A physical model is used to describe one or more fieldsize phenomena and, in representing them, it generates other quantities, which are the final output of the system. Sometimes the required thorough measurement of such quantities can lead to lose the total perception of the phenomenon under investigation. Digital image processing can not and should not obviously replace the measurements, but it can be a useful tool for capturing the global behaviour of the model-system and consequently of reality. This process can be especially effective in cases where, for certain configurations or for simplifying the model to its essential characteristics, the aim is to seek a simple representation, although not completely corresponding to reality.
Keywords-component: Image acquisition and analysis I. HYDRAULIC MODELSIn hydraulic models that are a scaled representation of hydraulic structure prototypes, gravity effects are usually predominant over surface tension effects and therefore Froude similitude is adopted.That is, the Froude number is the same for the model and the prototype, i.e.: p r m r F F = ⇒ p p 2 p m m 2 m L g V L g V =where the subscripts m and p refer to model and prototype parameters respectively. Based on the consideration that both the processes take place in the gravitational field, i.e. g m = g p . Equation (1) leads to the well-known velocity scale ratio V r , discharge scale ratio Q r , time scale ratio t r and roughness coefficient scale ratio k r (expressing the roughness parameter χ r using thewhere L r is the scale ratio. These relationships are the means which allow transferring the results measured in the model to the prototype scale for the phenomenon under investigation.The main difficult in constructing a physical model able to reproduce the full 3D prototype is to correctly scale the internal surface roughness. Equation (3) makes it possible to define the model roughness, once a roughness of the prototype and a scale parameter are assumed, Fig.1 . Figure 1Models are built according to design drawings and any subsequent updates, using commonly available materials. Limitations exist on the choice of the scale ratios, depending on the material type used for the model, as for small scale ratios (large models) it isn't possible to have fairly smooth surfaces for an adequate similitude, even with materials commonly used in laboratory, as PVC and PERSPEX.[6] [8] Each material has advantages that recommend its use for a particular purpose and disadvantages that prevent its universal use; in any case, all of them can be made smooth and waterresistant by appropriate treatments.