref007","doi-asserted-by":"crossref","first-page":"11","DOI":"10.1007\/s13735-023-00313-5","article-title":"A new CNN-based semantic object segmentation for autonomous vehicles in urban traffic scenes","volume":"13","author":"G Do\u011fan","year":"2024","journal-title":"Int J Multimed Inf Retr"},{"issue":"1","key":"pone.0332164.ref008","doi-asserted-by":"crossref","first-page":"366","DOI":"10.1007\/s12559-023-10206-8","article-title":"Attention-guided multi-scale fusion network for similar objects semantic segmentation","volume":"16","author":"F Yao","year":"2024","journal-title":"Cogn Comput"},{"issue":"5","key":"pone.0332164.ref009","doi-asserted-by":"crossref","first-page":"905","DOI":"10.1109\/LGRS.2020.2988294","article-title":"SCAttNet: semantic segmentation network with spatial and channel attention mechanism for high-resolution remote sensing images","volume":"18","author":"H Li","year":"2021","journal-title":"IEEE Geosci Remote Sensing Lett"},{"issue":"12","key":"pone.0332164.ref010","doi-asserted-by":"crossref","first-page":"4477","DOI":"10.3390\/s22124477","article-title":"Multiple-Attention Mechanism Network for Semantic Segmentation","volume":"22","author":"D Wang","year":"2022","journal-title":"Sensors (Basel)"},{"key":"pone.0332164.ref011","doi-asserted-by":"crossref","first-page":"107260","DOI":"10.1016\/j.engappai.2023.107260","article-title":"Image semantic segmentation approach based on DeepLabV3 plus network with an attention mechanism","volume":"127","author":"Y Liu","year":"2024","journal-title":"Eng Appl Artif Intell"},{"issue":"1","key":"pone.0332164.ref012","doi-asserted-by":"crossref","first-page":"23018","DOI":"10.1038\/s41598-025-07806-9","article-title":"Multi-scale fusion semantic enhancement network for medical image segmentation","volume":"15","author":"Z Zhang","year":"2025","journal-title":"Sci Rep"},{"issue":"2","key":"pone.0332164.ref013","doi-asserted-by":"crossref","first-page":"2557","DOI":"10.1007\/s40747-023-01279-x","article-title":"Enhanced multi-scale networks for semantic segmentation","volume":"10","author":"T Li","year":"2023","journal-title":"Complex Intell Syst"},{"key":"pone.0332164.ref014","doi-asserted-by":"crossref","first-page":"7887","DOI":"10.1109\/TMM.2024.3372835","article-title":"Boundary-Guided Lightweight Semantic Segmentation With Multi-Scale Semantic Context","volume":"26","author":"Q Zhou","year":"2024","journal-title":"IEEE Trans Multimedia"},{"key":"pone.0332164.ref015","doi-asserted-by":"crossref","first-page":"12138","DOI":"10.1109\/JSTARS.2024.3416623","article-title":"A Method of Water Body Extraction Based on Multiscale Feature and Global Context Information","volume":"17","author":"R Miao","year":"2024","journal-title":"IEEE J Sel Top Appl Earth Observations Remote Sensing"},{"issue":"6","key":"pone.0332164.ref016","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1145\/3639053","article-title":"MCFNet: Multi-Attentional Class Feature Augmentation Network for Real-Time Scene Parsing","volume":"20","author":"X Wang","year":"2024","journal-title":"ACM Trans Multimedia Comput Commun Appl"},{"key":"pone.0332164.ref017","doi-asserted-by":"crossref","first-page":"3732","DOI":"10.1109\/CVPR52733.2024.00358","article-title":"Contextrast: Contextual Contrastive Learning for Semantic Segmentation","author":"C Sung","year":"2024","journal-title":"2024 IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR)"},{"key":"pone.0332164.ref018","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1109\/TGRS.2024.3362914","article-title":"ELGC-Net: Efficient Local\u2013Global Context Aggregation for Remote Sensing Change Detection","volume":"62","author":"M Noman","year":"2024","journal-title":"IEEE Trans Geosci Remote Sensing"},{"issue":"1","key":"pone.0332164.ref019","doi-asserted-by":"crossref","first-page":"14737","DOI":"10.1038\/s41598-025-99663-9","article-title":"A multi-scale remote sensing semantic segmentation model with boundary enhancement based on UNetFormer","volume":"15","author":"J Wang","year":"2025","journal-title":"Sci Rep"},{"key":"pone.0332164.ref020","doi-asserted-by":"crossref","unstructured":"Xu Z, Wang Z, Wang H, Liu C, Zhou Y, Sun M. Boosting Lightweight Camouflaged Object Detection with Multi-Scale Context and Boundary Awareness. In: ICASSP 2025 - 2025 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). 2025;1\u20135. https:\/\/doi.org\/10.1109\/icassp49660.2025.10888969","DOI":"10.1109\/ICASSP49660.2025.10888969"},{"key":"pone.0332164.ref021","doi-asserted-by":"crossref","first-page":"105482","DOI":"10.1016\/j.autcon.2024.105482","article-title":"CNN-based network with multi-scale context feature and attention mechanism for automatic pavement crack segmentation","volume":"164","author":"J Liang","year":"2024","journal-title":"Autom Constr"},{"key":"pone.0332164.ref022","doi-asserted-by":"crossref","first-page":"293","DOI":"10.1016\/j.neucom.2019.12.007","article-title":"Traffic scene semantic segmentation using self-attention mechanism and bi-directional GRU to correlate context","volume":"386","author":"M Yan","year":"2020","journal-title":"Neurocomputing"},{"key":"pone.0332164.ref023","doi-asserted-by":"crossref","first-page":"9174","DOI":"10.1109\/JSTARS.2022.3214889","article-title":"High-resolution remote sensing image semantic segmentation via multiscale context and linear self-attention","volume":"15","author":"P Yin","year":"2022","journal-title":"IEEE J Sel Top Appl Earth Observations Remote Sensing"},{"key":"pone.0332164.ref024","doi-asserted-by":"crossref","unstructured":"Strudel R, Garcia R, Laptev I, Schmid C. Segmenter: Transformer for Semantic Segmentation. In: 2021 IEEE\/CVF International Conference on Computer Vision (ICCV). 2021;7242\u201352. https:\/\/doi.org\/10.1109\/iccv48922.2021.00717","DOI":"10.1109\/ICCV48922.2021.00717"},{"key":"pone.0332164.ref025","first-page":"1251","author":"F Chollet","year":"2017"},{"key":"pone.0332164.ref026","first-page":"12021","author":"J Chen","year":"2023"},{"key":"pone.0332164.ref027","doi-asserted-by":"crossref","unstructured":"Wang Q, Wu B, Zhu P, Li P, Zuo W, Hu Q. ECA-Net: Efficient Channel Attention for Deep Convolutional Neural Networks. In: 2020 IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR). 2020;11531\u20139. https:\/\/doi.org\/10.1109\/cvpr42600.2020.01155","DOI":"10.1109\/CVPR42600.2020.01155"},{"key":"pone.0332164.ref028","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1109\/LGRS.2024.3370299","article-title":"Hybrid convolutional and attention network for hyperspectral image denoising","volume":"21","author":"S Hu","year":"2024","journal-title":"IEEE Geosci Remote Sensing Lett"},{"issue":"2","key":"pone.0332164.ref029","doi-asserted-by":"crossref","first-page":"1489","DOI":"10.1109\/TPAMI.2022.3164083","article-title":"Contextual Transformer Networks for Visual Recognition","volume":"45","author":"Y Li","year":"2023","journal-title":"IEEE Trans Pattern Anal Mach Intell"},{"key":"pone.0332164.ref030","unstructured":"OpenMMLab. MMSegmentation: OpenMMLab semantic segmentation toolbox and benchmark [Internet]. 2020 [cited 2024 Apr 3]. Available from: https:\/\/github.com\/open-mmlab\/mmsegmentation"},{"key":"pone.0332164.ref031","doi-asserted-by":"crossref","unstructured":"Rezatofighi H, Tsoi N, Gwak J, Sadeghian A, Reid I, Savarese S. Generalized Intersection Over Union: A Metric and a Loss for Bounding Box Regression. In: 2019 IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2019. 658\u201366. https:\/\/doi.org\/10.1109\/cvpr.2019.00075","DOI":"10.1109\/CVPR.2019.00075"},{"key":"pone.0332164.ref032","doi-asserted-by":"crossref","unstructured":"Martin D, Fowlkes C, Tal D, Malik J. A database of human segmented natural images and its application to evaluating segmentation algorithms and measuring ecological statistics. In: Proceedings Eighth IEEE International Conference on Computer Vision. ICCV 2001. 416\u201323. https:\/\/doi.org\/10.1109\/iccv.2001.937655","DOI":"10.1109\/ICCV.2001.937655"},{"key":"pone.0332164.ref033","doi-asserted-by":"crossref","unstructured":"Cordts M, Omran M, Ramos S, Rehfeld T, Enzweiler M, Benenson R, et al. The Cityscapes Dataset for Semantic Urban Scene Understanding. In: 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). 2016;3213\u201323. https:\/\/doi.org\/10.1109\/cvpr.2016.350","DOI":"10.1109\/CVPR.2016.350"},{"key":"pone.0332164.ref034","doi-asserted-by":"crossref","unstructured":"Liu Z, Lin Y, Cao Y, Hu H, Wei Y, Zhang Z, et al. Swin Transformer: Hierarchical Vision Transformer using Shifted Windows. In: 2021 IEEE\/CVF International Conference on Computer Vision (ICCV). 2021;9992\u201310002. https:\/\/doi.org\/10.1109\/iccv48922.2021.00986","DOI":"10.1109\/ICCV48922.2021.00986"}],"container-title":["PLOS One"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0332164","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,5,14]],"date-time":"2026-05-14T17:40:47Z","timestamp":1778780447000},"score":1,"resource":{"primary":{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0332164"}},"subtitle":[],"editor":[{"given":"Baohua","family":"Wen","sequence":"first","affiliation":[],"role":[{"role":"editor","vocabulary":"crossref"}]}],"short-title":[],"issued":{"date-parts":[[2026,5,14]]},"references-count":34,"journal-issue":{"issue":"5","published-online":{"date-parts":[[2026,5,14]]}},"URL":"https:\/\/doi.org\/10.1371\/journal.pone.0332164","relation":{},"ISSN":["1932-6203"],"issn-type":[{"value":"1932-6203","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,5,14]]}}}