ACOUSTICAL RESONANT SCATTERING AND MATERIAL CHARACTERIZATION

Abstract Object Recognition is nowadays a question of major importance. To solve this question (sometimenamed the "Inverse Problem") it is necessary to determine both the shape and the material of the object. Since many years our team has been studying Resonant Scattering from axisymmetrical "targets". We have recently succeeded in solving the "Inverse Problem" for such scatterers, using two different techniques: Artificial Intelligence and a parametrical approach of resonance frequencies variations followed by an inversion of the results. After an overlook of such procedures, we focus this paper upon the material characterization of cylindrical objects. For an homogeneous, isotropic material, acoustic wave resonances arc sensitive to the shear and longitudinal sound velocities, and to the density of the material. The frequencies of surface acoustic wave resonances of the object are related to these velocities and to the characteristic dimensions of the scatterer while the resonance width is related to the allenualion of the resonating surface wave and, consequently. depends upon the density of the material. Our previous studies of the direct problem allow us to select the characteristic parameters having the strongest influence on a given resonance of the object. We show that the easiest parameter to characterize is the shear wave velocity, then the longitudinal one and, finally, the density of the material. Typical examples of material characterization are given. KEY WORDS: Ultrasonicmaterial characterizationresonancesinverse problems