Image de-hazing by solving transmission value

摘要

An image processing server performs haze-removal from images. Global atmospheric light is estimated and an initial transmission value is estimated. In one embodiment, a solver is applied to an objective function to recover a scene radiance value based on the estimated atmospheric light and estimated transmission value. The scene radiance value is used to construct an image without haze. In a simplified method that avoids using a solver, bilateral filtering is performed on the transmission image in order to construct an image without haze.

主权项

1. A computer-implemented method of removing haze from an image, the method performed by a computer and comprising:
estimating global atmospheric light from the image, the estimating comprising:
performing k-means clustering on colors of the image to obtain a plurality of mean values; andselecting a largest value from the plurality of mean values as an estimate of global atmospheric light; estimating an initial transmission value t from the image; estimating a change in haze across neighboring areas of the image based, at least in part on the estimated global atmospheric light from the image and the estimated initial transmission value, the change in haze represented by a measure of similarity between the neighboring areas and a change in appearance between the neighboring areas in the image; and constructing the image without haze using the estimated change in haze.