Robust underwater image enhancement with cascaded multi-level sub-networks and triple attention mechanism

Basit Öğe Kaydı
dc.authoridLin, Zifan/0000-0002-7046-3102
dc.contributor.authorZhang, Dehuan
dc.contributor.authorWu, Chenyu
dc.contributor.authorZhou, Jingchun
dc.contributor.authorZhang, Weishi
dc.contributor.authorLin, Zifan
dc.contributor.authorPolat, Kemal
dc.contributor.authorAlenezi, Fayadh
dc.date.accessioned2024-09-25T19:57:36Z
dc.date.available2024-09-25T19:57:36Z
dc.date.issued2024
dc.departmentAbant İzzet Baysal Üniversitesien_US
dc.description.abstractWith the growing exploration of marine resources, underwater image enhancement has gained significant attention. Recent advances in convolutional neural networks (CNN) have greatly impacted underwater image enhancement techniques. However, conventional CNN-based methods typically employ a single network structure, which may compromise robustness in challenging conditions. Additionally, commonly used UNet networks generally force fusion from low to high resolution for each layer, leading to inaccurate contextual information encoding. To address these issues, we propose a novel network called Cascaded Network with Multi-level Sub-networks (CNMS), which encompasses the following key components: (a) a cascade mechanism based on local modules and global networks for extracting feature representations with richer semantics and enhanced spatial precision, (b) information exchange between different resolution streams, and (c) a triple attention module for extracting attention-based features. CNMS selectively cascades multiple subnetworks through triple attention modules to extract distinct features from underwater images, bolstering the network's robustness and improving generalization capabilities. Within the sub-network, we introduce a Multi-level Sub-network (MSN) that spans multiple resolution streams, combining contextual information from various scales while preserving the original underwater images' high-resolution spatial details. Comprehensive experiments on multiple underwater datasets demonstrate that CNMS outperforms state-of-the-art methods in image enhancement tasks.en_US
dc.description.sponsorshipNational Natural Science Foundation of China [62301105]; Liaoning Provincial Natural Science Foundation of China [20170520196]; Fundamental Research Funds for the Central Universities, China [3132019205, 3132019354]; Cultivation Program for the Excellent Doctoral Dissertation of Dalian Maritime Universityen_US
dc.description.sponsorshipThis work was supported in part by the National Natural Science Foundation of China (No. 61702074) , the National Natural Science Foundation of China (No. 62301105) , the Liaoning Provincial Natural Science Foundation of China (No. 20170520196) , and the Fundamental Research Funds for the Central Universities, China (Nos. 3132019205 and 3132019354) , and the Cultivation Program for the Excellent Doctoral Dissertation of Dalian Maritime University.en_US
dc.identifier.doi10.1016/j.neunet.2023.11.008
dc.identifier.endpage697en_US
dc.identifier.issn0893-6080
dc.identifier.issn1879-2782
dc.identifier.pmid37972512en_US
dc.identifier.scopus2-s2.0-85177166864en_US
dc.identifier.scopusqualityQ1en_US
dc.identifier.startpage685en_US
dc.identifier.urihttps://doi.org/10.1016/j.neunet.2023.11.008
dc.identifier.urihttps://hdl.handle.net/20.500.12491/13494
dc.identifier.volume169en_US
dc.identifier.wosWOS:001166234000001en_US
dc.identifier.wosqualityQ1en_US
dc.indekslendigikaynakWeb of Scienceen_US
dc.indekslendigikaynakScopusen_US
dc.indekslendigikaynakPubMeden_US
dc.language.isoenen_US
dc.publisherPergamon-Elsevier Science Ltden_US
dc.relation.ispartofNeural Networksen_US
dc.relation.publicationcategoryMakale - Uluslararası Hakemli Dergi - Kurum Öğretim Elemanıen_US
dc.rightsinfo:eu-repo/semantics/closedAccessen_US
dc.snmzYK_20240925en_US
dc.subjectDeep convolutional networken_US
dc.subjectUnderwater image enhancementen_US
dc.subjectCascading mechanismen_US
dc.subjectMulti-scale feature representationen_US
dc.titleRobust underwater image enhancement with cascaded multi-level sub-networks and triple attention mechanismen_US
dc.typeArticleen_US

Dosyalar

Koleksiyon

WoS İndeksli Yayınlar Koleksiyonu
PubMed İndeksli Yayınlar Koleksiyonu
Scopus İndeksli Yayınlar Koleksiyonu