An approach to utilize a no-reference image quality metric and fusion technique for the enhancement of color images

This thesis presents a color image enhancement method that uses an existing method (developed for grayscale images) to adjust the luminance component of a color image towards intensifying details without debasing the images\u27 perceptual quality. For color modifications, the proposed method uses Adaptive and Integrated Neighborhood-Dependent Approach for Nonlinear Enhancement (AINDANE), a superior method originally defined for improving color images captured under low-light. Furthermore, utilization of an appropriate white balancing technique gives rise to a color temperature controlled final image. It is shown that a no-reference quality metric scores higher for final enhanced images of selected diverse images and a comparison with AINDANE shows the competitiveness of the proposed method. Finally, a statistical factor based graphical representation displays a convergence towards the visually optimal range