Deep learning-based blind image super-resolution with iterative kernel reconstruction and noise estimation

Basit Öğe Kaydı
dc.authorid0000-0003-0779-9620
dc.contributor.authorAteş, Hasan Fehmi
dc.contributor.authorYıldırım, Süleyman
dc.contributor.authorGüntürk, Bahadır Kürşat
dc.date.accessioned2023-07-10T10:09:17Z
dc.date.available2023-07-10T10:09:17Z
dc.date.issued2023
dc.departmentİstanbul Medipol Üniversitesi, Mühendislik ve Doğa Bilimleri Fakültesi, Elektrik ve Elektronik Mühendisliği Bölümü
dc.description.abstractBlind single image super-resolution (SISR) is a challenging task in image processing due to the ill-posed nature of the inverse problem. Complex degradations present in real life images make it difficult to solve this problem using naive deep learning approaches, where models are often trained on synthetically generated image pairs. Most of the effort so far has been focused on solving the inverse problem under some constraints, such as for a limited space of blur kernels and/or assuming noise-free input images. Yet, there is a gap in the literature to provide a well-generalized deep learning-based solution that performs well on images with unknown and highly complex degradations. In this paper, we propose IKR-Net (Iterative Kernel Reconstruction Network) for blind SISR. In the proposed approach, kernel and noise estimation and high-resolution image reconstruction are carried out iteratively using dedicated deep models. The iterative refinement provides significant improvement in both the reconstructed image and the estimated blur kernel even for noisy inputs. IKR-Net provides a generalized solution that can handle any type of blur and level of noise in the input low-resolution image. IKR-Net achieves state-of-the-art results in blind SISR, especially for noisy images with motion blur.
dc.identifier.citationAteş, H. F., Yıldırım, S. ve Güntürk, B. K. (2023). Deep learning-based blind image super-resolution with iterative kernel reconstruction and noise estimation. Computer Vision and Image Understanding, 233. https://dx.doi.org/10.1016/j.cviu.2023.103718
dc.identifier.doi10.1016/j.cviu.2023.103718
dc.identifier.issn1077-3142
dc.identifier.issn1090-235X
dc.identifier.scopus2-s2.0-85162834377
dc.identifier.scopusqualityQ1
dc.identifier.urihttps://dx.doi.org/10.1016/j.cviu.2023.103718
dc.identifier.urihttps://hdl.handle.net/20.500.12511/11175
dc.identifier.volume233
dc.identifier.wos001010560700001en_US
dc.identifier.wosqualityQ2
dc.indekslendigikaynakWeb of Science
dc.indekslendigikaynakScopus
dc.institutionauthorGüntürk, Bahadır Kürşat
dc.language.isoen
dc.publisherAcademic Press Inc.
dc.relation.ispartofComputer Vision and Image Understandingen_US
dc.relation.publicationcategoryMakale - Uluslararası Hakemli Dergi - Kurum Öğretim Elemanı
dc.relation.tubitakinfo:eu-repo/grantAgreement/TUBITAK/SOBAG/119E566
dc.rightsinfo:eu-repo/semantics/embargoedAccess
dc.subjectSuper-Resolution
dc.subjectBlind
dc.subjectIterative
dc.subjectDeep Network
dc.titleDeep learning-based blind image super-resolution with iterative kernel reconstruction and noise estimation
dc.typeArticle

Dosyalar

Orijinal paket
Listeleniyor 1 - 1 / 1
Küçük Resim Yok
İsim:
Gunturk-Bahdir-2023.pdf
Boyut:
3.88 MB
Biçim:
Adobe Portable Document Format
Açıklama:
Tam Metin / Full Text
İndir
Lisans paketi
Listeleniyor 1 - 1 / 1
Küçük Resim Yok
İsim:
license.txt
Boyut:
1.44 KB
Biçim:
Item-specific license agreed upon to submission
Açıklama:
İndir

Koleksiyon

Makale Koleksiyonu
Scopus İndeksli Yayınlar Koleksiyonu
WoS İndeksli Yayınlar Koleksiyonu