prikaz prve stranice dokumenta Automatsko prepoznavanje rukom pisanog teksta
No public access
undergraduate thesis
Automatsko prepoznavanje rukom pisanog teksta
Zagreb: University of Zagreb, Faculty of Electrical Engineering and Computing, 2020. urn:nbn:hr:168:527075
Vnučec, Nikola
University of Zagreb
Faculty of Electrical Engineering and Computing

Institutional repository: FER Repository

Cite this document

Vnučec, N. (2020). Automatsko prepoznavanje rukom pisanog teksta (Undergraduate thesis). Zagreb: University of Zagreb, Faculty of Electrical Engineering and Computing. Retrieved from https://urn.nsk.hr/urn:nbn:hr:168:527075

Vnučec, Nikola. "Automatsko prepoznavanje rukom pisanog teksta." Undergraduate thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, 2020. https://urn.nsk.hr/urn:nbn:hr:168:527075

Vnučec, Nikola. "Automatsko prepoznavanje rukom pisanog teksta." Undergraduate thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, 2020. https://urn.nsk.hr/urn:nbn:hr:168:527075

Vnučec, N. (2020). 'Automatsko prepoznavanje rukom pisanog teksta', Undergraduate thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, accessed 28 December 2024, https://urn.nsk.hr/urn:nbn:hr:168:527075

Vnučec N. Automatsko prepoznavanje rukom pisanog teksta [Undergraduate thesis]. Zagreb: University of Zagreb, Faculty of Electrical Engineering and Computing; 2020 [cited 2024 December 28] Available at: https://urn.nsk.hr/urn:nbn:hr:168:527075

N. Vnučec, "Automatsko prepoznavanje rukom pisanog teksta", Undergraduate thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, Zagreb, 2020. Available at: https://urn.nsk.hr/urn:nbn:hr:168:527075

Metadata
TitleAutomatsko prepoznavanje rukom pisanog teksta
Title (english)Automatic recognition of handwritten text
AuthorNikola Vnučec
MentorMarko Čupić (mentor)
Committee memberMarko Čupić (predsjednik povjerenstva)
Committee memberRudolf Lovrenčić (član povjerenstva)
Committee memberKarlo Knežević (član povjerenstva)
GranterUniversity of Zagreb
Faculty of Electrical Engineering and Computing
Zagreb
Defense date and country2020-07-15, Croatia
Scientific / art field,
discipline and subdiscipline		TECHNICAL SCIENCES
Computing
Abstract
Ovaj rad obrađuje postupke za automatsko prepoznavanje pisanog teksta. Cijeli rad možemo podijeliti u dva glavna dijela: segmentacija slike i klasifikacija slika. Segmentacija slike je područje digitalne obrade slika i računalnog vida. Cilj segmentacije je podijeliti sliku u više dijelova, odnosno segmenata nad kojima je lakše provoditi analizu i daljnju obradu. Segmentacija slike se uobičajeno koristi za pronalaženje objekata i granica objekata (linije, krivulje, itd.) u slikama. U okviru ovog rada implementirana je metoda segmentacije slike pomoću histograma vertikalne projekcije. Jednom kada je slika teksta podijeljena na pojedinačna slova, možemo provoditi klasifikaciju. Klasifikacija je problem koji se rješava algoritmima nadziranog strojnog učenja. Cilj klasifikacije je odrediti klasu y kojoj pripada ulazni primjer x. Algoritmi klasifikacije obrađeni u ovom radu su algoritam k – najbližih susjeda i klasifikacija pomoću konvolucijske neuronske mreže. Preciznost klasifikacija k – najbližim susjedima iznosi oko 66% s parametrom k=2, dok je kod klasifikacije pomoću konvolucijske mreže preciznost jednaka 92.46%.
Abstract (english)
This thesis describes procedures for automatic text recognition from images. The whole work can be divided into two main parts: image segmentation and image classification. Image segmentation is the field of digital image processing and computer vision. The goal of segmentation is to divide the image into several parts, i.e. segments on which is analysis and further processing easily performed. Image segmentation is usually used to find objects and object boundaries (lines, curves, etc.) in images. Within this paper, image segmentation is implemented using method of vertical projection histogram. Once the image is divided into individual letters, we can perform a classification. Classification is a problem that is solved by a supervised machine learning algorithms. The aim of classification is to determine the class to which the input example belongs. Classification algorithms discussed in this paper are the k - nearest neighbors algorithm and the classification using convolutional neural networks. The accuracy of the k - the nearest neighbors algorithm is about 66% with the parameter k = 2, while in the classification using convolutional network the accuracy is equal to 92.46%.
Keywords
Keywords (english)
Languagecroatian
URN:NBNurn:nbn:hr:168:527075
Study programmeTitle: Computing
Study programme type: university
Study level: undergraduate
Academic / professional title: sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka računarstva (sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka računarstva)
Type of resourceText
File originBorn digital
Access conditionsClosed access
Terms of useLicence In copyright icon
Public note
Created on2023-01-14 21:58:08