An optimization algorithm for the reconstruction of a surface profile is presented. This algorithm uses the far-field transmitted intensity to retrieve the surface profile of a thin film. We approach the inverse transmission problem as a constrained optimization problem. A mathematical representation of the surface based on B-spline curves and an optimum profile search with the self-adaptation evolutionary strategy (ES) are adopted. As the input data for the ES algorithm for surface inversion, the transmitted intensity has been measured by both a laser bidirectional reflectometer instrument and an in-line digital speckle camera system. The reconstructed 1D surfaces are compared with the profile measured by an atomic force microscope.