In one form, the method comprises obtaining scanned values of the image, scaling the scanned values to obtain scaled values, applying a non-linear correction function to the scaled values to obtain a first approximation of corrected reflectance values, filtering the corrected reflectance values by applying a piecewise Gaussian filter, applying the non-linear correction function to obtain a successive approximation of the corrected reflectance values, determining whether the successive approximation of the corrected reflectance values are sufficiently converged, selectively repeating the filtering and applying if the corrected reflectance values are insufficiently converged and selecting the corrected reflectance values for subsequent processing if the corrected reflectance values are sufficiently converged.