Institutional Repository of Vilnius University
Klasifikavimo metodais grindžiami skaitmeninių vaizdų glaudinimo sprendimai /
Abstract
dc:descriptionIn this thesis the universal technology for compression of groups of digital images is proposed, which allow saving space occupied by the images while maintaining the pre-defined quality. The main idea of the technology is: with the application of classification, possible impact of the compression algorithm on the quality of the image is predicted, where the quality is assessed by objective quality assessment methods. With the application of the proposed technology two classification-based approaches for the compression of images in the JPEG format by classifying stored images into two or three classes prior to the application of the compression algorithm are proposed and analysed in the thesis. The approach for computing and selection of feature describing images, used for a maximally precise classification of images in accordance with their quality after they have been processed by compression algorithms, is defined in the thesis. The main results of the dissertation were published in 10 research papers: four papers are published in periodicals, reviewed scientific journals, one paper are published in conference proceedings, and five papers are published in conference abstracts. The main results have been presented and discussed at three national and six international conferences. Dissertation amount: 122 pages, 38 formulas, 35 figures, 28 tables and list of literature that contains 92 references.
Degree
thesis:*- Grantor dc:publisher
- Institutional Repository of Vilnius University
- Year dc:date
- 2018
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Tichonov, Jevgenij,
- Contributors dc:contributor
-
- Kurasova, Olga
Subjects
dc:subject × 1Rights
dc:rights- Statement dc:rights
-
- info:eu-repo/semantics/openAccess
- Language dc:language
- lit
Identifiers
dc:identifier.*- Repository record dc:identifier
- https://repository.vu.lt/VU:ELABAETD29430906&prefLang=en_US