Bilateral Filtering Embed with Edge Flow Feature

Abstract

The bilateral filtering is a blurring method that aims to preserve the strong edges in the image. This paper purposes the scheme that collaborate between the edge flow feature and the Gaussian function applied for the bilateral filter. From the edge flow feature, it sets up the reference points from the image to support the consideration of the bilateral filter. The reference points of edge flow feature consists of the magnitude and the direction of the edges of the image. Both features are employed to discriminate the objects in the image. These reference points of the edge flow feature are cooperated in the bilateral filter to improve the quality of the filter. The results of our scheme will compare with the generic of the bilateral filter that our scheme can perform very well.