Design of an unmanned aerial vehicle 1
This report presents the design and implementation of an unmanned aerial vehicle (UAV) that is enhanced with reconnaissance and surveillance capabilities. The main focus of the report will be on human detection, this will be achieved by the use of a miniature computer installed on-board the UAV. Th...
Saved in:
Main Author: | |
---|---|
Other Authors: | |
Format: | Final Year Project |
Language: | English |
Published: |
2016
|
Subjects: | |
Online Access: | http://hdl.handle.net/10356/67861 |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Institution: | Nanyang Technological University |
Language: | English |
id |
sg-ntu-dr.10356-67861 |
---|---|
record_format |
dspace |
spelling |
sg-ntu-dr.10356-678612023-07-07T16:06:19Z Design of an unmanned aerial vehicle 1 Fun, Joel Bo Wen Cheng Tee Hiang School of Electrical and Electronic Engineering DRNTU::Engineering This report presents the design and implementation of an unmanned aerial vehicle (UAV) that is enhanced with reconnaissance and surveillance capabilities. The main focus of the report will be on human detection, this will be achieved by the use of a miniature computer installed on-board the UAV. There are many methods to achieve human detection, such as with the use of pyroelectric infrared sensors. However, in recent years, the field of computer vision has made significant progress, thus this report will explore human detection using research in computer vision. The implementation of the human detection was done using the Histogram of Oriented Gradient (HOG) feature descriptor. It works on the principle that an image will be partitioned into smaller segments know as a cell, and these cells are represented by a ‘star’ that shows the strength of the edge orientations. Together, these cells form a histogram of gradient directions. A Support Vector Machine (SVM) uses these histograms for classification with the aid of a sliding window detector. In addition, a second method of human detection using haar-like features will also be explored. Methods to reduce the load imposed on the central processing unit (CPU) were also explored, such as increasing the step size of the detection window as well as reducing the amount of processing that had to be done by the CPU. In doing so, detection accuracy decreases, thus optimization was done in an attempt to improve performance while not affecting detection accuracy to a large extent. Bachelor of Engineering 2016-05-23T04:19:47Z 2016-05-23T04:19:47Z 2016 Final Year Project (FYP) http://hdl.handle.net/10356/67861 en Nanyang Technological University 77 p. application/pdf |
institution |
Nanyang Technological University |
building |
NTU Library |
continent |
Asia |
country |
Singapore Singapore |
content_provider |
NTU Library |
collection |
DR-NTU |
language |
English |
topic |
DRNTU::Engineering |
spellingShingle |
DRNTU::Engineering Fun, Joel Bo Wen Design of an unmanned aerial vehicle 1 |
description |
This report presents the design and implementation of an unmanned aerial vehicle (UAV) that is enhanced with reconnaissance and surveillance capabilities. The main focus of the report will be on human detection, this will be achieved by the use of a miniature computer installed on-board the UAV.
There are many methods to achieve human detection, such as with the use of pyroelectric infrared sensors. However, in recent years, the field of computer vision has made significant progress, thus this report will explore human detection using research in computer vision.
The implementation of the human detection was done using the Histogram of Oriented Gradient (HOG) feature descriptor. It works on the principle that an image will be partitioned into smaller segments know as a cell, and these cells are represented by a ‘star’ that shows the strength of the edge orientations. Together, these cells form a histogram of gradient directions. A Support Vector Machine (SVM) uses these histograms for classification with the aid of a sliding window detector. In addition, a second method of human detection using haar-like features will also be explored.
Methods to reduce the load imposed on the central processing unit (CPU) were also explored, such as increasing the step size of the detection window as well as reducing the amount of processing that had to be done by the CPU. In doing so, detection accuracy decreases, thus optimization was done in an attempt to improve performance while not affecting detection accuracy to a large extent. |
author2 |
Cheng Tee Hiang |
author_facet |
Cheng Tee Hiang Fun, Joel Bo Wen |
format |
Final Year Project |
author |
Fun, Joel Bo Wen |
author_sort |
Fun, Joel Bo Wen |
title |
Design of an unmanned aerial vehicle 1 |
title_short |
Design of an unmanned aerial vehicle 1 |
title_full |
Design of an unmanned aerial vehicle 1 |
title_fullStr |
Design of an unmanned aerial vehicle 1 |
title_full_unstemmed |
Design of an unmanned aerial vehicle 1 |
title_sort |
design of an unmanned aerial vehicle 1 |
publishDate |
2016 |
url |
http://hdl.handle.net/10356/67861 |
_version_ |
1772825599269666816 |