Institutional Repository of Vilnius University
Evoliuciniai optimizavimo algoritmai: taikymas nuotraukų koliažo sudarymo problemos sprendime /
Abstract
dc:descriptionIn today's technologically driven world, computers are essential for performing numerous tasks with exact accuracy. Nonetheless, there are certain complex problems where deterministic algorithms fall short. Creating a collage, which involves strategically placing several images on a fixed canvas without leaving any gaps or overlaps, is one such problem and is known to be NP-Hard. This thesis investigates the use of Particle Swarm Optimization (PSO) and Harmony Search Optimisation (HSO) as evolutionary optimization algorithms to address the Image Collage Composition problem. It delves deeply into the functionalities of both PSO and HSO, examining their effectiveness through the development of a unit test generator that divides an image into $n$ segments and uses the PSO algorithm to optimally or near-optimally arrange these segments on the canvas. The performance of the PSO algorithm is thoroughly tested and critically assessed using various measures. The study wraps up with a discussion of the findings and suggestions for improving the PSO algorithm in future research endeavors. This thesis enhances the understanding of evolutionary optimization algorithms and demonstrates their practical utility in solving intricate optimization issues like collage creation.
Degree
thesis:*- Grantor dc:publisher
- Institutional Repository of Vilnius University
- Year dc:date
- 2024
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Jasaitis, Kazimieras,
Rights
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:ELABAETD210648365&prefLang=en_US