In this paper we propose an original solution to combine the scales of multi-resolution shape descriptors. More precisely, a classifier fusion scheme is applied to a set of shape descriptors obtained from the ridgelets transform. The Ridgelets coefficients are grouped into different descriptors according to their resolution. Then a classifier is trained for each descriptor and a final classification is obtained using the classifier fusion scheme. We have applied this approach to symbol recognition using the GREC 2003 database. In this perspective, we increase the recognition rates of previous works on ridgelets-based descriptors.