Lightweight fire detection algorithm based on deep learning

Wenhao Bi; Bangxu Li; Bo Lei

doi:10.1117/12.3022816

18 March 2024 Lightweight fire detection algorithm based on deep learning

Wenhao Bi, Bangxu Li, Bo Lei

Proceedings Volume 13104, Advanced Fiber Laser Conference (AFL2023); 1310429 (2024) https://doi.org/10.1117/12.3022816
Event: Advanced Fiber Laser Conference (AFL2023), 2023, Shenzhen, China

Abstract

Fire detection is an important measure to protect public safety, avoid casualties and property damage. With the increasing development of artificial intelligence algorithms, deep learning methods based on convolutional neural networks have been applied to fire detection. Although deep learning based fire detection algorithms have made significant progress, there are still problems such as insufficient and imbalanced data, low model accuracy, and insufficient real-time performance. Therefore, we designs a lightweight fire detection algorithm based on deep learning. This method is based on the object detection algorithm YOLOv5. Firstly, a data augmentation algorithm is used to solve the dataset problem, followed by a lightweight improvement on the algorithm's backbone network to improve detection speed. Finally, a feature fusion improvement is performed on the algorithm's neck to improve accuracy. The experimental results show that the above method slightly improves accuracy while reducing weight, and can be better applied to actual fire detection scenarios.

(2024) Published by SPIE. Downloading of the abstract is permitted for personal use only.

Citation Download Citation

Wenhao Bi, Bangxu Li, and Bo Lei "Lightweight fire detection algorithm based on deep learning", Proc. SPIE 13104, Advanced Fiber Laser Conference (AFL2023), 1310429 (18 March 2024); https://doi.org/10.1117/12.3022816

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available

Members: $17.00

Non-members: $21.00 ADD TO CART

PROCEEDINGS
6 PAGES

DOWNLOAD PAPER SAVE TO MY LIBRARY

GET CITATION

RIGHTS & PERMISSIONS

Get copyright permission Get copyright permission on Copyright Marketplace

KEYWORDS

Fire

Detection and tracking algorithms

Evolutionary algorithms

Object detection

Convolution

Deep learning

Forest fires

Show All Keywords

Keywords/Phrases

Search In:

Publication Years