Sustav za prepoznavanje gesti ruku temeljen na podacima s mikrokontrolerske pločice i neuronskoj mreži

undergraduate thesis

• Title (english): System for recognizing hand gestures based on microcontroller board data and neural network
• Author: Eugen Vušak
• Mentor: Alan Jović (mentor)
• Committee member: Alan Jović (predsjednik povjerenstva)
• Committee member: Nikolina Frid (član povjerenstva)
• Committee member: Ivan Kovačević (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2018-07-13, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

U ovom radu pokriven je čitav proces izrade sustava koji koristi neuronske mreže.
Mikrokontrolerska pločica, Arduino Nano, sa senzorom MPU-6050, korištena je kao izvor
podatka. Pomoću nje prikupljeni su podaci o gestama. Za to je bila izrađena aplikacija u
programskom jeziku Processing, koja tražili podatke od mikrokontrolera te dodaje
odgovarajuću oznaku geste na te podatke. Podaci su zatim zapisani u .csv datoteku.
Neuronska mreža koja je implementirana je unaprijedna neuronska mreža, a algoritam za
učenje je algoritam propagacije pogreške unatrag. Nakon što smo istrenirali mrežu i dobili
model koji prepoznaje geste, a taj model je ugrađen u aplikaciju za prepoznavanje u
stvarnom vremenu.

Abstract (english)

This paper covers the entire process of making system using neural networks. The Arduino
Nano, microcontroller board, with the MPU-6050 sensor, was used as a source of data.
Using it gestures data was collected. For this, an application was developed in the
Programming language Processing, that asked for data from the microcontroller and added
the corresponding gesture label to data. The data was then written to a .csv file. The neural
network that was implemented is a feedforward neural network, and the algorithm for
learning is the backpropagation. After we've trained the network and got a model for
recognizing gestures, this model was embedded in the real-time recognition app.

• Keywords: unaprijedne neuronske mreže; mikrokontroler; prepoznavanje; geste; digitalni filter; algoritam propagacije unatrag; gradijentni spust; klasifikacija
• Keywords (english): feedforward neural network; microcontroller; recognition; gesture; digital filter; backpropagation; gradient descent; classification
• Language: croatian
• URN:NBN: urn:nbn:hr:168:601180
• 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: 2019-03-14 02:15:53