Fog Removal From Images And Video

Current vision systems are designed to perform in normal weather conditions. However, no one can escape from bad weather conditions. Bad weather reduces scene contrast and visibility, which results in a degradation in the performance of various computer vision algorithms such as object tracking, segmentation, and recognition. Under fog weather conditions, the contrast and color of the images are drastically degraded. The degradation level increases with the distance from the camera to the object. Two fundamental phenomena that cause loss of visibility are attenuation and airlight. The light beam coming from a scene point gets attenuated by scattering due to atmospheric particles. This phenomenon is termed attenuation which reduces contrast in the scene. The light coming from the source is scattered towards the camera and leads to the shift in color. This phenomenon is termed as airlight. Airlight increases with the distance from the object. In order to remove the fog,  first preprocessing (i.e. histogram equalization) is performed over the foggy image. This preprocessing increases the contrast of the image prior to the fog removal and results in a better estimation of the airlight map. The initial value of the airlight map is estimated by the dark channel prior. Once the airlight map is obtained, the image is restored. Histogram stretching of the output image is performed as post-processing. This histogram stretched image is the final defogged image. A fast fog removal algorithm for the image can be tried for the video fog removal. However, there are many challenges in the design of the video fog removal algorithm. For real-time implementation, one needs to find the fastest algorithm with quality output. For real-time implementation, the conventional frame by frame approach is computationally expensive. If temporal correlations between neighboring frames can be utilized for restoration, it promises a large amount of saving in computation.

Highlights:

1. National Award for Technology Start-Ups sponsored by Technology Development Board, DST in 2021

2. Chunauti 3.0 from STPI in 2022

3. Our start-up named ProficientVision Solutions Private Limited received CII Statrtupreneur Award, 2019 in manufacturing. 

4. INAE has acknowledged the work of technology transition by awarding Abdul Kalam Technology Innovation National Fellowship to Prof. S Mukhopadhyay (2018).

5. "Real time Fog Removal from Video and Rain Removal from Video" received FICCI R&D Awards 2017 under GLOBAL R&D SUMMIT 2017 on April 6-7, 2017 at The Lalit Ashok, Kumara Krupa High Grounds, Bengaluru, India.

6. "Real Time Fog Removal from Video in Application to Safety and Surveillance" ranked 5th in the list of Top 50 Innovators in the 2016 DST - Lockheed Martin India Innovation Growth Programme

Patents:

1. Bijaylaxmi Das, Sudipta Mukhopadhyay, " Method and apparatus for haze/fog removal from a single image with color correction", Patent filing no. TEMP/E-1/63469/2022-KOL dated  27-09-2022.  

2.A. K. Tripathi and S. Mukhopadhyay, Method and System for Removal of Fog from the Images and Videos, Indian Patent (App. No. 1029/KOL/2011), PCT: App. No. PCT/IN2012/000017

Publications:

1. B Das, JP Ebenezer, S Mukhopadhyay, 'A comparative study of single image fog removal methods', The Visual Computer, Springer, November 2020, pp 1-17.

2. JP Ebenezer, B Das, S Mukhopadhyay, 'Single Image Haze Removal Using Conditional Wasserstein Generative Adversarial Networks', 27th European Signal Processing Conference (EUSIPCO), 1-5,2019

3. A. K. Tripathi and S. Mukhopadhyay, 'Efficient fog removal from video", Signal image and video processing, Springer, SIViP (2014) 8:1431–1439

4.  A. K. Tripathi and S. Mukhopadhyay, 'Single Image Fog Removal Using Anisotropic Diffusion", IET image processing, Vol. 6, Issue 7, October 2012, pp. 966-975.

5. A. K. Tripathi and S. Mukhopadhyay, 'Removal of fog from images: A review", IETE Technical Review, Vol. 29, No. 2, pp. 148-156, 2012.

6. A. K. Tripathi and S. Mukhopadhyay, 'Single Image Fog Removal Using Bilateral Filter", International Conference on Signal processing, computing and control, Shimla, India, March 2012.

.