This project presents an analysis of the current state-of-the-art of the Wow-
bagger’s Face Detection, in order to find a better approach for detecting victims,
in our case dolls. Wowbagger is the name of the robot, who will send regularly
to competitions. The focus is mainly on image processing for autonomous robots
operating for detecting victims in crisis area. The thesis discuses the methods that
are more adequate to be applied in a rescue scenario and some parts will be chosen
to implement an optimal doll detection for our robot.
It will be discussed one of the main objectives of rescue missions is to rescue the
disaster victims. Of course, in some situations it is too risky to send human agents
in the unstable environment. Therefore, robotic agents can be essential in such
situations and important collaborators aside human agents.
One crucial requirement in above-mentioned rescue scenarios is to identify victims
from all other objects.