{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,2,17]],"date-time":"2026-02-17T07:14:12Z","timestamp":1771312452737,"version":"3.50.1"},"reference-count":45,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,2,17]],"date-time":"2026-02-17T00:00:00Z","timestamp":1771286400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"Anhui Province Mid-Career and Young Teachers Training Initiative-Outstanding Young Teacher Cultivation Project","award":["YQYB2023163"],"award-info":[{"award-number":["YQYB2023163"]}]},{"name":"Anhui Vocational and Adult Education Association Planning Project","award":["AZCJ2024208"],"award-info":[{"award-number":["AZCJ2024208"]}]},{"name":"School Level Quality Engineering Project \u201cTeaching Resource Library for Software Technology Major\u201d","award":["2023jxzyk01"],"award-info":[{"award-number":["2023jxzyk01"]}]},{"name":"Teacher\u2019s Internship Program for Hanging Jobs in Industry and Enterprises","award":["xjgz2024009"],"award-info":[{"award-number":["xjgz2024009"]}]},{"name":"Chuzhou Polytechnic Natural Science Research Project"},{"name":"Anhui Provincial Natural Science Research Project for Higher Education Institutions","award":["ZKZ-2022-02"],"award-info":[{"award-number":["ZKZ-2022-02"]}]},{"name":"Anhui Provincial Natural Science Research Project for Higher Education Institutions","award":["2024AH051439"],"award-info":[{"award-number":["2024AH051439"]}]},{"name":"Anhui Provincial Natural Science Research Project for Higher Education Institutions","award":["2023AH053088"],"award-info":[{"award-number":["2023AH053088"]}]},{"name":"Chuzhou Polytechnic Science and Technology Innovation Platform Project","award":["YJP-2023-02"],"award-info":[{"award-number":["YJP-2023-02"]}]},{"name":"Anhui Provincial Quality Engineering Project for Higher Education Institutions","award":["2022jnds043"],"award-info":[{"award-number":["2022jnds043"]}]}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Neurorobot."],"abstract":"<jats:sec>\n                    <jats:title>Problem<\/jats:title>\n                    <jats:p>Deep learning technology promotes the development of single-image dehazing. However, many existing methods fail to fully consider the haze density and its spatial distribution, which limits the improvement of dehazing performance.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Proposed solution<\/jats:title>\n                    <jats:p>To address this issue, we propose an attention-based multi-scale feature aggregation network (AMSA-Net) for single-image dehazing.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Method<\/jats:title>\n                    <jats:p>AMSA-Net is an encoding and decoding structure. Its encoder and decoder are composed of multi-scale hybrid attention feature aggregation module (MSHA-FAM). The module can perceive the haze density and spatial information in the haze image, which helps to improve the dehazing effect. MSHA-FAM is composed of two key components: the scale-aware coordinate residual module (SCRM) and multi-scale feature refinement residual module (MSFRRM). SCRM uses improved coordinate attention to effectively capture haze density and spatial characteristics, thus significantly improving dehazing effect. MSFRRM extracts semantic features through up-sampling and down-sampling, and uses improved pixel attention mechanism to enhance key features. In the overall MSHA-FAM pipeline, SCRM first learns the density and spatial distribution characteristics of haze, then refines it through MSFRRM, so as to remove haze more effectively.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Key results<\/jats:title>\n                    <jats:p>The experimental results demonstrate that our proposed AMSA-Net is superior to the comparison methods in terms of dehazing quality. Ablation studies further verify the effectiveness of the proposed modules.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Impact<\/jats:title>\n                    <jats:p>In this work, we present AMSA-Net, which has achieved good dehazing performance and can provide high-quality input for subsequent computer vision tasks.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.3389\/fnbot.2026.1698100","type":"journal-article","created":{"date-parts":[[2026,2,17]],"date-time":"2026-02-17T06:33:40Z","timestamp":1771310020000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["AMSA-Net: attention-based multi-scale feature aggregation network for single image dehazing"],"prefix":"10.3389","volume":"20","author":[{"given":"Shanqin","family":"Wang","sequence":"first","affiliation":[{"name":"School of Information Engineering, Chuzhou Polytechnic","place":["Chuzhou, China"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Mengjun","family":"Miao","sequence":"additional","affiliation":[{"name":"School of Information Engineering, Chuzhou Polytechnic","place":["Chuzhou, China"]},{"name":"School of Computer, Qinghai Normal University","place":["Xining, China"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Miao","family":"Zhang","sequence":"additional","affiliation":[{"name":"School of Information Engineering, Chuzhou Polytechnic","place":["Chuzhou, China"]}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"1965","published-online":{"date-parts":[[2026,2,17]]},"reference":[{"key":"ref1","doi-asserted-by":"publisher","first-page":"999","DOI":"10.1109\/TIP.2017.2771158","article-title":"Single image dehazing using color ellipsoid prior","volume":"27","author":"Bui","year":"2017","journal-title":"IEEE Trans. Image Process."},{"key":"ref2","first-page":"52","article-title":"All-in-one image dehazing based on attention mechanism","author":"Dai","year":"2023"},{"key":"ref3","first-page":"2157","article-title":"Multi-scale boosted dehazing network with dense feature fusion","author":"Dong","year":"2020"},{"key":"ref4","doi-asserted-by":"publisher","first-page":"5024","DOI":"10.3390\/s23115024","article-title":"Deep learning-based anomaly detection in video surveillance: a survey","volume":"23","author":"Duong","year":"2023","journal-title":"Sensors"},{"key":"ref5","doi-asserted-by":"publisher","first-page":"104747","DOI":"10.1016\/j.imavis.2023.104747","article-title":"Single image dehazing using extended local dark channel prior","volume":"136","author":"Dwivedi","year":"2023","journal-title":"Image Vis. Comput."},{"key":"ref6","doi-asserted-by":"publisher","first-page":"331","DOI":"10.1007\/s41095-022-0271-y","article-title":"Attention mechanisms in computer vision: a survey","volume":"8","author":"Guo","year":"2022","journal-title":"Comput. Vis. Media"},{"key":"ref7","first-page":"5812","article-title":"Image dehazing transformer with transmission-aware 3D position embedding","author":"Guo","year":"2022"},{"key":"ref8","doi-asserted-by":"publisher","first-page":"2341","DOI":"10.1109\/TPAMI.2010.168","article-title":"Single image haze removal using dark channel prior","volume":"33","author":"He","year":"2010","journal-title":"IEEE Trans. Pattern Anal. Mach. Intell."},{"key":"ref9","first-page":"1670","article-title":"TransER: hybrid model and ensemble-based sequential learning for non-homogenous dehazing","author":"Hoang","year":"2023"},{"key":"ref10","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1109\/TGRS.2024.3360516","article-title":"Spatiotemporal enhancement and interlevel fusion network for remote sensing images change detection","volume":"62","author":"Huang","year":"2024","journal-title":"IEEE Trans. Geosci. Remote Sens."},{"key":"ref11","doi-asserted-by":"publisher","first-page":"320","DOI":"10.3724\/SP.J.1089.2023.19275","article-title":"Hazy image dehazing algorithm based on two branch residual feature fusion","volume":"35","author":"Ji","year":"2023","journal-title":"J. Comput. Aided Des. Comput. Graph."},{"key":"ref12","doi-asserted-by":"publisher","first-page":"1870","DOI":"10.1109\/LSP.2024.3430066","article-title":"FGN: a fully guided network for image dehazing","volume":"31","author":"Ju","year":"2024","journal-title":"IEEE Signal Process Lett."},{"key":"ref13","first-page":"1","article-title":"Single image dehazing via multi-scale large kernel convolutional neural networks","author":"Li","year":"2024"},{"key":"ref14","first-page":"4770","article-title":"AOD-Net: all-in-one dehazing network","author":"Li","year":"2017"},{"key":"ref15","doi-asserted-by":"publisher","first-page":"492","DOI":"10.1109\/TIP.2018.2867951","article-title":"Benchmarking single-image dehazing and beyond","volume":"28","author":"Li","year":"2018","journal-title":"IEEE Trans. Image Process."},{"key":"ref16","doi-asserted-by":"publisher","first-page":"3238","DOI":"10.1109\/TIP.2023.3279980","article-title":"Single image dehazing using saturation line prior","volume":"32","author":"Ling","year":"2023","journal-title":"IEEE Trans. Image Process."},{"key":"ref17","doi-asserted-by":"publisher","first-page":"102416","DOI":"10.1016\/j.displa.2023.102416","article-title":"MFID-Net: multi-scaled feature-fused image dehazing via dynamic weights","volume":"78","author":"Liu","year":"2023","journal-title":"Displays"},{"key":"ref18","first-page":"7314","article-title":"GridDehazeNet: attention-based multi-scale network for image dehazing","author":"Liu","year":"2019"},{"key":"ref19","doi-asserted-by":"publisher","first-page":"1897","DOI":"10.1049\/ipr2.12455","article-title":"Single image dehazing using generative adversarial networks based on an attention mechanism","volume":"16","author":"Ma","year":"2022","journal-title":"IET Image Process."},{"key":"ref20","doi-asserted-by":"publisher","first-page":"5858","DOI":"10.3390\/app15115858","article-title":"LKD-Net: lightweight single image dehazing via multi-head large kernel attention","volume":"15","author":"Moon","year":"2025","journal-title":"Appl. Sci."},{"key":"ref21","doi-asserted-by":"publisher","first-page":"713","DOI":"10.1109\/TPAMI.2003.1201821","article-title":"Contrast restoration of weather degraded images","volume":"25","author":"Narasimhan","year":"2003","journal-title":"IEEE Trans. Pattern Anal. Mach. Intell."},{"key":"ref22","first-page":"11908","article-title":"FFA-Net: feature fusion attention network for single image dehazing","author":"Qin","year":"2020"},{"key":"ref23","article-title":"Rethinking the elementary function fusion for single-image dehazing","author":"Rohn","year":"2024"},{"key":"ref24","doi-asserted-by":"publisher","first-page":"8423","DOI":"10.1007\/s11042-020-10035-z","article-title":"PSNR vs SSIM: imperceptibility quality assessment for image steganography","volume":"80","author":"Setiadi","year":"2021","journal-title":"Multimed. Tools Appl."},{"key":"ref25","doi-asserted-by":"publisher","first-page":"103327","DOI":"10.1016\/j.dsp.2021.103327","article-title":"Multi-scale residual attention network for single image dehazing","volume":"121","author":"Sheng","year":"2022","journal-title":"Digit. Signal Process."},{"key":"ref26","doi-asserted-by":"publisher","first-page":"1927","DOI":"10.1109\/TIP.2023.3256763","article-title":"Vision transformers for single image dehazing","volume":"32","author":"Song","year":"2023","journal-title":"IEEE Trans. Image Process."},{"key":"ref27","doi-asserted-by":"publisher","first-page":"10","DOI":"10.1016\/j.neunet.2023.03.017","article-title":"Multi-level feature interaction and efficient non-local information enhanced channel attention for image dehazing","volume":"163","author":"Sun","year":"2023","journal-title":"Neural Netw."},{"key":"ref28","doi-asserted-by":"publisher","first-page":"108458","DOI":"10.1016\/j.engappai.2024.108458","article-title":"The evolution of object detection methods","volume":"133","author":"Sun","year":"2024","journal-title":"Eng. Appl. Artif. Intell."},{"key":"ref29","doi-asserted-by":"publisher","first-page":"682","DOI":"10.1016\/j.procs.2022.08.082","article-title":"A novel encoder-decoder network with guided transmission map for single image dehazing","volume":"204","author":"Tran","year":"2022","journal-title":"Procedia Comput. Sci."},{"key":"ref30","doi-asserted-by":"publisher","first-page":"359","DOI":"10.1007\/s00371-024-03330-5","article-title":"Encoder-decoder networks with guided transmission map for effective image dehazing","volume":"41","author":"Tran","year":"2025","journal-title":"Vis. Comput."},{"key":"ref31","doi-asserted-by":"publisher","first-page":"5876","DOI":"10.1007\/s00034-025-03058-0","article-title":"Single image dehazing based on haze prior residual perception learning","volume":"44","author":"Wang","year":"2025","journal-title":"Circuits Syst. Signal Process."},{"key":"ref32","first-page":"1398","article-title":"Multiscale structural similarity for image quality assessment","author":"Wang","year":"2003"},{"key":"ref33","first-page":"25479","article-title":"ODCR: orthogonal decoupling contrastive regularization for unpaired image dehazing","author":"Wang","year":"2024"},{"key":"ref34","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1109\/TGRS.2023.3325927","article-title":"Encoder-free multiaxis physics-aware fusion network for remote sensing image dehazing","volume":"61","author":"Wen","year":"2023","journal-title":"IEEE Trans. Geosci. Remote Sens."},{"key":"ref35","first-page":"10551","article-title":"Contrastive learning for compact single image dehazing","author":"Wu","year":"2021"},{"key":"ref36","doi-asserted-by":"publisher","first-page":"109486","DOI":"10.1016\/j.engappai.2024.109486","article-title":"Multi-stream feature aggregation network with multi-scale supervision for single image dehazing","volume":"139","author":"Wu","year":"2025","journal-title":"Eng. Appl. Artif. Intell."},{"key":"ref37","doi-asserted-by":"publisher","first-page":"109225","DOI":"10.1016\/j.sigpro.2023.109225","article-title":"Single UHD image dehazing via interpretable pyramid network","volume":"214","author":"Xiao","year":"2024","journal-title":"Signal Process."},{"key":"ref38","doi-asserted-by":"publisher","first-page":"3351","DOI":"10.3390\/electronics11203351","article-title":"A novel approach to maritime image dehazing based on a large kernel encoder-decoder network with multihead pyramids","volume":"11","author":"Yang","year":"2022","journal-title":"Electronics"},{"key":"ref39","first-page":"2628","article-title":"Y-Net: multi-scale feature aggregation network with wavelet structure similarity loss function for single image dehazing","author":"Yang","year":"2020"},{"key":"ref40","doi-asserted-by":"publisher","first-page":"107692","DOI":"10.1016\/j.engappai.2023.107692","article-title":"Visual attention and ODE-inspired fusion network for image dehazing","volume":"130","author":"Yin","year":"2024","journal-title":"Eng. Appl. Artif. Intell."},{"key":"ref41","doi-asserted-by":"publisher","first-page":"9308","DOI":"10.1109\/TITS.2024.3354102","article-title":"Bayesian calibration of the intelligent driver model","volume":"25","author":"Zhang","year":"2024","journal-title":"IEEE Trans. Intell. Transp. Syst."},{"key":"ref42","doi-asserted-by":"publisher","first-page":"290","DOI":"10.3390\/jimaging11090290","article-title":"Contrastive learning-driven image dehazing with multi-scale feature fusion and hybrid attention mechanism","volume":"11","author":"Zhang","year":"2025","journal-title":"J. Imaging"},{"key":"ref43","first-page":"727","article-title":"Single image dehazing using bounded channel difference prior","author":"Zhao","year":"2021"},{"key":"ref44","first-page":"5785","article-title":"Curricular contrastive regularization for physics-aware single image dehazing","author":"Zheng","year":"2023"},{"key":"ref45","doi-asserted-by":"publisher","first-page":"6236","DOI":"10.1109\/TCSVT.2023.3264717","article-title":"Spectral dual-channel encoding for image dehazing","volume":"33","author":"Zhu","year":"2023","journal-title":"IEEE Trans. Circuits Syst. Video Technol."}],"container-title":["Frontiers in Neurorobotics"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fnbot.2026.1698100\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,2,17]],"date-time":"2026-02-17T06:33:40Z","timestamp":1771310020000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fnbot.2026.1698100\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,2,17]]},"references-count":45,"alternative-id":["10.3389\/fnbot.2026.1698100"],"URL":"https:\/\/doi.org\/10.3389\/fnbot.2026.1698100","relation":{},"ISSN":["1662-5218"],"issn-type":[{"value":"1662-5218","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,2,17]]},"article-number":"1698100"}}