Laineteisenduspõhine piltidelt müraeemaldus
Failid
Kuupäev
2019
Autorid
Ajakirja pealkiri
Ajakirja ISSN
Köite pealkiri
Kirjastaja
Abstrakt
Käesolev töö uurib laineteisenduste kasutust piltide kvaliteedi parandamise eesmärgil, neilt müra eemaldades. See annab ülevaate erinevates müra tüüpidest ning müraeemaldusmeetoditest. Edasi keskendub töö laineteisenduspõhistele müraeemaldusskeemidele. Samuti uurib töö laineteisenduspõhise müraeemaldusmeetodi ning kokkupakkimise kombineerimise kasulikkust ja pakub välja uue lävendamise tüübi ning muudatuse eksisteerivale BayesShrink meetodile. Pakutud meetod implementeeritakse C# keeles ning selle implementatsiooni tulemusi, jõudlust ning optimaalseid parameetreid analüüsitakse eksperimentaalsete tulemuste abil.
This thesis studies the use of wavelets for the purpose of improving the quality of images by removing noise from them. It presents an overview of different types of noise and denoising methods. The thesis then focuses on the wavelet transform based denoising schemes. It explores the potential of combining wavelet-based denoising and compression, and presents a new thresholding type and a modification to the existing BayesShrink method. The proposed method is implemented in the C# language and its performance and optimal parameters are analyzed through experimental results.
This thesis studies the use of wavelets for the purpose of improving the quality of images by removing noise from them. It presents an overview of different types of noise and denoising methods. The thesis then focuses on the wavelet transform based denoising schemes. It explores the potential of combining wavelet-based denoising and compression, and presents a new thresholding type and a modification to the existing BayesShrink method. The proposed method is implemented in the C# language and its performance and optimal parameters are analyzed through experimental results.