Color image detail enhancement based on quaternion guided filter

Color image enhancement is an active research field in image processing. Currently, many image enhancement methods are capable of enhancing the details of the color image. However, these methods only process the red, green and blue (RGB) color channels separately, which leads to color distortion easily. In order to overcome this problem, the paper presents an approach to integrate the quaternion theory into the traditional guided filter to obtain a quaternion guided filter (QGF). This method makes full use of the color information of an image to realize the holistic processing of RGB color channels. So as to preserve color information while enhancing details, this paper proposes a color image detail enhancement algorithm based on the QGF. Experimental results show that the proposed algorithm is effective in the applications of the color image detail enhancement, and enables image's edges to be more prominent and texture clearer while avoiding color distortion. Compared with the existing image enhancement methods, the proposed method achieves better enhancement performance in terms of the visual quality and the objective evaluating indicators.