Online Inspection Of Defects In Polymeric Film Products

The cleanliness is one of most important parameters for polymeric film product. How to improve the cleanliness of polymeric film product is an important problem. To solve the problem, in the thesis, online inspecting system of defects in polymeric film and concerned techniques were studied. The digitized improvement and function promotion for existed inspecting system of defects were achieved, in which the real-time measuring technique and digital image processing technique were adopted.Signals of defects in polymeric film under scattered light illumination were obtained by online inspecting system with linear array CCD control circuit. The 1-D defect signals were collected and transferred using PCI data acquisition card, and the 1-D defect signals were analyzed and processed by computer. Thereby, the defects were precisely figure out, and its geometry can be displayed real-time.In the thesis, the principle of data acquisition hardware and the operational mechanism of Windows system were studied, and the programming method of driver for PCI data acquisition card written by DriverStudio was given, by which real-time and reliability in the progress of data acquisition and transmission were improved. On the basis of deeply analysing signal characteristics of defects, the algorithm for online inspecting of defects in polymeric film products was put forward: first signals from linear array CCD were preprocessed, and defects were inexactly located using dual threshold, then the curve fitting gradient method was used to accomplished subpixel edge location. Thereby, the online inspection for precise attribute and figure of defects was achieved. Depending on above algorithm, online inspecting system software of defects in polymeric film products was developed in VC++ programming language. Image processing and parameter computing of defects were achieved using the method of associated programming between VC++ and MATLAB. According to experimental results, defects within 50μm～1500μm can be detected effectively by online inspecting system. Error is less than 5%.Geometry of defects could be reconstructed accurately in addition.