| dc.description.abstract | In this paper we solve the problem of reconstructing a color image from sparse, noisy, and nonuniformly distributed color measurements. We apply a method for reconstructing a surface from sparse depth measurements to each of the R, G and B components of the color data, by treating each as a surface, with depth measurements being the R, G and B values. We apply this method to the reconstruction of nonuniformly distributed sparse color data from even 12.5% of the pixels, if no discontinuities are given and from 6.25% of the pixels, if the discontinuities are given. Also we present results of reconstructing a corrupted version of the original image with Gaussian noise of zero-mean and standard deviation 30 from 25% of the data, for color levels between 0 and 255. The applicability of the method is independent of the choice of the color space used. | en_US |
