Provided is an image capture method including: illuminating an object having known label data, capturing images of the object using illumination parameters, and acquiring captured images; generating an estimated image of the object at the time of illumination based on the illumination parameter on the basis of an image data set obtained by associating a captured image with an illumination parameter; setting a maximum number of captured images permitted for the object; and performing learning with a machine learning model for each number of images equal to or smaller than the maximum number of captured images, optimizing illumination parameters and inspection algorithm parameters on the basis of a result of comparison between a result of estimation of the machine learning model and the label data of the object, calculating an index representing inspection performance expected when the each number of images is captured, and presenting the index to a user.