An infrared image dehazing method based on modified dark channel prior

Siyu Yan; Jingwen Zhu; Kai Yun; Yuexing Wang; Chuangang Xu

doi:10.1117/12.2640246

9 May 2022 An infrared image dehazing method based on modified dark channel prior

Siyu Yan, Jingwen Zhu, Kai Yun, Yuexing Wang, Chuangang Xu

Proceedings Volume 12252, International Conference on Biometrics, Microelectronic Sensors, and Artificial Intelligence (BMSAI); 122520N (2022) https://doi.org/10.1117/12.2640246
Event: International Conference on Biometrics, Microelectronic Sensors, and Artificial Intelligence (BMSAI), 2022, Guangzhou, China

Abstract

To improve the effectiveness of the infrared image dehazing algorithm, we modify the DCP based on the observation of several outdoor haze-free images: in these images, pixels of most local patches no longer have low intensity but are close to high intensity. The transmission map is estimated through the MDCP, and the guided filter and CLAHE achieve enhancement. The haze infrared images can be recovered completely. Moreover, to thrive the infrared dehazing research, we introduce the IR Dense-haze dataset, which contains 13 pairs of synthetic haze infrared images and corresponding haze-free infrared images and six extra haze infrared images captured in natural scenes. Experiments on the IR Densehaze dataset show that our method achieves dominant performance. The average PSNR of the infrared image is 39.02% higher than the original DCP method, and the average SSIM is 28.65% higher than the original DCP method.

Citation Download Citation

Siyu Yan, Jingwen Zhu, Kai Yun, Yuexing Wang, and Chuangang Xu "An infrared image dehazing method based on modified dark channel prior", Proc. SPIE 12252, International Conference on Biometrics, Microelectronic Sensors, and Artificial Intelligence (BMSAI), 122520N (9 May 2022); https://doi.org/10.1117/12.2640246

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