|
|
Photo Noise Reduction Using Deep Neural Networks
Tichý, Jonáš ; Juránek, Roman (referee) ; Španěl, Michal (advisor)
Obrazový šum je fundamentálním problémem v digitální fotografii. Cílem této práce je studium redukce šumu ve fotografiích pomocí hlubokých neuronových sítí. Dvě vybrané metody založené na hlubokých neuronových sítích, DnCNN a BRDNet, byly implementovány a jejich výkon byl změřen v několika experimentech. Kromě toho byl navržen a proveden experiment na uživatelích s cílem vyhodnotit vnímanou kvalitu obrazu širokou veřejností. Experimenty ukázaly, že zatímco obě metody dosahují výborných výsledků v metrikách, jako je PSNR a SSIM, vnímaná vizuální kvalita ne vždy koreluje s numerickými metrikami. Výsledky prezentované v této práci zdůrazňují důležitost vhodných trénovacích dat a metrik kvality obrazu v odšumování digitálních fotografií.
|
|
|
HDR Image Acquisition on Mobile Devices
Mik, Filip ; Musil, Martin (referee) ; Nosko, Svetozár (advisor)
This bachelor thesis deals with aquisition of high dynamic range images. The resulting HDR image is created by processing of LDR image sequence - by merging and then tone mapping. This thesis includes description of all issues related to the process of acquisition and displaying HDR image. Whole process is then implemented in form of Android application.
|
|
|
Digital Photo Processing
Zdražil, Vít ; Sumec, Stanislav (referee) ; Potúček, Igor (advisor)
This document focuses on processing of RAW image data from digital cameras. In first section is described principle of image capturing by digital camera, common way of image processing in the device and what the RAW format is, its pros and cons. In the next section is described existing demosaicing methods, methods for additional processing of RAW image data and there is analyzed specific RAW format, Canon's CR2, including structure and guide for its conversion. Next sections contains proposal of the new improved demosaicing method and method for suppressing digital noise. On this basis a library for basic CR2 files processing was implemented. There is comparison of methods for processing RAW image data with this library in the next section. In the conclusion there is summary of finished work and there is also mentioned outline of future work.
|
|
|
Mobile App for Acquisition and Processing of HDR Images
Michalák, Patrik ; Brejcha, Jan (referee) ; Čadík, Martin (advisor)
The goal of this paper is to create an application that solves HDR image processing problems and focuses to interactivity with the user, offering him more opportunities in clear and minimalistic graphic user interface. Application generate HDR content by merging a series of low dynamic range images with different exposure time values. The solution provides work with HDR photography, four tone mapping methods and various tools for user.
|
|
|
PANASONIC FZ7 Digital Cameras Simulator and Trainer
Bielik, Branislav ; Drozdová, Martina (referee) ; Kunovský, Jiří (advisor)
The Bachelor's thesis Simulator and Trainer of the Panasonic FZ7 Camera deals with a design and a realization of the training manual and the trainer for this digital camera. The application will be used to support the education at the University of the Third Age. Adobe Flash and programming language ActionScript is used for implementation. Basics of photography and application creation in Flash are described. Simulation allows trying possibilities of a device in teaching and manual explains basic usage of the camera.
|
|
|
Výkonnostní a regresní testování nástroje Zoner Photo Studio X
Konopík, Matěj ; Kočí, Radek (referee) ; Smrčka, Aleš (advisor)
This thesis deals with the problematics of testing the manipulation of RAW format files from the modern digital cameras in Zoner Photo Studio X. Automatic regression tests are used together with the grey box method to continually control the quality of the output image date. A new testing system was created that is used to test the RAW conversion module. This system brings and open interface to easily and quickly integrate new image data tests, It also features automatic launching, analysis and reporting of the results. Thanks to this new system, the implementation of new test will become much easier and this will moreover increase the quality of the product.
|
|
|
Jitters
Schrijversová, Tereza ; Spáčilová Blažková, Kateřina (referee) ; Fišerová, Lucia (advisor)
In her bachelor's thesis, the author will focus on working with the photographs of her family archive, which she uses in her artistic practice. This archive depicts the author's past. In her work, the author will subjectively reflect on the possibilities of perceiving pictorial memories as two separate, ephemeral realities for which we cannot confirm or deny their reality. At the same time, she wants to avoid the possible patheticness that working with archives tends to lead to, which is why she will use more media than just a single photograph. The artistic space he will create should serve for subjective contemplation of the pictorial medium whose affinity between reality and a certain interiority denies the exclusivity of only one reality. It accentuates here a kind of leaving a trace whose stratification refers to the psychological development of the personality in its inherent features. The aim of the work will be a multimedia installation using not only the aesthetics of original cinema films.This opens up the possibility of working with the film material itself, as a carrier of visual information, as well as its digital copy. The subsequent combination of these media will bring the viewer closer to the intimate interior of the artist herself. The work will be realized in a gallery environment, but it can also be set in a less formal setting.
|
|
|
Photo Noise Reduction Using Deep Neural Networks
Tichý, Jonáš ; Juránek, Roman (referee) ; Španěl, Michal (advisor)
Obrazový šum je fundamentálním problémem v digitální fotografii. Cílem této práce je studium redukce šumu ve fotografiích pomocí hlubokých neuronových sítí. Dvě vybrané metody založené na hlubokých neuronových sítích, DnCNN a BRDNet, byly implementovány a jejich výkon byl změřen v několika experimentech. Kromě toho byl navržen a proveden experiment na uživatelích s cílem vyhodnotit vnímanou kvalitu obrazu širokou veřejností. Experimenty ukázaly, že zatímco obě metody dosahují výborných výsledků v metrikách, jako je PSNR a SSIM, vnímaná vizuální kvalita ne vždy koreluje s numerickými metrikami. Výsledky prezentované v této práci zdůrazňují důležitost vhodných trénovacích dat a metrik kvality obrazu v odšumování digitálních fotografií.
|
|
| |
|
|
Photography as a special visual resource and itś aplication to art education in kindergarten
Půtová, Martina ; Šmíd, Jan (advisor) ; Hazuková, Helena (referee)
The thesis deals with visual culture and digital photography in art education of pre-school children. The main objective is to uncover children's ability to work with a specific type of picture - photographic picture. The educational project is based on theoretical knowledge of digital media provided in the first part of the thesis and on practical experience gained during several years of practice with pre-school children. The project offers children to use digital photographs as another form of art expression. Talking with children and monitoring their reactions have shown that working with digital photographs is equally beneficial to them as any other art technique.
|
guest ::
