Razvoj prototipa za procesiranje slikovnega toka brezpilotnega letala

magistrsko delo Organizacija in management informacijskih sistemov

Jure Tarman (Avtor), Robert Leskovar (Mentor)

Povzetek

Raziskava obravnava razvoj prototipa za procesiranje video toka preko kamere na brezpilotnem letalu. Procesiranje video toka je osredotočeno na zaznavanje in prepoznavanje obrazov ali predmetov. Uporabljeno brezpilotno letalo se imenuje Ar.Drone Parrot 2.0. Podan je kratek pregled stanja na področju raziskovanja in razvoja brezpilotnih letal. Opisan je razvoj brezpilotnega letala Ar.Drone. Raziskani so dosežki procesiranja video toka na področju zaznave in prepoznave obrazov. Preizkušene so aplikacije Ardrone Weblight, Ar.Freeflight, Drone, Node JS controller and Camera Feed. Izbrana je najprimernejša aplikacija za osnovo prototipa, ki se imenuje Ardrone Weblight. Raziskano je zaznavanje obraza in sledenje ljudem s pomočjo brezpilotnega letala Ar.Drone. Opisani sta sestava in programska oprema brezpilotnega letala Ar.Drone. Opisan je postopek postavitve delovnega okolja za razvoj prototipa. Izbrana so vsa orodja, ki jih potrebujemo za razvoj. Podani so postopki za namestitev in konfiguracijo orodij. Popisan je test orodja OpenCV, ki skrbi za zaznavanje obrazov in predmetov. Izdelan je prototip, s katerim smo izvedli eksperiment. Za popis eksperimenta je uporabljen del standarda za testiranje ISO/IEEE/IEC 29119. Ugotovljeno je, da programska in strojna oprema ustrezata za rekreativno uporabo. Za zahtevnejšo uporabo je kamera na brezpilotnem letalu neustrezna.

Ključne besede

procesiranje video toka;testiranje programske opreme;ISO/IEEE/IEC 29119;

Podatki

Jezik:	Slovenski jezik
Leto izida:	2016
Tipologija:	2.09 - Magistrsko delo
Organizacija:	UM FOV - Fakulteta za organizacijske vede
Založnik:	[J. Tarman]
UDK:	004
COBISS:	7631635
Št. ogledov:	795
Št. prenosov:	81
Ocena:	0 (0 glasov)
Metapodatki:

Ostali podatki

Sekundarni jezik:	Angleški jezik
Sekundarni naslov:	PROTOTYPE DEVELOPMENT FOR DRONE VIDEO STREAM PROCESSING
Sekundarni povzetek:	The research addresses prototype development for video stream processing made by drone’s camera. Focus of video processing is detection of faces and objects. Product used as drone is Ar.Drone Parrot 2.0. Short analysis of current state in drone researching was given. Development of unmanned vehicle Ar.Drone was described. Achievements on face and object detection were explored. Following applications were tested: Ardrone Weblight, Ar.Freeflight, Drone, Node JS controller and Camera Feed. Ardrone Webflight was chosen as the best one for prototype basic platform. Face detection and tracking people with Ar.Drone was explored. Construction and software of Ar.Drone were described. Procedure of setting up the workplace and tools for prototype development was described. Procedures for installing and configuration tools were given. OpenCV is the tool that enables face and object detection. It has been tested. Prototype was developed and experiment has been done. A part of testing standard ISO/IEEE/IEC 29119 was used. It has been found out that software and hardware is good enough for recreational use. However it is not good enough for professional use.
Sekundarne ključne besede:	Video stream processing;Software testing;ISO/IEEE/IEC 29119;Javascript;Drone;
URN:	URN:SI:UM:
Vrsta dela (COBISS):	Magistrsko delo/naloga
Komentar na gradivo:	Univ. v Mariboru, Fak. za organizacijske vede
Strani:	49 f.
ID:	9148933