A questionnaire survey carried out among 25 visually impaired women indicates that they feel uncertainty to their makeup. It is observed that there is a need to develop a system to teach them how to apply makeup and to check their makeup. We have been developing a prototype of a makeup support system for visually impaired persons. The system provides information on makeup and imparts knowledge on how to apply makeup. Further, for checking the current makeup condition, an image processing system that can recognize the face and its parts and check for excess lipstick and the shape of the eyebrows has been developed. From a series of input images, the best image for checking can be selected on the basis of the sum of squares of residual errors after an affine transformation. Further, a usability assessment was carried out by considering eight visually impaired women, and the result could achieve a high score related to the content of information on makeup.