Usporedba algoritama za OCR pri različitim rezolucijama

master's thesis

• Author: Mario Mikšić
• Mentor: Darko Agić (mentor)
• Committee member: Darko Agić (član povjerenstva)
• Granter: University of Zagreb; Faculty of Graphic Arts; Zagreb
• Defense date and country: 2012-12-05, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Graphic Technology

Abstract

Optičko prepoznavanje znakova (OCR) je kombinacija mehaničke i elektoničke komunikacije za prijenos rukom pisanih ili otisnutih tekstova u računalni kod pomoću optičkih uređaja. Njegova povijest seže i sto godina unazad i do danas se razgranala tj. dio je jedne široke skupine optičkog prepoznavanja.Primjena njegovih mogućnosti dovela nas je do toga da nam se optičko prepoznavanje uvuklo u sve životne prostore kao što su prepoznavanje poslovnih kartica,interaktivne igre novog doba,osobna identifikacija itd.,a najzaslužniji za to su sve bolji algoritmi. Oni omogućavaju sve bržu i precizniju obradu (unos,raščiščavanje i usporedbu)no uvijek postoji mogućnost razvoja pogotovo kombinacijom više algoritama. Bitna stavka je da pri prepoznavanju algoritmi zadrže mogućnost"gledanja"na sliku bez obzira na rezoluciju i transformaciju slike (translataciju, kaliranje i rotaciju) tj. da stvori slikovne invarijacije i da se prepoznavanje dovede do razine ljudskih osjetila.

Abstract (english)

Optical Character Recognition(OCR) is combination of mechanical and electronic comunication for transmission of handwritten, typewritten or printed text into machine-encoded text using optical devices. It's histry goes hundred years back and today has become part of the group of optical recognition devices. Using it's possibilities led us to point where we have optical recognition in every area of our lives such as the recognition of business cards, interactive games of new age, personal identification, etc., and most responsible for it's progress are algorithms. They allow faster and more accurate processing (input, clearing and comparison), but there is always possibility of more development, especially with the combination of two or more algorithms. The important fact is that the algorithms for recognition keep the possibility of "seeing" the image, regardless of image resolution and transformation (translation, scaling and rotation)and to create image invariants and to bring recognition to the level of human senses.

• Keywords: Optičko prepoznavanje; slikovne invarijacije; rezolucija
• Keywords (english): Optical recognition; Image invariants; moment; resolution
• Language: croatian
• URN:NBN: urn:nbn:hr:216:899267
• Study programme: Title: Graphic technology; specializations in: Graphic technology, Design of graphic products; Course: Graphic technology; Study programme type: university; Study level: graduate; Academic / professional title: sveučilišni/a magistar/magistra inženjer/inženjerka grafičke tehnologije (sveučilišni/a magistar/magistra inženjer/inženjerka grafičke tehnologije)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2024-04-17 08:56:22