DPCB-CycleGAN: a heterologous image migration algorithm based on attention mechanism and CycleGAN

Xinjie Gu; Gang Liu; Xihong Zhou; Wenbin Zhang

doi:10.1117/12.2657833

20 October 2022 DPCB-CycleGAN: a heterologous image migration algorithm based on attention mechanism and CycleGAN

Xinjie Gu, Gang Liu, Xihong Zhou, Wenbin Zhang

Proceedings Volume 12451, 5th International Conference on Computer Information Science and Application Technology (CISAT 2022); 124514L (2022) https://doi.org/10.1117/12.2657833
Event: 5th International Conference on Computer Information Science and Application Technology (CISAT 2022), 2022, Chongqing, China

Abstract

Infrared images have significant thermal radiation information and high penetration power and are unaffected by illumination. However, the current reality is that infrared images are more challenging to obtain than visible images. The importance of large numbers of paired heterogeneous images is particularly evident in vision tasks such as image fusion. Therefore, it is essential to expand and migrate infrared images using visible images. In this paper, we propose an attentionbased mechanism and CycleGAN heterogenous image migration algorithm which is termed DPCB-CycleGAN. It removes the traditional convolution operation of the generator and combines CBAM with depthwise separable convolution. Feature extraction of infrared and visible source images in both spatial and channel dimensions, increasing the weight of features in salient regions and learning the laws of migration for both types of images, making the generated images closer to the real ones. The method is trained and tested on the TNO public dataset, and the experimental results show it is effective. It has achieved satisfactory results from both a quantitative and qualitative point of view and is very generalizable.

Citation Download Citation

Xinjie Gu, Gang Liu, Xihong Zhou, and Wenbin Zhang "DPCB-CycleGAN: a heterologous image migration algorithm based on attention mechanism and CycleGAN", Proc. SPIE 12451, 5th International Conference on Computer Information Science and Application Technology (CISAT 2022), 124514L (20 October 2022); https://doi.org/10.1117/12.2657833

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
4 PAGES

DOWNLOAD PAPER SAVE TO MY LIBRARY

GET CITATION

RIGHTS & PERMISSIONS

Get copyright permission Get copyright permission on Copyright Marketplace

KEYWORDS

Infrared imaging

Infrared radiation

Image fusion

Visible radiation

Convolution

Image enhancement

Feature extraction

Show All Keywords

Keywords/Phrases

Search In:

Publication Years