Gaussian filter threshold modulation for filtering flat and texture area of an image

Abstract

In digital images, there are some regions that correspond to flat area and some others correspond to texture area. Different image processing need to be performed depending upon the type of region. This paper presents threshold modulation of Gaussian filter for flat and texture area filtering of an image. Standard deviation is used to differentiate between flat and texture area. In texture area, the Gaussian adaptive filter behaves as a Normal distribution filter and in flat area, the Gaussian adaptive filter behaves as an average filter. Experimental results on various test images illustrate the capabilities of Gaussian filter threshold modulation approach in efficient noise reduction.