A stereoscopic technique for seafloor statistical characterization by acoustic remote sensing

The authors present a new method for performing statistical characterization of underwater surfaces by acoustic remote sensing. A wideband technique is outlined that, by exploiting stereoscopic observation, estimates the roughness height of underwater surfaces however oriented with respect to the instrument. The surface is ensonified by a wideband signal and simultaneously viewed from two or more slightly different directions of view. The corresponding echoes are compared in the frequency domain, and a roughness estimate is obtained. Once the instrument centerband frequency has been fixed, the scale factor in the estimation can be adjusted by varying the differences between the view directions. This permits a wide estimation interval, ranging from a small fraction to many times the wavelength corresponding to the centerband frequency. Preliminary experiments have been carried out in a water tank, and the estimated roughness values are in agreement with the theory.<>