Automatska klasifikacija broja dodijeljenih bodova na skeniranom obrascu za ispravljanje ispita

undergraduate thesis

• Title (english): Automatic Classification of the Number of Points Awarded on the Scanned Exam Form
• Author: Josip Mihaljević
• Mentor: Dario Bojanjac (mentor)
• Committee member: Dario Bojanjac (predsjednik povjerenstva)
• Committee member: Filip Turčinović (član povjerenstva)
• Committee member: Marin Kačan (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2022-07-08, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

U ovom završnom radu se opisuju osnovni koncepti dubokog učenja. Razrađuje se pojam umjetnih neuronskih mreža, opisuje se arhitektura mreže i najčešće korišteni algoritmi u dubokom učenju. Razrađuje se problem segmentacije slike te su ponuđeni postupci rješavanja problema pronalaska crvenog kvadratića i segmentacije znamenki. Dalje su opisani osnovni koncepti konvolucijskih neuronskih mreža te njihova primjena u rješavanju problema klasifikacije rukom pisanih znamenki. U nastavku je prikazan model konvolucijske neuronske mreže za rješavanje problema klasifikacije i dan je predložak korištene programske potpore. Na kraju su prikazani eksperimentalni rezultati.

Abstract (english)

This final thesis describes the basic concepts of deep and machine learning. The concept of artificial neural networks is elaborated, the architecture of the network and the most frequently used algorithms in deep learning are described. The problem of image segmentation is developed and procedures for solving the problem of finding the red square and segmenting digits are offered. The basic concepts of convolutional neural networks and their application in solving the problem of classification of handwritten digits are described below. Hereafter is a model of a convolutional neural network for solving classification problems and a template of the software used is given. Finally, the experimental results are presented.

• Keywords: duboko učenje; umjetne neuronske mreže; konvolucijske neuronske mreže; segmentacija slike; segmentacija rukom pisanih znamenki; klasifikacija
• Keywords (english): deep learning; artificial neural networks; convolutional neural networks; image segmentation; handwritten digits segmentation; classification
• Language: croatian
• URN:NBN: urn:nbn:hr:168:282124
• Study programme: Title: 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 resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2023-01-15 10:51:16