Image Dehazing using PCA Fusion Technique for Enhanced Road Visibility
Author(s) -
U. Pranali,
Samarth Borkar
Publication year - 2018
Publication title -
international journal of computer applications
Language(s) - English
Resource type - Journals
ISSN - 0975-8887
DOI - 10.5120/ijca2018917204
Subject(s) - computer science , visibility , computer vision , artificial intelligence , fusion , image (mathematics) , image fusion , computer graphics (images) , optics , philosophy , linguistics , physics
Presence of fog and haze significantly reduces the visibility of a scene. Better visibility is crucial for all computer vision applications thus recovering images impaired by haze or dehazing finds its application in the fields of surveillance, tracking, detection and restoration. In this paper fusion based approach using principle component analysis (PCA) technique has been adopted. The novelty of this algorithm is that it does not require any haze depth generation as normally required in many existing methods. Using the original image two images are derived on these images contrast adjustment, and contrast normalization techniques are performed. PCA fusion improves fused image quality and resolution. This method only requires the original image and is simple and easy to implement. As the haze impaired image appears whitish and blurry the details of the road become less evident thus making driving in foggy weather conditions unsafe. Thus, the proposed method concentrates on dehazing for better road visibility. The qualitative and quantitative comparison as compared with existing color fidelity and contrast reveals that our proposed novel method is better at restoring color fidelity and enhancing contrast. General Terms Haze, Air light, Contrast Color fidelity.
Accelerating Research
Robert Robinson Avenue,
Oxford Science Park, Oxford
OX4 4GP, United Kingdom
Address
John Eccles HouseRobert Robinson Avenue,
Oxford Science Park, Oxford
OX4 4GP, United Kingdom