THE ELEVENTH IEEE SENSOR ARRAY AND MULTICHANNEL SIGNAL PROCESSING WORKSHOP (IEEE SAM2020)

When the near-field underwater acoustic image measurement is carried out by the line array laid on the sea floor, the resolution of the conventional beamforming (CBF) acoustic image measurement method is poor, the sidelobe level is high, while the deconvolution algorithm has the effect of high resolution and low sidelobe. However, the direct deconvolution algorithm of the point spread function (PSF) shift-variant model has a large computational burden. This paper presents a non-uniform spatial resampling Richardson-Lucy (RL) fast algorithm, which based on energy distribution of conventional acoustic image measurement result make the original uniform space scanning to non-uniform spatial resampling. It can reduce the number of scanning grid, so as to reduce the amount of computation. Simulation results show that the fast RL algorithm can achieve the performance close to the original RL algorithm by reducing the computational amount by nearly an order of magnitude.

underwater passive acoustic image measurement; RL deconvolution; non-uniform spatial resampling; focused beamforming