Integration of Image Processing Algorithm and Path Planning for Search and Rescue Robot

The focus of this project was to explore algorithms and techniques used in motion detection, object recognition and facial recognition, path finding as well as obstacle avoidance. To apply these algorithms, OpenCV was used. In terms of hardware, Raspberry Pi were used to perform image processing...

Full description

Main Author: Wong, Chun Yew
Format: Final Year Project
Language: English
Institution: Universiti Teknologi Petronas
Record Id / ISBN-0: utp-utpedia.22926 /
Published: Universiti Teknologi PETRONAS 2017
Subjects:
Online Access: http://utpedia.utp.edu.my/22926/1/19984_WongChunYew_EE_Jan2017_FYP%20Final%20Report.pdf
http://utpedia.utp.edu.my/22926/
Tags: Add Tag
No Tags, Be the first to tag this record!
Summary: The focus of this project was to explore algorithms and techniques used in motion detection, object recognition and facial recognition, path finding as well as obstacle avoidance. To apply these algorithms, OpenCV was used. In terms of hardware, Raspberry Pi were used to perform image processing and robot movement. To perform image processing various OpenCV commands, such as cv2.GreyScale followed by cv2.GaussianBlur and cv2.DetectContour were combined. Path finding and obstacle avoidance were done by integrating an ultrasonic sensor into the system. Path finding is done by utilizing the coordinates of the bounding box. As a result, the robot turned around, 90 degree each time, to have a view of each of the four directions, in search of the target. Once motion was detected, the robot would stop at that direction and approach until the ultrasonic detected something. The robot would then run a scan on the target using facial recognition to determine whether it is human.