A motion vector detecting unit detects a motion vector of each block in a shot image, which is divided into a plurality of blocks, during an optical zooming operation. A linear function calculating unit calculates for each block a linear function that represents a straight line, the slope of which is the same as that of a line passing through the center of the shot image and a representative point and which passes through a vector detection point. A hand-movement motion vector calculating unit selects at least two blocks that have the calculated linear functions, the slope of which are different from each other, and obtains an intersection of graphs of the linear functions of the selected blocks so as to calculate a motion vector caused by hand movement. A hand-movement correcting unit corrects the hand movement in the shot image based on the calculated motion vector caused by the hand movement.