Aplikacija za prepoznavanje rukom pisanih brojeva pomoću konvolucijske neuronske mreže

master's thesis

• Title (english): Application for Recognition of Hand Written Numbers By Convolutional Neural Network
• Author: Mate Masnov
• Mentor: Tomislav Burić (mentor)
• Committee member: Tomislav Burić (predsjednik povjerenstva)
• Committee member: Ilko Brnetić (član povjerenstva)
• Committee member: Mate Puljiz (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2020-07-14, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

Glavna je tema ovog rada primjena metode optičkog prepoznavanja znakova (engl. optical character recognition - OCR) za prepoznavanje rukom pisanih brojeva. U ovom radu analizirani su algoritmi strojnog učenja i njihova primjena za prepoznavanje znakova, s naglaskom na neuronske mreže i prepoznavanje brojeva. Prepoznavanje brojeva ostvareno je pomoću modela konvolucijske neuronske mreže koji je povezan s iOS mobilnom aplikacijom koja dohvaća uzorke za prepoznavanje. Model je treniran na MNIST... More setu podataka. U radu su opisani i postupci ispitivanja modela te dobiveni rezultati za nekoliko različitih modela konvolucijskih mreža. Less

Abstract (english)

The main topic of this paper is the application of optical character recognition methods for handwritten number recognition. This paper analyzes machine learning algorithms and their application in character recognition, with emphasis on neural networks and number recognition. Number recognition is achieved using a convolutional neural network model connected to an iOS mobile application that retrieves samples used for recognition. The model was trained on the MNIST data set. The paper also... More describes the testing procedures and the results obtained using several different models of convolutional networks. Less

• Keywords: optičko prepoznavanje znakova; strojno učenje; konvolucijske neuronske mreže; MNIST set podataka; iOS aplikacija
• Keywords (english): optical character recognition; machine learning; convolutional neural networks; MNIST dataset; iOS application
• Language: croatian
• URN:NBN: urn:nbn:hr:168:617798
• 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 10:22:12