Umjetno degradiranje kvalitete slike

master's thesis

• Title (english): Artificial degradation of image quality
• Author: Marta Novak
• Mentor: Sonja Grgić (mentor)
• Committee member: Sonja Grgić (predsjednik povjerenstva)
• Committee member: Tomislav Kos (član povjerenstva)
• Committee member: Dario Bojanjac (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2020-07-14, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

Ovaj radi bavi se problematikom umjetne degradacije slika s naglaskom na primjenu u autonomnim vozilima. Degradacije kvalitete odnosi se na zamućenje, šum i izobličenja u slici te opstrukciju vidljivosti dijelova slike. Dan je pregled javno dostupnih baza za autonomna vozila i opis postojećih postupaka za degeneriranje slika. Napravljena je analiza karakteristika degradiranih slika i usporedba sa slikama bez degradacije. Na temelju provedene analize u programskom paketu MATLAB definiran je model za umjetno degradiranje slika. Dobivene slike korištene su za treniranje klasifikatora uz verifikaciju generaliziranja modela na stvarnim primjerima degradiranih slika.

Abstract (english)

This thesis addresses the problems of artificial image degradation with a focus on application in autonomous vehicles. Degradation of image quality refers to image haze, noise, distortion, and obstruction of visibility of particular image segments. An overview of publicly available databases for autonomous vehicles has been addressed, as well as descriptions of existing image degradation models. An analysis of the characteristics of degraded images, versus undegraded images has been made as well. Based on the carried out analysis in the program package MATLAB, a model for artificial image degradation has been defined. The resulting images were used for training classifier along with model verification on real examples of degraded images.

• Keywords: Degradacija slika; Analiza slika; Klasifikacija; AdaBoost
• Keywords (english): Image Degradation; Image Analysis; Classification; AdaBoost
• Language: croatian
• URN:NBN: urn:nbn:hr:168:098916
• Study programme: Title: Information and Communication Technology; Study programme type: university; Study level: graduate; Academic / professional title: magistar/magistra inženjer/inženjerka informacijske i komunikacijske tehnologije (magistar/magistra inženjer/inženjerka informacijske i komunikacijske tehnologije)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2023-01-13 09:17:49