Image super-resolution using deep feedforward neural networks in spectral domain

Simple item page
buir.advisorAksoy, Selim
dc.contributor.authorAydın, Onur
dc.date.accessioned2018-03-29T09:10:36Z
dc.date.available2018-03-29T09:10:36Z
dc.date.copyright2018-03
dc.date.issued2018-03
dc.date.submitted2018-03-28
dc.departmentDepartment of Computer Engineeringen_US
dc.descriptionCataloged from PDF version of article.en_US
dc.descriptionThesis (M.S.): Bilkent University, Department of Computer Engineering, İhsan Doğramacı Bilkent University, 2018.en_US
dc.descriptionIncludes bibliographical references (leaves 55-59)..en_US
dc.description.abstractWith recent advances in deep learning area, learning machinery and mainstream approaches in computer vision research have changed dramatically from hardcoded features combined with classi ers to end-to-end trained deep convolutional neural networks (CNN) which give the state-of-the-art results in most of the computer vision research areas. Single-image super-resolution is one of these areas which are considerably in uenced by deep learning advancements. Most of the current state-of-the-art methods on super-resolution problem learn a nonlinear mapping from low-resolution images to high-resolution images in the spatial domain using consecutive convolutional layers in their network architectures. However, these state-of-the-art results are obtained by training a separate neural network architecture for each di erent scale factor. We propose a novel singleimage super-resolution system with the limited number of learning parameters in spectral domain in order to eliminate the necessity to train a separate neural network for each scale factor. As a spectral transform function which converts images from the spatial domain to the frequency domain, discrete cosine transform (DCT) which is a variant of discrete Fourier transform (DFT) is used. In addition, in the post-processing step, an artifact reduction module is added for removing ringing artifacts occurred due to spectral transformations. Even if the peak signal-to-noise ratio (PSNR) measurement of our super-resolution system is lower than current state-of-the-art methods, the spectral domain allows us to develop a single model with a single dataset for any scale factor and relatively obtain better structural similarity index (SSIM) results.en_US
dc.description.degreeM.S.en_US
dc.description.provenanceSubmitted by Betül Özen (ozen@bilkent.edu.tr) on 2018-03-29T09:10:36Z No. of bitstreams: 1 thesis_onur_aydin.pdf: 28538459 bytes, checksum: a1f49543c50c3f76fbe5f5b9cfba9542 (MD5)en
dc.description.provenanceMade available in DSpace on 2018-03-29T09:10:36Z (GMT). No. of bitstreams: 1 thesis_onur_aydin.pdf: 28538459 bytes, checksum: a1f49543c50c3f76fbe5f5b9cfba9542 (MD5) Previous issue date: 2018-03en
dc.description.statementofresponsibilityby Onur Aydın.en_US
dc.format.extentxiii, 66 leaves : illustrations (some color) ; 30 cmen_US
dc.identifier.itemidB150135
dc.identifier.urihttp://hdl.handle.net/11693/36341
dc.language.isoEnglishen_US
dc.publisherBilkent Universityen_US
dc.rightsinfo:eu-repo/semantics/openAccessen_US
dc.subjectSuper-resolutionen_US
dc.subjectDeep learningen_US
dc.subjectFourier transformen_US
dc.titleImage super-resolution using deep feedforward neural networks in spectral domainen_US
dc.title.alternativeSpekral alanda derin ileri beslemeli sinir ağları kullanılarak görüntü süper çözünürlüğüen_US
dc.typeThesisen_US

Files

Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
thesis_onur_aydin.pdf
Size:
27.22 MB
Format:
Adobe Portable Document Format
Description:
Full printable version
Download
License bundle
Now showing 1 - 1 of 1
No Thumbnail Available
Name:
license.txt
Size:
1.71 KB
Format:
Item-specific license agreed upon to submission
Description:
Download

Collections

Dept. of Computer Engineering - Master's degree