An Improved Air-Light Estimation Scheme for Single Haze Images Using Color Constancy Prior

Hazy environment attenuates the scene radiance andcauses difficulty in distinguishing the color and texture of the scene.A crucial step in dehazing is the recovery of the global air-lightvector. Traditional methods usually interpret the RGB value ofthe brightest region in haze images as the air-light. In this letter,a new prior called ‘color constancy prior’ has been proposed toimprove the robustness of air-light estimation when varicoloredillumination exists.The prior utilizes the statistical observation thatdistant scenery objects become the most haze-opaque due to thepixel escalation towards the higher intensity side. The comparativeevaluation on a variety of haze images manifests that the proposedprior perform better than existing air-light recovery methods andcan be used for subsequent dehazing applications.