An image processing method is provided. This method comprises: a first information extraction step of extracting first information, including a registration signal used to correct the geometrical distortion of an image; a determination step of employing results obtained at the first information extraction step to determine whether second information is to be extracted from the image; a division step of dividing the image into at least one block; and a selection step of selecting the block. In this method, the first information and the second information are embedded in the image as invisible or less visible electronic watermarks, and are added to components of said image that are less easily discerned by a human's eyes. Further, the amount of said first information is smaller than the amount of said second information, and the first information is embedded more deeply in the image than is the second information.