Aplikacija za optičko prepoznavanje i normalizaciju teksta

master's thesis

• Title (english): Application for Optical Text Recognition and Normalization
• Author: Ivan Varga
• Mentor: Marin Vuković (mentor)
• Committee member: Marin Vuković (predsjednik povjerenstva)
• Committee member: Miljenko Mikuc (član povjerenstva)
• Committee member: Krešimir Pripužić (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2020-07-08, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

U ovom radu definiran je i implementiran sustav optičkog prepoznavanja i normalizacije teksta. Za svrhu optičkog prepoznavanja implementirana je aplikacija za operacijski sustav Android dok je za normalizaciju implementiran poslužitelj u tehnologiji Spring. Obje aplikacije pisane su u jeziku Kotlin. Nakon definicije sustava nalazi se opis svih tehnologija koje su korištene u sklopu rada. Također je definirano što je normalizacija te koje probleme susrećemo tijekom implementacije normalizacije teksta na hrvatskom jeziku. Za svaki problem ponuđeno je rješenje te je u kratko opisan način na koji je normalizacija implementirana. Normalizacija se provodi za kratice, akronime, datume, redne i glavne brojeve te simbole i vrijeme. Na kraju prikazana je konkretna implementacija nekih metoda i regularnih izraza na poslužitelju te je dan uvid u funkcionalnost klijenta te izgled korisničkog sučelja. Također dan je opis modela normalizacije ovisne o kontekstu te potencijalnog rješenja takvog problema.

Abstract (english)

This paper defines and implements a system for optical text recognition and normalization of text in Croatian language. For optical text recognition an Android OS application was developed. The server is implemented using Spring framework. Both applications are written in Kotlin programming language. After the system definition comes a detailed description of all technologies used in the implementation of both the client and the server applications. The definition of normalization is also given with known issues when trying to normalize text in Croatian language. For each issue the author proposes a solution with a short description of the exact implementation details. Normalization is being conducted for acronyms, abbreviations, decimal, ordinal and whole numbers as well as for symbols and time. At last a concrete implementation of some methods and regular expressions used on the server is described with an overview of the client application and its functionality and user interface. Finally, a model for normalization based on context is defined and a possible solution is proposed to solve such a problem.

• Keywords: normalizacija; optičko prepoznavanje teksta; Android; Spring Boot; Kotlin
• Keywords (english): normalization; optical character recognition; Android; Spring Boot; Kotlin
• Language: croatian
• URN:NBN: urn:nbn:hr:168:803422
• Study programme: Title: Computing; Study programme type: university; Study level: graduate; Academic / professional title: magistar/magistra inženjer/inženjerka računarstva (magistar/magistra inženjer/inženjerka računarstva)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2023-01-13 14:07:55