Kartezijsko genetsko programiranje u konvoluciji slike

master's thesis

• Title (english): Cartesian Genetic Programming for image convolution
• Author: Lukas Šestić
• Mentor: Domagoj Jakobović (mentor)
• Committee member: Domagoj Jakobović (predsjednik povjerenstva)
• Committee member: Marko Đurasević (član povjerenstva)
• Committee member: Marko Čupić (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2021-07-16, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

Ovaj rad bavi se obradom slike konvolucijskim metodama koristeći Kartezijsko genetsko programiranje (CGP).
Istražuju se mogućnosti i performanse razvoja filtera za obradu slike.
Prikazuju se rezultati pri micanju šuma i detekciji rubova, pokazujući prednosti koje dolaze s korištenjem CGP-a.
Nažalost, pokušaji generiranja podataka izvedbom autoenkodera CGP-om ne rezultiraju uspjehom.
Do izražaja dolaze određeni nedostatci genetskog programiranja koji ukazuju na korištenje prikladnijih tehnika za određene probleme.

Abstract (english)

This paper describes the process of image convolution using Cartesian genetic programming (CGP).
Possibilities and performance in evolving image filters are explored.
Results are shown with image denoising filters and edge detection, pointing to the benefits of using CGP.
Unfortunately, synthetic data generation using CGP as an autoencoder does not result in such success.
Various downsides of genetic programming make other, more appropriate deep learning techniques a better choice for some problems.

• Keywords: genetsko programiranje; Kartezijevsko genetsko programiranje; konvolucija; strojno učenje; obrada slike
• Keywords (english): genetic programming; Cartesian genetic programming; convolution; machine learning; image processing
• Language: croatian
• URN:NBN: urn:nbn:hr:168:627275
• Study programme: Title: Computing; Study programme type: university; Study level: graduate; Academic / professional title: magistar/magistra inženjer/inženjerka računarstva (magistar/magistra inženjer/inženjerka računarstva)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2023-01-13 16:15:47