Texture analysis: Comparison of autocorrelation-based with cumulant-based approaches

In this paper the use of 3^rd-order cumulants, i.e. triple correlations, is proposed for texture analysis. Properties of such features are derived, with particular attention to insensitivity to symmetrically distributed noises and statistical estimate stabilility. Experimental evaluation of 3^rd-order cumulants as descriptive features for textures is carried out in comparison with autocorrelation-based approaches.