prikaz prve stranice dokumenta Raspoznavanje znakova na registarskim tablicama
No public access
undergraduate thesis
Raspoznavanje znakova na registarskim tablicama
Zagreb: University of Zagreb, Faculty of Electrical Engineering and Computing, 2019. urn:nbn:hr:168:746917
Pandžić, Franko
University of Zagreb
Faculty of Electrical Engineering and Computing

Institutional repository: FER Repository

Cite this document

Pandžić, F. (2019). Raspoznavanje znakova na registarskim tablicama (Undergraduate thesis). Zagreb: University of Zagreb, Faculty of Electrical Engineering and Computing. Retrieved from https://urn.nsk.hr/urn:nbn:hr:168:746917

Pandžić, Franko. "Raspoznavanje znakova na registarskim tablicama." Undergraduate thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, 2019. https://urn.nsk.hr/urn:nbn:hr:168:746917

Pandžić, Franko. "Raspoznavanje znakova na registarskim tablicama." Undergraduate thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, 2019. https://urn.nsk.hr/urn:nbn:hr:168:746917

Pandžić, F. (2019). 'Raspoznavanje znakova na registarskim tablicama', Undergraduate thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, accessed 10 August 2024, https://urn.nsk.hr/urn:nbn:hr:168:746917

Pandžić F. Raspoznavanje znakova na registarskim tablicama [Undergraduate thesis]. Zagreb: University of Zagreb, Faculty of Electrical Engineering and Computing; 2019 [cited 2024 August 10] Available at: https://urn.nsk.hr/urn:nbn:hr:168:746917

F. Pandžić, "Raspoznavanje znakova na registarskim tablicama", Undergraduate thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, Zagreb, 2019. Available at: https://urn.nsk.hr/urn:nbn:hr:168:746917

Metadata
TitleRaspoznavanje znakova na registarskim tablicama
Title (english)Recognition of License Plate Characters
AuthorFranko Pandžić
MentorZoran Kalafatić (mentor)
Committee memberZoran Kalafatić (predsjednik povjerenstva)
Committee memberJosip Šarić (član povjerenstva)
Committee memberMarin Oršić (član povjerenstva)
GranterUniversity of Zagreb
Faculty of Electrical Engineering and Computing
Zagreb
Defense date and country2019-07-12, Croatia
Scientific / art field,
discipline and subdiscipline		TECHNICAL SCIENCES
Computing
Abstract
U okviru završnog rada obrađeno je raspoznavanje znakova postojećim modelom i raspoznavanje znakova vlastitim naučenim modelom programskim jezikom Python. Za postojeći model korišten je Tesseract OCR uz biblioteku OpenCV. Za izgradnju vlastitog modela korištena je biblioteka scikit-learn, također uz pomoć OpenCV-a. Slike nastale u stvarnom svijetu sadrže mnogo smetnje što nam otežava automatizirano očitavanje. Iz tog se razloga nad slikama vrši više postupaka u nadi eliminacije iste. Za treniranje vlastitog klasifikatora potrebna je baza slika raspoređenih po sadržaju. Nakon uspješnog treniranja, možemo klasifikator i testirati. Slike za testiranje istog su tipa kao i slike namijenjene treniranju klasifikatora. Ulazne slike koje korisnik odabire unutar grafičkog sučelja potrebno je segmentirati kako bi se znak po znak mogao slati klasifikatoru na raspoznavanje. Naposlijetku, usporedba rezultata dvaju modela nad istim ulaznim podacima prikazuje se unutar grafičkog korisničkog sučelja.
Abstract (english)
In this work, we covered character recognition using an existing model and character recognition using our trained model written in Python programming language. We used Tesseract OCR as an existing model with OpenCV library. scikit-learn library along with OpenCV was used to create our own trained model. Images generated in the real world tend to have a lot of noise which makes recoginiton harder. For this reason, we apply various methods to images in hopes of eliminating image noise.. A database of categorised pictures is needed for training and testing of our classifier. After successful training, we can test our classifier. Images for testing are the same type as images used for training. Input images selected by the user should be segmented so that characters could be sent individually to the classifier for recognition. Finally, comparison of the results of two models over the same input data is displayed within the graphical user interface.
Keywords
Keywords (english)
Languagecroatian
URN:NBNurn:nbn:hr:168:746917
Study programmeTitle: Computing
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 conditionsClosed access
Terms of useLicence In copyright icon
Public note
Created on2020-01-07 20:55:25