izpis_h1_title_alt

Uporaba različnih biometričnih tehnik pri spletnem preverjanju znanja
ID Stavanja, Jaka (Author), ID Lavbič, Dejan (Mentor) More about this mentor... This link opens in a new window

.pdfPDF - Presentation file, Download (1,52 MB)
MD5: 922FF6563828A4801240B0A9137EEEA1

Abstract
Identifikacijo študentov pri reševanju kvizov preko spletne učilnice lahko trenutno izvajamo s preverjanjem posamezne osebe v živo. Primerjati moramo njen izgled s tistim iz slike na osebni oziroma študentski izkaznici. Če hočemo zmanjšati čas, porabljen za verifikacijo identitete, potrebujemo drug, učinkovitejši način overjanja. Časovni problem odpravimo s pomočjo razvoja vtičnika za programsko opremo Moodle in njemu pripadajoče zaledne logike. Ta zna s pomočjo spletnih obrazcev zajemati zapise stila tipkanja in slike obraza. Skupaj z vtičnikom razvijemo tudi zaledni sistem, ki zna primerjati slike obraza s pomočjo odprtokodne knjižnice OpenFace in računati razlike med zapisi stila tipkanja. Zanesljivost našega sistema na koncu tudi testiramo na prostovoljcih in ugotovimo, da je točnost sistema, ob uporabi obeh tipov primerjave, na prostovoljcih zelo visoka.

Language:Slovenian
Keywords:biometrika, identiteta, spletno preverjanje znanja
Work type:Bachelor thesis/paper
Organization:FRI - Faculty of Computer and Information Science
Year:2018
PID:20.500.12556/RUL-102883 This link opens in a new window
Publication date in RUL:11.09.2018
Views:706
Downloads:148
Metadata:XML RDF-CHPDL DC-XML DC-RDF
:
Copy citation
Share:Bookmark and Share

Secondary language

Language:English
Title:Use of biometric techniques in online exams
Abstract:
The identification of students in exams is currently done by physically checking each and every student if their appearance and identity match the one on their student ID cards. If we wish to shorten the time needed to check students' identities, we need a more efficient way of verification. We solve this problem by developing a plugin for the Moodle learning management software, which is able to record prints of typing styles and images of faces by displaying an input form. Along with the plugin, we develop a backend system, that can distinguish faces using the OpenFace open-source library and typing styles. Finally, we test the system's reliability on volunteers and see that its accuracy is very high when using both comparison techniques.

Keywords:biometrics, identity, web, quiz, knowledge

Similar documents

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

Back