Cover Image

Study of machine learning in computer vision using Raspberry Pi

TA1634.C46 2017

Saved in:
Bibliographic Details
Main Author: Chong, Kah Luo
Format: text::Final Year Project
Language:English US
Published: 2024
Subjects:
Computer vision
Machine learning
Raspberry Pi
Tags: Add Tag
No Tags, Be the first to tag this record!
id my.uniten.dspace-33128
record_format dspace
spelling my.uniten.dspace-331282024-08-04T02:04:47Z Study of machine learning in computer vision using Raspberry Pi Chong, Kah Luo Computer vision Machine learning Raspberry Pi TA1634.C46 2017 Modernisation has made the world to develop in a very rapid pace. As the technology becomes more advanced and innovative, a need in computer vision along with the help of machine learning significantly affects the way of how a typical thing or process is handled and completed especially in monitoring system. Smart monitoring system can reduce the effort of human in performing various tasks and increase the sense of security. This thesis presents and reports the author's effort in studying the use of machine learning algorithms in processing computer vision data using Raspberry Pi, which is a microcomputer. A monitoring system with the capability of detecting motion and human was created using various modules at the end of the project. The major modules used for motion detection were Motion module and OpenCV module. They detected the presence of motion by calculating the change in pixel values and background subtraction in a stream of videos. Pictures of the detected motion were captured and stored for human analysation process. For human detection algorithms, Clarifai, an online cloud machine learning system and Histogram of Oriented Gradient (HOG) along with Linear Support Vector Machine (SVM) in OpenCV were applied and integrated into motion detection modules. Both had machine learning capability where they could determine the presence of human in the supplied images by comparing the extracted features from the images and their own respective database. The positive response returned from these algorithms triggered the notification to the user either through e-mail or through Dropbox. These algorithms were tested and analysed using different images and working situation to discover their efficiencies, accuracies, CPU usages and limitations by adjusting and tuning the parameters presented in the modules. 2024-07-31T08:04:25Z 2024-07-31T08:04:25Z 2017 Resource Types::text::Final Year Project https://irepository.uniten.edu.my/handle/123456789/33128 en_US application/pdf
institution Universiti Tenaga Nasional
building UNITEN Library
collection Institutional Repository
continent Asia
country Malaysia
content_provider Universiti Tenaga Nasional
content_source UNITEN Institutional Repository
url_provider http://dspace.uniten.edu.my/
language English US
topic Computer vision
Machine learning
Raspberry Pi
spellingShingle Computer vision
Machine learning
Raspberry Pi
Chong, Kah Luo
Study of machine learning in computer vision using Raspberry Pi
description TA1634.C46 2017
format Resource Types::text::Final Year Project
author Chong, Kah Luo
author_facet Chong, Kah Luo
author_sort Chong, Kah Luo
title Study of machine learning in computer vision using Raspberry Pi
title_short Study of machine learning in computer vision using Raspberry Pi
title_full Study of machine learning in computer vision using Raspberry Pi
title_fullStr Study of machine learning in computer vision using Raspberry Pi
title_full_unstemmed Study of machine learning in computer vision using Raspberry Pi
title_sort study of machine learning in computer vision using raspberry pi
publishDate 2024
_version_ 1806518000104243200
score 13.222552

Similar Items