Image processing based quality control of the impermeable seams in multilayered aseptic packages

摘要: We propose an approach to detect unsteady seams in multilayered aseptic packages.It is the first study on quality control of impermeable seams.Segmentation and edge detection are evaluated longitudinal transverse seams.One feature extracted from images classified by linear regression SVM.The proposed system can overlook packet only once a 96years. Multilayered material which guarantees extended shelf life liquid foods turned into package folding packaging machines. Some problems machines causes packages lose impermeability property therefore lead deterioration food. The leak test, was performed determine problem, has been carried out expert employees. process steps random selection food packages, opening it properly, observing distribution injected ink human eye. It known that this cause serious damage due human-based errors. With study, aimed perform test using combination image processing classification techniques. Through data obtained real production environment, experiments were results evaluated. As result, be said distinctive as being its field with promising results.