A feature amount extraction unit 27 extracts a feature amount from each learning image generated under the same combined condition information, and a classifier generation unit 35 generates a classifier by learning with the extracted feature amount as training data. A storage unit 43 stores the generated classifier in association with the combined condition information of the learning image. A classifier selection unit 37, upon input of the combined condition information at the time of generation of an examination image, searches for the most closely related stored combined condition image and selects the classifier that is associated with the searched combined condition image. The classifier selection unit 37 then extracts the most appropriate image processing method for the examination image using the selected classifier. The appropriate image processing method for the examination image is extracted on the basis of the feature amount of the examination image received by a classifier C. In this case, the appropriate image process can be promptly applied to the examination image under any set of combined conditions. Additionally, the quality of the X-ray image after the image processing can be kept constant regardless of the operators skill.
