VGU RESEARCH REPOSITORY
Please use this identifier to cite or link to this item:
https://epub.vgu.edu.vn/handle/dlibvgu/1000
DC Field | Value | Language |
---|---|---|
dc.contributor.author | Tran Hong Dai Phat | en_US |
dc.date.accessioned | 2021-03-09T03:29:05Z | - |
dc.date.available | 2021-03-09T03:29:05Z | - |
dc.date.issued | 2020 | - |
dc.identifier.uri | http://epub.vgu.edu.vn/handle/dlibvgu/1000 | - |
dc.description.abstract | In the modern society, the attention relating to human lives and properties protection is constantly increasing. Early fire hazard detection approaches become the most important topic in many research realms. Therefore, in this thesis, Internet of Things (IoT) concepts, machine learning technologies are implemented for constructing an innovative, affordable and effective approach in order to early detect then alert fire outbreaks. The general idea of the approach is calculating a possibility of the fire hazard using a machine learning model based on environment’s metrics collected from IoT devices. When the probability exceeds a pre-defined threshold, users are alerted to respond accordingly. In addition, the machine learning model has a self-improvement module which automatically adapts to user’s feedback to enhance the accuracy. In order to prove the initial hypothesis, a demonstrative application is introduced in this thesis. The application collects the environment’s metrics by implementing a combination of embedded devices. A machine learning model is integrated to train and analyze the input vectors to output the possibility of fire, before it happens. Dataset for training machine learning model was mainly generated by employing embedded IoT devices of the system. In order to allow machine learning, improve from user’s responses, notification service is applied for collecting feedback. For visualizing the proposed concept, an interactive web and mobile web application is designed and created. The concept is suitable for deploying in apartments, households or small businesses because the system is only effective for detecting fire outbreak in moderate areas. | en_US |
dc.language.iso | en | en_US |
dc.title | Early fire detection system with machine learning and embedded system | en_US |
dc.type | Thesis | en_US |
item.grantfulltext | restricted | - |
item.fulltext | With Fulltext | - |
item.languageiso639-1 | other | - |
Appears in Collections: | Computer Science (CS) |
Files in This Item:
File | Description | Size | Format | Existing users please Login |
---|---|---|---|---|
Early fire detection system with machine learning and embedded system.pdf | 2.47 MB | Adobe PDF |
Page view(s)
121
checked on May 5, 2024
Download(s)
43
checked on May 5, 2024
Google ScholarTM
Check
Items are protected by © Copyright of Vietnamese - German University Library