Shell increments contain information related to the evolution of the environment in which the organism grew during its biomineralization. To extract the information from variations in shell topography, a new and promising technique is presented, involving multi-scale analysis of the shell topography using a B-spline wavelet transform. An accurate non-contact optical system, based on laser triangulation, is used to map the shell surface. The resulting range image is treated as a grey-level image by using a multi-resolution approach based on the generalization of the cascade algorithm. This method allows reconstruction of non-subsampled images that correspond to the projection onto the space of the chosen scale of detail. This new approach provides an efficient tool for analyzing multi-scale information contained in growth increment rings and/or within quasi-periodic features. In conclusion, this approach can be applied to any 3D object, in order to extract features such as rhythmic information, color variations or object envelope.