The aim of this work is to design algorithm that will detect pupil from video. The theoretical knowledge necessary for proper pupil detection is also described in this master’s thesis. Detection is done on 24 videos that are converted to single images. The complete result the dependence of the pupil diameter on the used dioptre. The overall success rate of the algorithm is 88,13 %. The overall error is 11,87 %. For 17 out of 24 patients, it is confirmed that the greater the dioptre, the larger the pupil.
Identifer | oai:union.ndltd.org:nusl.cz/oai:invenio.nusl.cz:401002 |
Date | January 2019 |
Creators | Musilová, Kateřina |
Contributors | Harabiš, Vratislav, Kolář, Radim |
Publisher | Vysoké učení technické v Brně. Fakulta elektrotechniky a komunikačních technologií |
Source Sets | Czech ETDs |
Language | Czech |
Detected Language | English |
Type | info:eu-repo/semantics/masterThesis |
Rights | info:eu-repo/semantics/restrictedAccess |
Page generated in 0.0017 seconds