Design Of Quality Inspection System For Printed Bills

Computer vision is widely used in industrial inspection.This system takes printed bills as the research object,and designs a quality line detection system which can be widely applied to various printing machines.The system has the characteristics of advanced,portable and real-time.The system has done the following work from two aspects: hardware selection and software design.1.A small printing runner is taken as an object,and the hardware design of the detection system is carried out according to the actual industrial situation.Firstly,the calculation method of camera parameters is studied,and the camera type is selected according to the parameters.After the selection of the camera,the lens is selected according to the actual parameters of the camera and the selection principle.The illumination scheme of the light source for the defect detection of printed matter is introduced and the appropriate light source and matching electronic circuit are selected;in order to facilitate the adjustment of the system.The parameters of the system and the portability of the system must be equipped with a speed measuring device.The core of the device is the rotary encoder.The selection principle of the rotary encoder is given.In order to control the shooting time of the camera,the trigger device is designed.The sensor is the core of the device,and the sensor is selected according to the field conditions.Finally,the system adopts Basler ra L4096-80 km CMOS linear array camera takes pictures of the printed matter,the Schneider XENON-Emerald 2.8/28-L lens,the FU-A05 D type reflective photoelectric sensor produced by Keyence Company,and the E6C3-CWZ3 EH 2500P/R 1M incremental encoder of OMPRON Corporation.2.The image processing flow of defect detection is studied in the software part,including image preprocessing,image registration,defect detection,morphological processing and Blob analysis and so on.Based on the classification and shape analysis of printed defects,the process of image processing is given.Algorithm selection is adopted to determine the algorithm type by comparing common algorithms.After detecting defects,the location and size of defects are determined by morphological principle,and the defects are marked and output results of defects are finally detected.The software platform of this system is based on the Windows operating system,using the VC++ language to package the modules of the acquisition card driver and the image detection,and the system software is completed.The image processing method adopts the OpenCV visual library.Through the installation,debugging and testing of the system,the results show that the system can meet the needs of practical application.