INPUT COORDINATE DECIDING DEVICE

摘要

PURPOSE: To stably detect an error between an input image and input coordinates by adding lightness information to a structure model generated from the input image and its coordinates and making a comparison with reference data. CONSTITUTION: A structure model matching part 5 deforms the structure model in a reference structure model part 4 by using position information obtained by a coordinate input part 1 to be matched. Further, a lightness calculation part 6 calculates the lightness of each patch of the obtained structure model. The deformed structure model correctly matches the image in an image storage part 2 on condition that inputted position information is correct, so it is considered that the lightness distribution of respective patches is similar as long as the inputted image is of the same kind with the image stored in a reference data part 7. For the purpose, the lightness distribution of the reference image of the same kind as the data of the reference data part 7 and a decision part 8 calculates a difference from the lightness obtained by the lightness calculation part 6 and detects the difference between the position information and input image from the calculation result.