Studies On Industrial Vision Inspection Method

Although image-based inspection has been applied to a wide range of industrial applications, inspection accuracy remains a challenging issue due to the complexity involved in industrial inspection. The common method adopted in industry is to use a template image as a good image template to inspect each live image on a pixel-by-pixel basis. In this thesis, a tolerance method is presented to replace the template image method. The said tolerance is formed by two indices computed from an image, instead of using the whole image for inspection. To ensure an accurate tolerance zone, a Neural Network method is used to take into consideration the noise and uncertainties in the parts under inspection. To improve speed, the Orthogonal Arrays are adopted to select a minimum number of the sample images needed for training. Once a tolerance zone is obtained, a live image is inspected against it. If the indices fall inside the tolerance zone, it is regarded as good, otherwise faulty. The inspection accuracy achieved was 95%. Three examples are given, one for label inspection and the other two for auto part inspection.