Vrednovanje raspoznavanja koštica maslina konvolucijskim neuronskim mrežama

undergraduate thesis

• Title (english): Experimental evaluation of recognizing olive pits by convolutional neural networks
• Author: Dario Pažin
• Mentor: Siniša Šegvić (mentor)
• Committee member: Siniša Šegvić (predsjednik povjerenstva)
• Committee member: Mladen Karan (član povjerenstva)
• Committee member: Goran Glavaš (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2015-07-13, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

U radu su predstavljene umjetne neuronske mreže i njihovo proširenje konvolucijske neuronske mreže koje su korištene u radu. Opisan je način rada, izgled arhitekture koja je korištena i predstavljeni eksperimentalni rezultati. Cilj ovog rada bila je klasifikacija sorti maslina pomoću pripadajućih koštica. Kao fizikalno svojstvo po kojem se koštice masline razlikuju odabran je oblik masline. Pronađene su točke na konturi masline i predane kao ulaz konvolucijskoj neuronskoj mreži. Za rad s mrežom korištena je biblioteka Caffe, dane su upute za njenu instalaciju, objašnjeno korištenje te opisani njeni slojevi i parametri. Na kraju su predložene ideje za poboljšanje točnosti klasifikacijskog postopka.

Abstract (english)

This paper is about artificial neural networks and convolutional neural network that was used in the experimental work. The main goal of this paper was classification of different types of olives using their pits. We can difference olive sort by looking at the shape of the pits, so the plan was to find points on the pit contour. Points became convolutional neural network input. Framework Caffe was used to work with convolutional neural networks. Caffe installation, user instructions and network architecture were presented in this paper. In the end, experimental results of classification were shown and some ideas for classification results improvement were given.

• Keywords: umjetna neuronska mreža; konvolucijaska neuronska mreža; maslina; koštica; Caffe; višeklasna klasifikacija; Google Protocol Buffer
• Keywords (english): artificial neural network; convolutional neural network; olive; pit; Caffe; multiclass classification; Google Protocol Buffer
• Language: croatian
• URN:NBN: urn:nbn:hr:168:307082
• 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: 2018-02-06 11:07:35