prikaz prve stranice dokumenta Sustav za raspoznavanje teksta u fotografijama
No public access
undergraduate thesis
Sustav za raspoznavanje teksta u fotografijama
Zagreb: University of Zagreb, Faculty of Electrical Engineering and Computing, 2020. urn:nbn:hr:168:733659
Bićanić, Miroslav
University of Zagreb
Faculty of Electrical Engineering and Computing

Institutional repository: FER Repository

Cite this document

Bićanić, M. (2020). Sustav za raspoznavanje teksta u fotografijama (Undergraduate thesis). Zagreb: University of Zagreb, Faculty of Electrical Engineering and Computing. Retrieved from https://urn.nsk.hr/urn:nbn:hr:168:733659

Bićanić, Miroslav. "Sustav za raspoznavanje teksta u fotografijama." Undergraduate thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, 2020. https://urn.nsk.hr/urn:nbn:hr:168:733659

Bićanić, Miroslav. "Sustav za raspoznavanje teksta u fotografijama." Undergraduate thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, 2020. https://urn.nsk.hr/urn:nbn:hr:168:733659

Bićanić, M. (2020). 'Sustav za raspoznavanje teksta u fotografijama', Undergraduate thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, accessed 22 November 2024, https://urn.nsk.hr/urn:nbn:hr:168:733659

Bićanić M. Sustav za raspoznavanje teksta u fotografijama [Undergraduate thesis]. Zagreb: University of Zagreb, Faculty of Electrical Engineering and Computing; 2020 [cited 2024 November 22] Available at: https://urn.nsk.hr/urn:nbn:hr:168:733659

M. Bićanić, "Sustav za raspoznavanje teksta u fotografijama", Undergraduate thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, Zagreb, 2020. Available at: https://urn.nsk.hr/urn:nbn:hr:168:733659

Metadata
TitleSustav za raspoznavanje teksta u fotografijama
Title (english)System for Text Recognition in Photographs
AuthorMiroslav Bićanić
MentorZoran Kalafatić (mentor)
Committee memberZoran Kalafatić (predsjednik povjerenstva)
Committee memberIvona Brajdić (član povjerenstva)
Committee memberKarlo Slovenec (član povjerenstva)
GranterUniversity of Zagreb
Faculty of Electrical Engineering and Computing
Zagreb
Defense date and country2020-07-06, Croatia
Scientific / art field,
discipline and subdiscipline		TECHNICAL SCIENCES
Computing
Abstract
U ovome radu analizira se postupak optičkog prepoznavanja teksta na slikama. Navodi se općeniti opis postupka te njegova podjela na osnovne korake: predprocesiranje, segmentaciju, klasifikaciju i postprocesiranje. Spominju se i OCR sustavi koji već postoje na tržištu, kao što je Tesseract, te povijesni postupci kojima se problem pojednostavljuje, kao što su posebno prilagođeni fontovi poput OCR-A. Posebna pažnja dana je koracima segmentacije i klasifikacije, koji čine osnovu optičkog prepoznavanja teksta. Opisani su različiti načini na koje je moguće izvršiti te korake, njihove prednosti i nedostatci, te česti problemi s kojima se razvojni programeri susreću u implementaciji. Kroz cijeli rad postepeno se uvodi i analizira implementirani sustav za optičko prepoznavanje teksta, a posebno njegovi podsustavi za segmentaciju i klasifikaciju. Podsustav za segmentaciju implementiran je algoritamski, dok se podsustav za klasifikaciju oslanja na strojno učenje koristeći neuronsku mrežu. Na kraju su prikazani rezultati razvijenog sustava na nekoliko primjera slika s tekstom.
Abstract (english)
This paper analyses the process of optically recognizing text from images. The general description of the process is given, as well as the division into its basic steps: pre-processing, segmentation, classification and post-processing. Existing OCR systems, such as Tesseract, are mentioned along with historical methods of simplifying the problem, such as the utilization of specially adapted fonts like OCR-A. Special attention is given to the steps of segmentation and classification, which represent the core of optical text recognition. Various different approaches to these steps are described, together with their benefits and drawbacks, and frequent problems with which developers are faced when implementing such systems. Throughout the paper, the implemented system for optical text recognition is gradually introduced and described, especially its subsystems for segmentation and classification. The segmentation subsystem utilizes an algorithmic approach, while the classification subsystem relies on machine learning in the form of using a neural network. In the end, the results of the developed system are shown on several examples of images that contain text.
Keywords
Keywords (english)
Languagecroatian
URN:NBNurn:nbn:hr:168:733659
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 on2023-01-14 08:28:49