Thesis Details

Rozpoznávání písmen pomocí neuronové sítě

Bachelor's Thesis Student: Kluknavský František Academic Year: 2007/2008 Supervisor: Šilhavá Jana, Ing.
English title
Neural Network Letter Recognition
Language
Czech
Abstract

This work uses handwritten character recognition as a model problem for using multilayer perceptron,error backpropagation learning algorithm and finding their optimal parameters, hidden layer size,learning rate and length, ability to handle damaged data. Results were acquired by repeated simulationand testing the neural network using 52,152 English lowercase letters. Best results, smallest networkand shortest learning time was at 60 neurons in the hidden layer and learning rate of 0.01. Biggernetworks achieved the same ability to recognize unknown patterns and higher robustness at highlydamaged data processing.

Keywords

neuron, neural network, letter, overfitting, hidden layer, ocr, backpropagation, learning rate

Department
Department of Computer Graphics and Multimedia FIT BUT
Degree Programme
Information Technology
Files
Status
defended, grade D
Date
10 June 2008
Reviewer
Hradiš Michal, Ing., Ph.D.
Committee
Černocký Jan, prof. Dr. Ing. (DCGM FIT BUT), předseda
Herout Adam, prof. Ing., Ph.D. (DCGM FIT BUT), člen
Herout Pavel, doc. Ing., Ph.D. (WBU in Pilsen), člen
Lukáš Roman, Ing., Ph.D. (DIFS FIT BUT), člen
Martinek David, Ing. (DIFS FIT BUT), člen
Zbořil František V., doc. Ing., CSc. (DITS FIT BUT), člen
Citation
KLUKNAVSKÝ, František. Rozpoznávání písmen pomocí neuronové sítě. Brno, 2008. Bachelor's Thesis. Brno University of Technology, Faculty of Information Technology. 2008-06-10. Supervised by Šilhavá Jana. Available from: https://www.fit.vut.cz/study/thesis/6613/
BibTeX 
@bachelorsthesis{FITBT6613,
    author = "Franti\v{s}ek Kluknavsk\'{y}",
    type = "Bachelor's thesis",
    title = "Rozpozn\'{a}v\'{a}n\'{i} p\'{i}smen pomoc\'{i} neuronov\'{e} s\'{i}t\v{e}",
    school = "Brno University of Technology, Faculty of Information Technology",
    year = 2008,
    location = "Brno, CZ",
    language = "czech",
    url = "https://www.fit.vut.cz/study/thesis/6613/"
}

Theses

Back to top