The Development Of Adhesive Products Print Defect Detection And Control System

Nowadays, the users are expecting much higher quality of all sorts of products and this helps to improve the quality continuously. The label, as the identification of some kind of product, has also undergone much quality improvement. In order to raise the automation level of label quality detection and solve the problem of any limitation of human inspection, this thesis, based on machine vision technology, researches on the image processing algorithm and theory for adhesive print products and finally develops a detection and control system for adhesive products print defect.Based on the analysis of the defects and detection requirements of adhesive print products, this thesis firstly constructs a set of equipment integrated with tension control and image processing to automatically detect the print defection.After a study of the existing tension control mechanical structure, this thesis proposes to establish an optimized structure by placing the tension control between the volume roller and the rewinding roller. The optimized structure will use converter, encoder and analog control to build a position closed loop control between the volume motor and the rewinding motor and then the tension control algorithm can be realized in the detection process of label defections. According to the speed feedback of the coding line of this new system, the positioning of converter control motor can be finished and this further realizes the positioning of defected labels in rewinding.According to the features of adhesive label detection, we need to increase the instantaneity of the image processing algorithm. This can be realized by image mosaic, sheet label extraction and gray processing to reduce the data quantity of image processing. Besides, other algorithms such as image shifting, image rotation, image corrosion and swelling also contributes to the development of adhesive products print defect detection and control system.