prikaz prve stranice dokumenta Raspoznavanje tipa terena na temelju satelitskih snimki
  Download
PDF 1.06 MB
undergraduate thesis
Raspoznavanje tipa terena na temelju satelitskih snimki
2019. urn:nbn:hr:200:944175
Mučaj, Klaudija
Josip Juraj Strossmayer University of Osijek
Faculty of Electrical Engineering, Computer Science and Information Technology Osijek
Department of Software Engineering
Chair of Programming Languages and Systems

Cite this document

Mučaj, K. (2019). Raspoznavanje tipa terena na temelju satelitskih snimki (Undergraduate thesis). Osijek: Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek. Retrieved from https://urn.nsk.hr/urn:nbn:hr:200:944175

Mučaj, Klaudija. "Raspoznavanje tipa terena na temelju satelitskih snimki." Undergraduate thesis, Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek, 2019. https://urn.nsk.hr/urn:nbn:hr:200:944175

Mučaj, Klaudija. "Raspoznavanje tipa terena na temelju satelitskih snimki." Undergraduate thesis, Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek, 2019. https://urn.nsk.hr/urn:nbn:hr:200:944175

Mučaj, K. (2019). 'Raspoznavanje tipa terena na temelju satelitskih snimki', Undergraduate thesis, Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek, accessed 19 April 2024, https://urn.nsk.hr/urn:nbn:hr:200:944175

Mučaj K. Raspoznavanje tipa terena na temelju satelitskih snimki [Undergraduate thesis]. Osijek: Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek; 2019 [cited 2024 April 19] Available at: https://urn.nsk.hr/urn:nbn:hr:200:944175

K. Mučaj, "Raspoznavanje tipa terena na temelju satelitskih snimki", Undergraduate thesis, Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek, Osijek, 2019. Available at: https://urn.nsk.hr/urn:nbn:hr:200:944175

Please login to the repository to save this object to your list.
Metadata
TitleRaspoznavanje tipa terena na temelju satelitskih snimki
Title (english)Land cover classification based on satellite images
AuthorKlaudija Mučaj
MentorZdravko Krpić (mentor)
MentorBruno Zorić (mentor)
Committee memberZdravko Krpić (predsjednik povjerenstva)
Committee memberTomislav Matić (član povjerenstva)
Committee memberGoran Martinović (član povjerenstva)
GranterJosip Juraj Strossmayer University of Osijek
Faculty of Electrical Engineering, Computer Science and Information Technology Osijek
(Department of Software Engineering)
(Chair of Programming Languages and Systems)
Osijek
Defense date and country2019-10-01, Croatia
Scientific / art field,
discipline and subdiscipline		TECHNICAL SCIENCES
Computing
Program Engineering
Abstract
U teorijskom dijelu ovog rada objašnjen je problem raspoznavanja tipa terena, te su opisani postupci i koraci strojnog učenja za raspoznavanje tipa terena. U praktičnom je dijelu prikazan program kojim su se za problem raspoznavanja tipa terena uspoređivali četiri klasifikatora. Ti klasifikatori su: SVM, DT, RF, te kNN. Korištena je baza satelitskih snimaka SAT-4. U njoj su označena četiri tipa terena: neplodna zemlja, drveće, travnjak i klasa koju čine svi ostali tipovi terena. Klasifikatori na temelju značajki slika predviđaju o kojem se tipu terena radi. Podaci od 100,000 slika se prvo dijele na trening i testni skup. Klasifikatori na trening skupu uče prepoznavati tipove terena, a na testnom skupu predviđaju o kojem se tipu terena radi. Na temelju rezultata predviđanja testnog skupa, izračunavaju se iduće mjere: F-mjera, preciznost, odziv i točnost. Klasifikatori se vrednuju s obzirom na te mjere. Usporedbom rezultata tih mjera zaključuje se kako se klasifikator RF pokazao kao najuspješniji u raspoznavanju tipa terena.
Abstract (english)
The problem of terrain type recognition is explained in the theoretical part of this thesis. With it, the procedures and steps of creating a machine learning models for terrain type recognition are explained. The practical part of this thesis shows a program which is used for comparing four classifiers on the terrain type recognition problem. The four classifiers are: SVM, DT, RF, and kNN.. The SAT-4 satellite imagery database is used in this thesis. The database contains pictures which are labeled as barren land, trees, grassland or other (which represents all other land cover classes). Based on image features, the classifiers predict the land type. A dataset of 100,000 pictures is first divided into a training and test set. On the training set the classifiers learn to recognize the types of terrain, while they are tested on the test set. Based on the results of the test set predictions, the following measures are calculated: F-score, precision, recall and accuracy. With these measures the classifiers are evaluated. By comparing the results of these measures, the RF classifier was shown to be the most successful in recognizing terrain type.
Keywords
Keywords (english)
Languagecroatian
URN:NBNurn:nbn:hr:200:944175
Study programmeTitle: Undergraduate University Study Programme in Computer Engineering
Study programme type: university
Study level: undergraduate
Academic / professional title: sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka računarstva (sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka računarstva)
Type of resourceText
File originBorn digital
Access conditionsOpen access
Terms of useLicence In copyright icon
Created on2019-12-20 14:54:09