An image processing apparatus includes: an image input section via which a medical image picked up of a mucous surface of a living body is inputted; a setting section that sets a region of interest as a first region and a second region including a unit region in the medical image; a feature value calculating section that calculates a first feature value relating to a shape of the region of interest based on pixel values of the region of interest and a region in a periphery thereof and a second feature value relating to a shape of the unit region based on pixel values of the unit region and a region in a periphery thereof; and an matching section that matches the region of interest with a predetermined shape based on distributions of the first feature value and the second feature value calculated by the feature value calculating section.