Research on low-light image enhancement algorithm based on filtering and data-driven composite

Jiang Wu; Zhengdong Cheng; Yi Chen; Bo Xie

doi:10.1117/12.3016282

1 February 2024 Research on low-light image enhancement algorithm based on filtering and data-driven composite

Jiang Wu, Zhengdong Cheng, Yi Chen, Bo Xie

Proceedings Volume 13068, Fifth International Conference on Optoelectronic Science and Materials (ICOSM 2023); 1306822 (2024) https://doi.org/10.1117/12.3016282
Event: Fifth International Conference on Optoelectronic Science and Materials (ICOSM 2023), 2023, Hefei, China

Abstract

Low-light images have low visibility, poor contrast, and weakened details, which pose significant obstacles for subsequent computer vision tasks. When enhancing low-light images, multiple factors such as brightness, contrast, artifacts, and noise need to be considered, making this problem challenging. Low-light enhancement algorithms based on the Retinex theory are important methods in the field of image enhancement, with a wide range of applications and practicality. In this paper, starting from the low-light image imaging model, we address the issues of color distortion and limited brightness improvement exhibited by traditional Retinex-based algorithms and deep learning algorithms in actual low-light scenarios. We propose a filtering and data-driven composite enhancement algorithm (FADDC) that performs weighted fusion on output and intermediate quantities before outputting them. Experimental results combining SCI-Net with traditional algorithms SSR and MSR demonstrate that this fusion method can effectively enhance image brightness without color distortion, achieving better enhancement results.

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

Citation Download Citation

Jiang Wu, Zhengdong Cheng, Yi Chen, and Bo Xie "Research on low-light image enhancement algorithm based on filtering and data-driven composite", Proc. SPIE 13068, Fifth International Conference on Optoelectronic Science and Materials (ICOSM 2023), 1306822 (1 February 2024); https://doi.org/10.1117/12.3016282

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