Share Email Print
cover

Proceedings Paper • new

IFPSS: intelligence fire point sensing systems in AIoT environments
Author(s): Kun-Ming Yu; Yen-Chiu Chen; Chung-Hsing Liu; Huan-Po Hsu; Ming-Yuan Lei; Nancy Tsai
Format Member Price Non-Member Price
PDF $17.00 $21.00

Paper Abstract

The situation at the scene of the fire is changing rapidly. How to collect and analyze the most immediate fire information, providing the most effective information for disaster decision-making has always been an important issue. This paper proposes an Intelligent Fire Point Sensing System (IFPSS), which proposes fire condition prediction based on artificial intelligence technology as well as large amounts of gas and temperature data in fire scenes collected by IoT devices. The IFPSS collected actual gas and temperature data from the simulation room where the actual fire test was conducted. Taking carbon monoxide (CO) and hydrogen sulfide (H2S) data as an example, the artificial intelligence analysis of IFPSS uses linear regression algorithm to establish artificial intelligence model. After training and testing the model, an accuracy of up to 84.4% predicts whether the fire process is in the very early stages of a fire.

Paper Details

Date Published: 27 November 2019
PDF: 6 pages
Proc. SPIE 11321, 2019 International Conference on Image and Video Processing, and Artificial Intelligence, 113212T (27 November 2019); doi: 10.1117/12.2550315
Show Author Affiliations
Kun-Ming Yu, Chung Hua Univ. (Taiwan)
Yen-Chiu Chen, Chung Hua Univ. (Taiwan)
Chung-Hsing Liu, Chung Hua Univ. (Taiwan)
Huan-Po Hsu, Chung Hua Univ. (Taiwan)
Ming-Yuan Lei, Architecture and Building Research Institute (Taiwan)
Nancy Tsai, Architecture and Building Research Institute (Taiwan)


Published in SPIE Proceedings Vol. 11321:
2019 International Conference on Image and Video Processing, and Artificial Intelligence
Ruidan Su, Editor(s)

© SPIE. Terms of Use
Back to Top