prikaz prve stranice dokumenta Interaktivna semantička segmentacija rukom pisanih znakova
No public access
undergraduate thesis
Interaktivna semantička segmentacija rukom pisanih znakova
Zagreb: University of Zagreb, Faculty of Electrical Engineering and Computing, 2017. urn:nbn:hr:168:959856
Blažić, Katarina
University of Zagreb
Faculty of Electrical Engineering and Computing

Institutional repository: FER Repository

Cite this document

Blažić, K. (2017). Interaktivna semantička segmentacija rukom pisanih znakova (Undergraduate thesis). Zagreb: University of Zagreb, Faculty of Electrical Engineering and Computing. Retrieved from https://urn.nsk.hr/urn:nbn:hr:168:959856

Blažić, Katarina. "Interaktivna semantička segmentacija rukom pisanih znakova." Undergraduate thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, 2017. https://urn.nsk.hr/urn:nbn:hr:168:959856

Blažić, Katarina. "Interaktivna semantička segmentacija rukom pisanih znakova." Undergraduate thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, 2017. https://urn.nsk.hr/urn:nbn:hr:168:959856

Blažić, K. (2017). 'Interaktivna semantička segmentacija rukom pisanih znakova', Undergraduate thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, accessed 26 December 2024, https://urn.nsk.hr/urn:nbn:hr:168:959856

Blažić K. Interaktivna semantička segmentacija rukom pisanih znakova [Undergraduate thesis]. Zagreb: University of Zagreb, Faculty of Electrical Engineering and Computing; 2017 [cited 2024 December 26] Available at: https://urn.nsk.hr/urn:nbn:hr:168:959856

K. Blažić, "Interaktivna semantička segmentacija rukom pisanih znakova", Undergraduate thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, Zagreb, 2017. Available at: https://urn.nsk.hr/urn:nbn:hr:168:959856

Metadata
TitleInteraktivna semantička segmentacija rukom pisanih znakova
Title (english)Interactive semantic segmentation of handwritten text
AuthorKatarina Blažić
MentorSiniša Šegvić (mentor)
Committee memberSiniša Šegvić (predsjednik povjerenstva)
Committee memberPetra Bevandić (član povjerenstva)
Committee memberIvan Krešo (član povjerenstva)
GranterUniversity of Zagreb
Faculty of Electrical Engineering and Computing
Zagreb
Defense date and country2017-07-10, Croatia
Scientific / art field,
discipline and subdiscipline		TECHNICAL SCIENCES
Computing
Abstract
U ovom radu su obrađena dva pristupa semantičkoj segmentaciji rukom pisanog teksta. Prvi pristup je statistički temeljen na histogramima boja i uvjetnim vjerojatnostima. Pritom je analiziran utjecaj koji veličina histograma i prag imaju na klasifikaciju. Drugi pristup je temeljen na konvolucijskim neuronskim mrežama. Pronađena je pojednostavljena arhitektura konvolucijske mreže VGG koja uz malo treniranja daje dobre rezultate. Prikazan je način i intenzitet označavanja ulaznih slika te su oba pristupa testirana na ulaznom skupu slika pri čemu su dali zadovoljavajuće rezultate.
Abstract (english)
This paper analyzes two approaches to the semantic segmentation of a handwritten text. The first approach is statistically based on color histograms and conditional probabilities. Thereby the effect of histogram size and threshold on the classification is analyzed. The second approach is based on convolutional neural networks. We simplified the architecture of the VGG convolution network that gives good results with little training. The way and intensity of labeling input images is shown and both approaches are tested and provide satisfactory results.
Keywords
Keywords (english)
Languagecroatian
URN:NBNurn:nbn:hr:168:959856
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 on2019-03-13 19:08:57