Izvajanje modelov strojnega učenja na mobilnih napravah z uporabo programskega ogrodja ML Kit

diplomsko delo

Duško Hristovski (Author), Domen Verber (Mentor)

Abstract

V zaključnem delu smo predstavili programsko ogrodje “ML Kit” (Machine Learning Kit – komplet strojnega učenja) ter njegove modele, tehnologije in lastnosti. Ogrodje omogoča izvajanje vnaprej omogočenih modelov strojnega učenja na mobilnih napravah. V teoretičnem delu smo pojasnili, kaj je strojno učenje in umetna inteligenca in zakaj ogrodje ne potrebuje znanje nobenih. V nadaljevanju smo razložili obstoječe rešitve, podrobnosti posameznih modelov in tehnologij ML Kit-а. Za praktičnem delu smo naredili mobilna aplikacija za skeniranje računov v Android Studio, tako da smo s pomočjo analize in načrtovanja uspešno izvedli model na napravi za prepoznavanje besedila iz slik.

Keywords

strojno učenje;programsko orodje ML Kit;mobilne naprave;diplomske naloge;

Data

Language:	Slovenian
Year of publishing:	2020
Typology:	2.11 - Undergraduate Thesis
Organization:	UM FERI - Faculty of Electrical Engineering and Computer Science
Publisher:	[D. Hristovski]
UDC:	621.397.7-026.26:004.85(043.2)
COBISS:	37258243
Views:	456
Downloads:	53
Average score:	0 (0 votes)
Metadata:

Other data

Secondary language:	English
Secondary title:	Implementing machine learning models on mobile devices using the ML Kit software framework
Secondary abstract:	In this thesis we presented the software framework “ML Kit” with its models, technologies and features. The framework enables the implementation of pre-enabled machine learning models on mobile devices. In the theoretical part, we explained what machine learning and artificial intelligence are and why the framework does not need the knowledge of neither. Below we explained the details of individual models and technologies of ML Kit, as well as other existing solutions. For the practical part, we made a mobile application for scanning invoices in Android Studio. With the help of analysis and planning, we successfully implemented the on device model for recognizing text in images.
Secondary keywords:	machine learning;ML Kit;Android;
Type (COBISS):	Bachelor thesis/paper
Thesis comment:	Univ. v Mariboru, Fak. za elektrotehniko, računalništvo in informatiko, Informatika in tehnologije komuniciranja
Pages:	VIII, 45 f.
ID:	11855055