An estimated value of sound velocity of a subject interior is derived more stably and with higher accuracy. A phasing processing section provides, to each of received signals that were generated by piezoelectric elements respectively, respective delay times that were computed on the basis of each of plural set sound velocities, and phases the received signals per set sound velocity. A degree of similarity deriving section derives, for each of the set sound velocities, a degree of mutual similarity among received signals that were phased by the phasing processing section. An optimal set sound velocity deriving section derives, as an optimal set sound velocity, an estimated value of sound velocity of the subject interior on the basis of the degree of mutual similarity among the received signals of each of the set sound velocities.
