Analysis Method Research And Software System Development For The Flat Glass Defect Detection

With the development of economy,people’s demand for glass is more and more,and the quality requirements are higher and higher.However,at present,most domestic manufacturers still use manual inspection methods,which leads to the problems of low production efficiency,high cost and poor quality control.In this paper,the image processing technology is used to detect plane glass defects automaticallyIn this paper,flat glass as the object to be detected.The surface defects in this paper mainly include scratch defects,bubble defects and chip defects,extracting the characteristic parameters of various defects,used as an important index to assess whether or not qualified for flat glass,to realize the automatic,real-time,accurate detection and identification of defect types.In this paper,an image acquisition experimental device consisting of a linear CCD camera,a lighting system,an IEEE 1394 bus,and a 1394 image acquisition card is used to acquire the image of the glass defect.Defect images are processed by pre-processing techniques such as filtering,denoising and binarization to obtain defect images for subsequent detection.In the scratch defects,this paper mainly studied the X-type scratch,U-type scratch and single scratch,and proposed the corresponding recognition algorithm.In the process of X-shaped flaw detection,the main algorithms such as the skeleton extraction,the change of the existing mark value,the maximum width and the length are proposed,which can basically realize the identification and analysis of the defects.In the process of U-shaped flaw detection,the algorithm of coverage and pruning is added,and the method of calculating the maximum width and length is changed,so as to detect the flaw and complete the integration and recognition of the three types of flaws.In the process of bubble defect detection,a new threshold segmentation method,hole filling and maximum size calculation are proposed based on the gray image features.The defect features such as the circularity,the elongation and the centroid are extracted,to complete the identification and analysis of bubble defects.In the process of chip detection,the main algorithms such as contour extraction and minimum external rectangle extraction of contour are proposed to identify and analyze this flaw..Based on the difference between different defects,the types of scratches,air bubbles and explosive boundary defects can be identified to achieve the simultaneous detection of these types of defects.The quality of the glass is evaluated according to the parameters of the national standard of glass defects.In the C++ software,we use the technology of object-oriented and modular programming to write the system login interface,complete the input interface of the national standard or the factory standard defect parameters,design the image processing menu and the test result output interface,so that the user can use it more conveniently,the basic realization of the flat glass automatic detection requirements.