Loading [MathJax]/extensions/MathMenu.js
VROHI: Visibility Recovery for Outdoor Hazy Image in Scattering Media | IEEE Journals & Magazine | IEEE Xplore

VROHI: Visibility Recovery for Outdoor Hazy Image in Scattering Media


Impact Statement:“In this paper, a modified additive haze model (MAHM) is introduced for fast hazy image restoration. Based on this MAHM, a visibility recovery technique for single hazy i...Show More

Abstract:

Additive haze model (AHM), due to its high simplicity, has a potential to increase the efficiency of the restoration procedure of images degraded by scattering media. How...Show More
Impact Statement:
“In this paper, a modified additive haze model (MAHM) is introduced for fast hazy image restoration. Based on this MAHM, a visibility recovery technique for single hazy images called VROHI is further developed. By combining the low-frequency component of hazy image and the saturation prior, VROHI only needs to determine one unknown constant to achieve visibility recovery, and can work well on both mist, heavy haze, inhomogeneous haze, and gray white scene images.”

Abstract:

Additive haze model (AHM), due to its high simplicity, has a potential to increase the efficiency of the restoration procedure of images degraded by scattering media. However, AHM is designed for hazy remote sensing data and is not suitable to be used on outdoor images. In this paper, according to the low-frequency feature (LFC) of haze, AHM is modified via gamma correction technique to make it suitable for modeling outdoor images. Benefitting from the modified AHM (MAHM), a simple yet effective method called VROHI is proposed to enhance the visibility of an outdoor hazy image. In specific, a low complexity LFC extraction method is designed by utilizing characteristic of the discrete cosine transform. Subsequently, by constructing the linear function of unknown parameters and imposing the saturation prior on MAHM, the image dehazing problem can be derived into a global optimization function. Experiments reveal that the proposed VROHI is superior to the other state-of-the-art techniques in terms of both the processing efficiency and recovery quality.
Published in: IEEE Photonics Journal ( Volume: 12, Issue: 6, December 2020)
Article Sequence Number: 7001015
Date of Publication: 10 November 2020

ISSN Information:

Funding Agency:


References

References is not available for this document.