Prepoznavanje simbola učinjenih gestama miša

master's thesis

• Title (english): Recognition of Mouse-based Gestures
• Author: Dario Sikirica
• Mentor: Siniša Srbljić (mentor)
• Committee member: Siniša Srbljić (predsjednik povjerenstva)
• Committee member: Stjepan Groš (član povjerenstva)
• Committee member: Dejan Škvorc (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2015-09-25, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

Problem prepoznavanja simbola učinjenih gestama miša može se riješiti jednom od metoda strojnog učenja. Svaki simbol se može predstaviti nizom značajki pomoću kojih se onda trenira model za prepoznavanje simbola. Za prepoznavanje se koristi višeklasna klasifikacija i model stroja s potpornim vektorima. Napravljen je sustav za prepoznavanje riječi koji korisniku omogućava crtanje simbola na digitalnom platnu. Na određenu akciju miša pokreće se sustav za prepoznavanje te se ispisuje nacrtana riječ u tekstualnom obliku. Sustav za prepoznavanje riječi i njegovi podsustavi su ostvareni pomoću programskog jezika Python i radnog okvira za strojno učenje, Scikit. Sustav se vrednuje na skupu primjera za učenje i provjeru.

Abstract (english)

The task of recognizing symbols drawn with mouse-based gestures can be solved with one of the methods of machine learning. Each symbol can be represented as a sequence of features which are used for model training. Multiclass classification and Support Vector Machines are used for symbol recognition. The word recognition system enables the user to draw letters on a digital canvas. After a specific mouse action, the recognized letters are printed out in textual form. Word recognition system and its subsystems are implemented in Python, using the Scikit framework for machine learning. The correctness of the system is validated on the training and test set.

• Keywords: Strojno Učenje; Stroj s Potpornim Vektorima; Višeklasna klasifikacija; Prepoznavanje simbola; Python; Scikit
• Keywords (english): Machine Learning; Support Vector Machines; Multiclass classification; Symbol recognition; Python; Scikit
• Language: croatian
• URN:NBN: urn:nbn:hr:168:579696
• 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: 2018-02-06 11:45:34