Digitalizacija arhivskoga gradiva s uključenim procesom optičkog prepoznavanja znakova na primjeru Muzičkih novina Hrvatskog državnog konzervatorija

master's thesis

• Title (english): Digitization of archival material with the included process of optical character recognition on the example of Muzičke novine
• Author: Ilija Kitin
• Mentor: Željko Trbušić (mentor)
• Committee member: Hrvoje Stančić (predsjednik povjerenstva)
• Committee member: Ana Barbarić (član povjerenstva)
• Committee member: Željko Trbušić (član povjerenstva)
• Granter: University of Zagreb; Faculty of Humanities and Social Sciences; (Department of information and Communication sciences); Zagreb
• Defense date and country: 2023-12-04, Croatia
• Scientific / art field, discipline and subdiscipline: SOCIAL SCIENCES; Information and Communication Sciences; Library Science

Abstract

Diplomski rad donosi temu procesa digitalizacije u knjižnici Muzičke akademije u Zagrebu s uključenim procesom optičkog prepoznavanja znakova na primjeru Muzičkih novina Hrvatskog državnog konzervatorija. Istražen je utjecaj rezolucije na brzinu digitalizacije, a time i troškove. Rezultate se izrazilo u radnim satima i kao točnost OCR-a, a korištene su dvije verzije temeljnog teksta. Uzorak se sastojao od dva broja novina, tj. deset stranica. Odabir postavki digitalizacije može dugoročno... More smanjiti troškove, a uz testiranje točnosti OCR-a i zadržati prihvatljive razine vjerodostojnosti. Važna je ravnoteža između troška digitalizacije i postignute razine točnosti. Najviša je točnost postignuta sa ScanSnap skenerom na 300 dpi i drugim temeljnim tekstom. Najniža je postignuta upotrebom kopirnog uređaja pri 600 dpi i drugog temeljnog teksta. Sveukupno je viša točnost omogućena sa ScanSnapom. Prvi temeljni tekst u 66,67% slučajeva vodio je k višoj točnosti OCR-a. Najbolja opcija u svim aspektima osim točnosti je kopirni uređaji na 200 dpi. Less

Abstract (english)

The thesis presents the digitization process in the library of the Academy of Music in Zagreb with OCR on “Muzičke novine”. The impact of the resolution on the speed and the costs was investigated. The results were expressed in working hours and OCR accuracy. Two versions of GT were used. The sample consisted of two issues, i.e. ten pages. Choosing digitization settings can reduce costs in the long run, and with OCR accuracy testing, maintain acceptable levels of accuracy. The balance... More between the cost of and the level of accuracy is important. The highest accuracy was achieved with the ScanSnap scanner at 300 dpi and the second version of GT. The lowest was achieved using a copier at 600 dpi and the second version of GT. Overall higher accuracy was observed with the ScanSnap scanner. The first version of GT led to higher accuracy in 66.67% of cases. The best option in all aspects except accuracy is the copier at 200 dpi. Less

• Keywords: digitalizacija; OCR; temeljni tekst; Muzičke novine; ocreval; ISRI alati
• Keywords (english): digitization; OCR; Ground Truth; Muzičke novine; ocreval; ISRI Tools
• Language: croatian
• URN:NBN: urn:nbn:hr:131:671735
• Study programme: Title: Information Sciences; specializations in: Archivistics Study, Library Science, Information Sciences (research), Information Sciences (teaching), Museology and heritage management, Informatology; Course: Library Science; Study programme type: university; Study level: graduate; Academic / professional title: magistar/magistra bibliotekarstva (magistar/magistra bibliotekarstva)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Open access
• Created on: 2024-01-25 09:36:46