Očitavanje 7-segmentnih identifikacijskih oznaka studenata

master's thesis

• Title (english): Automatic Recognition of 7-Segment Student Identifiers
• Author: Shadi Zahran
• Mentor: Marko Čupić (mentor)
• Committee member: Marko Čupić (predsjednik povjerenstva)
• Committee member: Jan Šnajder (član povjerenstva)
• Committee member: Leonardo Jelenković (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2020-07-15, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

Automatsko očitavanje rukom pisanih brojeva i tekstova na temelju slike je računalno težak problem. Kako bi se u obrazovnim institucijama riješio problem unosa identifikacijskih brojeva, studentima je predočen predložak koji se sastoji od 7-segmentnih znakova kojeg onda oni moraju popuniti sa svojim brojem. U okviru ovog diplomskog rada, istraženi su mogući načini za ostvarivanje jednog sustava za identifikaciju studentskih oznaka sa popunjenih predložaka. U sklopu tog istraživanja, proučene su razne metode segmentacije slike koja je ključni dio obrade slike. Osim toga su proučavane i umjetne neuronske mreže te načini na koji mogu pomoći ovome problemu. Uz segmentaciju slike dobiveni su ulazi za napravljenu neuronsku mrežu koja je onda trenirana te je dobiven model čija je točnost na skupu za testiranje koji sačinjavaju slike od 300 dpi-ja 100%.

Abstract (english)

Automatic handwriting and text recognition are a computationally very difficult problem. Since students in universities have to write their personal identifying numbers during exams or other activities that problem is made simpler by offering them a template consisting of 7-segment characters which they are able to fill in order to form their identifying number, removing the handwriting recognition problem in the process. In this thesis, ways to implement such a system are researched as well as ways to extract data from an image like image segmentation. After performing image segmentation on the filled templates an artificial neural network is developed and trained on the training set. The resulting model is tested on the test set which consisted of 300 dpi images and the model performed exceptionally well with the accuracy of it being at the maximum possible, 100%.

• Keywords: Analiza slike; segmentacija slike; automatsko prepoznavanje znakova; strojno učenje; umjetne neuronske mreže; 7-segmentni prikaz
• Keywords (english): Image analysis; image segmentation; automatic character recognition; machine learning; artificial neural networks; 7-segment display
• Language: croatian
• URN:NBN: urn:nbn:hr:168:794956
• 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 10:54:44