The patent discloses an automated system and algorithm for evaluation of imperfections in the metal pieces that form the body of a car, in order to reduce the standard deviation in the quality control of the experts.
The patent includes: The hardware implementation of the automated system, an optical system based on light reflection for defect enhance, an algorithm which evaluates the signals profiles are extracted from the image and a defect classification system based on fuzzy logic.
The method involves calculating initial, intermediate and following profiles from a segment defined on an image of a zone to be analyzed of a piece, where pixel of the profiles is equal to a variation in gray level in a corresponding pixel of the segment along a length and a direction of the segment. Each of initial and following measuring points is a pixel for which the variation of gray levels along the length and the direction of corresponding measurement profile is maximum, where the pixel is computed by a gray levels maximum variation determining algorithm. Independent claims are also included for the following: a computer program product comprising a set of instructions for implementing a method for characterizing a surface defect of a stamped piece an installation for characterizing a surface defect of a stamped piece.