izpis_h1_title_alt

Prepoznava obrazov na mobilnih platformah
ID FERBIŠEK, GORAN (Author), ID Batagelj, Borut (Mentor) More about this mentor... This link opens in a new window

.pdfPDF - Presentation file, Download (1,88 MB)
MD5: 5181E750BB88E535F7EFDF6125BBA194
PID: 20.500.12556/rul/16e6b0a1-ca09-40a8-9a90-aa102a035929

Abstract
Področje prepoznave obrazov se največ ukvarja s programi, ki so namenjeni varnostnim sistemom, malo pa je aplikacij namenjenih zabavi. Večina teh programov je napisanih za namizne računalnike. Zato smo se odločili izdelati aplikacijo zabavne narave, ki deluje na mobilnem operacijskem sistemu Android. Aplikacija je namenjena primerjavi z znanimi nogometnimi igralci. Uporabnik usmeri kamero pametnega telefona tako, da ujame svoj ali pa prijateljev obraz. Aplikacija prikaže kateremu znanemu nogometašu v bazi slik je zajeti obraz najbolj podoben. Naredili smo pregled obstoječih aplikacij za prepoznavo obrazov na mobilnih platformah. Pregledali smo različne metode za prepoznavo obrazov, med njimi pa smo opisali implementacijo metode glavnih komponent. Algoritem za prepoznavo obrazov smo testirali na bazi obraznih slik FERET. Opisali smo postopek za pripravo obrazov na prepoznavo, tako da dajejo najboljše rezultate.

Language:Slovenian
Keywords:prepoznava obrazov, metoda glavnih komponent, Android, openCV, FERET
Work type:Bachelor thesis/paper
Organization:FRI - Faculty of Computer and Information Science
Year:2017
PID:20.500.12556/RUL-90989 This link opens in a new window
Publication date in RUL:06.03.2017
Views:2221
Downloads:353
Metadata:XML DC-XML DC-RDF
:
Copy citation
Share:Bookmark and Share

Secondary language

Language:English
Title:Face recognition on mobile platforms
Abstract:
The field of face recognition is dealing mostly with security system applications. Only a small portion of these applications are made with entertainment in mind. Most of the software on the market is written for personal computers. This was our motivation to make an entertainment application for Android mobile operating system. The application use is intended for comparing persons faces to celebrities. In our case, we chose football players. The user directs the smart phone's camera towards a persons face and captures it. As a result, the application shows which face in the database, it resembles the most. We made an overview of similar applications currently available. We also described several face recognition methods. Our implementation of Principal Component Analysis was described in detail. Our method was tested on FERET database of facial images. We described the procedure of image preparation for best results possible.

Keywords:face recognition, Principal Component Analysis, Android, openCV, FERET

Similar documents

Similar works from RUL:
Similar works from other Slovenian collections:

Back