Imaging method, learning method, learning data generating device, and data structure

A learning data generating device comprising:

• an imaging device including an imaging element having a light-receiving area, an optical system configured to form an image from incident first light in the light-receiving area, and an optical element configured to guide second light to the light-receiving area, an angle formed between an optical axis of the optical system and a principle ray incident to the optical system being different for the first light and the second light;
• a display located within a field of view of the imaging device;
• and
• an information processing device configured to cause the imaging device to capture a third image, the third image being an image obtained when a first image and a second image are simultaneously displayed on the display, the first image formed by an object point radiating the first light, the second image formed by an object point radiating the second light.

A data structure used to train an image processing model configured to separate a second superimposed image into a first image component and a second image component, the first image component and the second image component being superimposed in the second superimposed image, the first image component corresponding to an object point radiating first light, and the second image component corresponding to an object point radiating second light, the data structure comprising:

• a first image formed by an object point radiating the first light;
• a second image formed by an object point radiating the second light;
• a first superimposed image created by superimposing the first image and the second image;
• and
• correspondence information representing a correspondence between the first image, the second image, and the first superimposed image, and
• the image processing model being trained using a combination of the first image, the second image, the first superimposed image, and the correspondence information as training data.