| | SLO | ENG | Cookies and privacy

Bigger font | Smaller font

Show document

Title:Aplikacija odprtokodne knjižnice za optično prepoznavo znakov na operacijskem sistemu Android
Authors:Nemanič, Tina (Author)
Leskovar, Robert (Mentor) More about this mentor... New window
Files:.pdf MAG_Nemanic_Tina_2017.pdf (1,72 MB)
 
Language:Slovenian
Work type:Master's thesis/paper (mb22)
Organization:FOV - Faculty of Organizational Sciences in Kranj
Abstract:Magistrska naloga obravnava odprtokodno knjižnico Tesseract za optično prepoznavo znakov (OCR, optical character recognition) na operacijskem sistemu Android. Podane so osnove in pregled aktualnih raziskav na področju optične prepoznave znakov. Raziskave so usmerjene v: a) prepoznavanje ročne pisave, b) samostojne aplikacije na mobilnih napravah, c) delovanje v realnem času v kombinaciji s prevodom, d) večjezične aplikacije in e) strojno učenje. Razlogi za odločitev za uporabo knjižnice Tesseract so: odprta koda, najboljši rezultati pri neodvisnih merjenjih in prilagojenost za operacijski sistem Android. Kot razvojno orodje je bil uporabljen Android Studio zaradi enostavnosti uporabe, ugodne krivulje učenja, možnosti uporabe čarovnika in močnih orodjih za razhroščevanje. Razvita rešitev je aplikacija za operacijski sistem Android 5.0 Lollipop, ki uporablja API 21. V letu 2016 bi razvito aplikacijo lahko uporabilo približno 40,5% naprav. Aplikacija deluje popolnoma avtonomno, čas prepoznave je pod eno minuto. Rezultati testiranja so pokazali, da v slabih svetlobnih pogojih, kot je na primer slika plinskega ali električnega števca, dobimo zelo nizko stopnjo prepoznanih številk. To stopnjo je možno povečati z dodatnim treningom aplikacije, vendar so tudi tu pomembne omejitve.
Keywords:odprta koda, OCR-optična prepoznava znakov, Android, aplikacija, knjižnica Tesseract
Year of publishing:2017
Source:Maribor
COBISS_ID:7939603 Link is opened in a new window
NUK URN:URN:SI:UM:DK:BTMJIZ21
Views:271
Downloads:52
Metadata:XML RDF-CHPDL DC-XML DC-RDF
Categories:FOV
:
  
Average score:(0 votes)
Your score:Voting is allowed only for logged in users.
Share:AddThis
AddThis uses cookies that require your consent. Edit consent...

Hover the mouse pointer over a document title to show the abstract or click on the title to get all document metadata.

Secondary language

Language:English
Title:Application of open-source library for optical character recognition on operating system Android
Abstract:Master's thesis addresses open-source library Tesseract for optical character recognition (OCR) on Android operating system. Foundations and survey of contemporary research on the field of optical character recognition are presented. Research is focused in: a) recognition of handwriting, b) autonomous applications on mobile devices c) real-time recognition in combination with translation in other languages, d) multilingual applications and e) machine learning. Rationales to choose Tesseract OCR library are open source code, the best results in independent measurement and suitability for Android operating system. Android Studio was used as a development tool due to simplicity of use, steep learning curve and powerful debugging tools. Application is developed for Android 5.0 Lollipop operating system, which uses API level 21. Approximately 40.5 % of devices in year 2016 could use developed application. Application works autonomously, time of optical recognition is under one minute. Results of a tests shows that the rate of character recognition on gas or electricity meter in poor light conditions is very low. It is possible to increase this rate with additional training of application, however with significant limitations.
Keywords:open source code, OCR – optical character recognition, Android, application, Tesseract library


Comments

Leave comment

You have to log in to leave a comment.

Comments (0)
0 - 0 / 0
 
There are no comments!

Back
Logos of partners University of Maribor University of Ljubljana University of Primorska University of Nova Gorica