A novel approach for color image enhancement based on NTF

Abstract

A modified technique for image enhancement based on a specifically designed nonlinear transfer function (NTF) is presented in this paper. The proposed algorithm improves the visual quality of digital images captured under non-uniform illuminated condition. This paper proposes a method for enhancing the color images based on nonlinear transfer function. The present study exploits Hue-Saturation-Value (HSV) color space, since the HSV space separates color from intensity. The image enhancement technique is applied to the value component of the image, and saturation component is enhanced by stretching its dynamic range to get rich color display. The hue component has been kept constant in order to avoid color distortion. In this work, the value component of the HSV has been modified in two steps. First, the non-linear transfer function is applied on the overlapping and nonoverlapping window in order to enhance luminance part of the image. Further, the 2-D Gaussian filter has been used for contrast enhancement. The modified value is combined with preserved hue and enhanced saturation components in order to obtain enhanced HSV image. Finally, the image from HSV color space is converted into RGB color space. The subjective and objective evaluation shows that this method yields better result than other conventional enhancement methods.