This article refers to algorithms based on finite difference schemes for computing mean and affine curvature evolutions of digital images, introduced by Alvarez and Morel [L. Alvarez, J.M. Morel, “Formalization and computational aspects of image analysis”, Acta Numerica, pp. 159, 1994]. We discuss consistency, stability and convergence. Our analysis focuses on some possible choices of the parameters, choices that generate multiple variants in the implementations. Meaningful visual examples on how the algorithms actually work are provided.