master's thesis
Impulse Noise Removal in Image by Applying Selective Median Filter with Noise Detection

Adriana Matijas (2016)
Josip Juraj Strossmayer University of Osijek
Faculty of Electrical Engineering, Computer Science and Information Technology Osijek
Department of Communications
Laboratory for High Frequency Measurements
Metadata
TitleUklanjanje impulsnog šuma iz slike primjenom selektivnog medijan filtra s detekcijom šuma
AuthorAdriana Matijas
Mentor(s)Snježana Rimac-Drlje (thesis advisor)
Abstract
Standardni medijan filter poznat je kao učinkovito riješenje problema uklanjanja šumova iz slika, međutim kada se radi o slikama koje su teže oštećene šumma, standardni medijan filter gubi na učinkovitosti. U ovom je radu napravljen program za uklanjanje impulsnog sol i papar šuma primjenom medijan filtra, adaptivog medijan filtra i adaptivog medijan filtra s BDND algoritmom. Analizirana je kvaliteta slika nakon uklanjanja impulsnog šuma navedenim algoritmima te je utvrđeno da adaptivni medijan filtar učinkovito otklanja šum u slikama s manje detalja i s izraženim rubovima. Za slike s mnoštvom detalja najbolje rezultate pokazuje adaptivni medijan filtar s BDND algoritmom, ali ovaj način otklanjanja šuma zahtjeva značajno više procesorskog vremena i nije pogodan kod visokih postotaka šumom oštećenih elemenata slike.
Keywords- impulse noise adaptive median filter standard median filter noise detection
Parallel title (English)Impulse Noise Removal in Image by Applying Selective Median Filter with Noise Detection
Committee MembersSnježana Rimac-Drlje (committee chairperson)
Mario Vranješ (committee member)
Irena Galić (committee member)
GranterJosip Juraj Strossmayer University of Osijek
Faculty of Electrical Engineering, Computer Science and Information Technology Osijek
Lower level organizational unitsDepartment of Communications
Laboratory for High Frequency Measurements
PlaceOsijek
StateCroatia
Scientific field, discipline, subdisciplineTECHNICAL SCIENCES
Computing
Process Computing
Study programme typeuniversity
Study levelgraduate
Study programmeGraduate University Study Programme in Computer Engineering
Academic title abbreviationmag.ing.comp.
Genremaster's thesis
Language Croatian
Defense date2016-10-12
Parallel abstract (English)
Standard median filter is known as an effective solution in noise removing from corrupted images, however when it comes to badly corrupted images standard median filter loses its effectiveness. In this paper a program has been made to remove salt and pepper noise by applying a median filter, adaptive median filter and adaptive median filter with BDND algorithm. Quality of the image has been analyzed after removing the impulse noise using mentioned algorithms and found that adaptive median filter effectively eliminates noise in images with less details and with pronounced edges. For pictures with lot of details best results shows adaptive median filter with BDND algorithm, but this method of eliminating noise requires significantly more processing time and is not suitable for the large percentage of damaged picture elements
Parallel keywords (Croatian)Impulsni šum adaptivni medijan filter standardni medijan filter detekcija šuma
Resource typetext
Access conditionOpen access
Terms of usehttp://rightsstatements.org/vocab/InC/1.0/
URN:NBNhttps://urn.nsk.hr/urn:nbn:hr:200:543002
CommitterAnka Ovničević