Super resolution of large field of view infrared image based on sparse residual auto-encoder

Yu-dan Chen; Gu-Yue Chen; Sheng-jin Gu; Gang Li

doi:10.1117/12.2687943

6 November 2023 Super resolution of large field of view infrared image based on sparse residual auto-encoder

Yu-dan Chen, Gu-Yue Chen, Sheng-jin Gu, Gang Li

Proceedings Volume 12921, Third International Computing Imaging Conference (CITA 2023); 129210N (2023) https://doi.org/10.1117/12.2687943
Event: Third International Computing Imaging Conference (CITA 2023), 2023, Sydney, Australia

Abstract

Large field of view infrared images have low resolution and few high-frequency details for its imaging characteristics. So, the super-resolution reconstruction algorithms that has good results in visible image may not be suitable. Therefore, it is important to study the super-resolution of large field of view infrared images. Based on convolutional auto-encoders, a residual sparse Auto-encoders was constructed. The residual sparse auto-encoders was composed of ten convolution layers, two maxpooling layers and two upsampling layers. Consulting the residual theory, the first layer was added to the last layer and the second layer was added to the second to last layer. To achieve a better result, the spare was also added in the model. In order to analyzed the result of super-resolution, the classical algorithms of Bicubic and SRCNN were discussed. With the image and the data, it can be showed that result of the residual sparse Auto-encoders was better than the Bicubic and SRCNN.

Citation Download Citation

Yu-dan Chen, Gu-Yue Chen, Sheng-jin Gu, and Gang Li "Super resolution of large field of view infrared image based on sparse residual auto-encoder", Proc. SPIE 12921, Third International Computing Imaging Conference (CITA 2023), 129210N (6 November 2023); https://doi.org/10.1117/12.2687943

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