An image forming method uses an optical coherence tomography as to an optical axis direction of plural pieces of image information of an object. First image information of an object is obtained at a first focus with respect to an optical axis direction to then object. A focusing position is changed by dynamic focusing from the first focus to a second focus along the optical axis. The second image information of the object is obtained at the second focus. A third image information, tomography image information of the object and including a tomography image of the first focus or the second focus, is obtained by Fourier domain optical coherence tomography. A tomography image or a three-dimensional image of the object is formed in positional relation, in the optical axis direction, between the first image information and the second image information using the third image information.