In this work, optical imaging is used to quantify the mixing dynamics of granular materials. Two different methods (GLCM and Multivariate RGB analysis) are combined to extract surface information and the time to achieve a specific degree of mixing. In particular, the effect of density, friction coefficient, surface quality and particle geometry was studied. The results show that the Froude number alone is not enough to completely characterize the rolling regime. In addition, the filling ratio must be in a specific range which depends on the material properties. Crown Copyright (C) 2011 Published by Elsevier B.V. All rights reserved.