Author(s)
Term
4. term
Publication year
2013
Submitted on
2013-06-04
Pages
118 pages
Abstract
Dette speciale beskriver et system til at detektere tilskadekomne i et katastrofescenarie, som kan implementeres i software ved anvendelse af velkendte computer vision metoder. Målet har været at skabe et system som kan behandle video input fra kameraet på en robot. Det første kapitel leverer baggrund for katastrofehåndtering på verdensplan og undersøger de situationer hvor robotter har været anvendt i redningsarbejdet efter en katastrofe er indtruffet. Til dags dato har robotter endnu ikke spillet nogen central rolle i søgning- og redningsaktioner efter en større katastrofe. Formålet med detektionssystemet er at det skal kunne anvendes som et visuelt hjælpemiddel for en person som styrer robotten. Dette kan forhåbentlig forøge chancerne for at finde tilskadekomne i situationer hvor operatøren er belastet mentalt, hvilket har været tilfældet under aktioner med robotter. Det endelige detektionssystem består af 3 detektionsmoduler som alle søger efter det menneskelige hoved eller ansigt i billeder. Den første detektor anvender HOG descriptor’en og en SVM classifier til at bestemme om en mindre billedregion indeholder et hoved. Den anden detektor anvender Viola-Jones object detection framework til at finde ansigter i billedet. Den tredje detektor anvender template matching til at søge efter hovedet ved anvendelse af kvartcirkelformer. Kombinationen af detektorer opnåede en detektionsrate på 62 % med et varieret billedsæt med mennesker. På et simuleret katastrofesæt opnåede detektorerne kun en detektionrate på 14 %.
This thesis describes how a system for detecting victims in a disaster scenario can be implemented in software using known methods from computer vision. The goal has been to create a system that can process video input from the camera on a robot. The first chapter provides background on disaster management worldwide and explores the cases where robots have been deployed for rescue work after a disaster has struck. To this day, robots have not played a central role in search and rescue operations after any large disaster. The purpose of the human detection system is to work as a visual aid to a human robot operator. This can hopefully increase the chance of locating victims if the operator is burdened by mental fatigue as has been the case during robot operations. The final detection system consists of 3 detection modules that all search for the human head or face in images. The first detector uses the HOG descriptor and an SVM classifier to determine if a human head is present within a smaller region of the image. The second detector uses the Viola-Jones object detection framework to detect human faces. The third detector uses template matching to search for the head using quarter circle shapes. The combined detection system achieved a detection rate of up to 62 % on a varied set of image with humans. On a simulated disaster image set the detector has a detection rate of only 14 %.
Keywords
Documents
Colophon: This page is part of the AAU Student Projects portal, which is run by Aalborg University. Here, you can find and download publicly available bachelor's theses and master's projects from across the university dating from 2008 onwards. Student projects from before 2008 are available in printed form at Aalborg University Library.
If you have any questions about AAU Student Projects or the research registration, dissemination and analysis at Aalborg University, please feel free to contact the VBN team. You can also find more information in the AAU Student Projects FAQs.