Fire image segmentation based on transfer learning and fully convolutional neural network

Yuan-Bin Wang; Zong-You Duan; Hai-Long Huang

doi:10.1117/12.2652593

6 September 2022 Fire image segmentation based on transfer learning and fully convolutional neural network

Yuan-Bin Wang, Zong-You Duan, Hai-Long Huang

Proceedings Volume 12332, International Conference on Intelligent Systems, Communications, and Computer Networks (ISCCN 2022); 1233211 (2022) https://doi.org/10.1117/12.2652593
Event: International Conference on Intelligent Systems, Communications, and Computer Networks (ISCCN 2022), 2022, Chengdu, China

Abstract

Aiming at the problem of low accuracy for fire image segmentation, this paper proposes a method based on full convolution neural network. Firstly, on the basis of SegNet network, this method reduces the complexity by appropriately deleting the number of network layers, and the SegNet model becomes a lightweight full convolution neural network model. Then, the parameters of the first four convolution layers of VGG-16 trained based on multiple data sets are transferred to the coding part of the full convolution neural network. According to the training, verification and test, the model based on the optimal parameters is selected to segment the fire area of the image, and the segmentation results are obtained. Finally, the experimental results show that this method can completely segment the fire area from the image. Compared with other common methods, the segmentation accuracy of this method is higher, its accuracy can reach 93%, and the time-consuming is 0.12s.

Citation Download Citation

Yuan-Bin Wang, Zong-You Duan, and Hai-Long Huang "Fire image segmentation based on transfer learning and fully convolutional neural network", Proc. SPIE 12332, International Conference on Intelligent Systems, Communications, and Computer Networks (ISCCN 2022), 1233211 (6 September 2022); https://doi.org/10.1117/12.2652593

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
11 PAGES

DOWNLOAD PAPER SAVE TO MY LIBRARY

GET CITATION

RIGHTS & PERMISSIONS

Get copyright permission Get copyright permission on Copyright Marketplace

KEYWORDS

Image segmentation

Data modeling

Image processing algorithms and systems

Convolution

Neural networks

Convolutional neural networks

Deconvolution

Show All Keywords

Keywords/Phrases

Search In:

Publication Years