Boosting the stability of wavelet-based contrast enhancement of color images through gamma transformations

Contrast enhancement in color images can be problematic due to many causes as, e.g. generation of unnatural colors, inversion of the order between level lines of the three chromatic channels and so on. Since the human visual system automatically realizes an efficient local contrast enhancement, it is natural to take inspiration from its features to build models that improve contrast in digital color images. In recent years, it has been proven that it is possible to build a variational framework which serves as a unified home for color correction models which comply with the basic perceptual features of color perception. This framework was built in the spatial domain and then extended in the local frequency domain thanks to the use of wavelets. A noticeable characteristic of the wavelet perceptual color correction algorithm is the existence of a range of parameters for which its output images are artefact-free. In this paper we show how to boost the stability of this method through a suitable selection of a functional parameter.