　　With the increasingly widespread application fields of modern glass such as panel glass, automotive glass, architectural glass, optical glass and other fields, glass surface quality problems directly affect the quality of the product and subsequent user use. Modern manufacturing industry has increasingly stringent product requirements, and more and more More and more companies are beginning to pay attention to surface defect detection and control during the glass online production process. Due to the influence of production process, environment and other factors, the surface of glass is prone to crystal points, scratches, stains, impurities, black spots, clots and other defects during the production process, which seriously affects the appearance quality of the glass surface and brings problems to enterprises. Serious economic losses. Traditional companies use artificial eyes to detect glass surface defects. However, due to the slow detection speed, low accuracy and fatigue of human eyes, they cannot meet modern production requirements. The SIMV glass surface defect online detection system can completely replace the artificial naked eye for high-speed, accurate 7*24 hours real-time online detection of glass production, automatically alarm and display specific information of current defects (picture display, diameter, coordinate distribution, defect type, etc.) , enabling workers to deal with defects in a timely manner. At the same time, the system automatically prints and automatically saves the test report of each roll of products based on the product batch number, which facilitates users’ later inquiries, calls and printing, thereby achieving product quality rating and classification, evidence-based traceability of customer complaints, and effective savings. Reduce labor costs and improve product quality. Therefore, the glass surface defect online detection system has become an indispensable part of the automated film production line.

【Detection Principle】

　　For glass materials with high light transmittance, the system uses the transmission lighting detection method for detection, that is, the light source is below the glass, and the camera performs real-time detection above the glass (for opaque materials, the reflection lighting method is used, that is, the light source On the same side as the camera to be detected). When the production line is running, the system collects the production line operating status information in real time through the encoder and starts detection. The system uses the images collected by the camera to process defects through SIMV image analysis software. Because the defects and normal product images are obviously different on the gray scale. Differences allow the system to detect defects and determine the size, location, type and other information of the defects through further calculation and analysis.

Detection Object	Panel Glass, Automotive Glass, Architectural Glass, Optical Glass
Typical Defects	Defects such as crystal points, Scratches, Stains, Impurities, Black spots, Clots, Etc.
Detection Speed	Max:800m/min
Detection Width	Adapt to any width (customized according to the production line)
Detection Accuracy	0.05mm~1mm (depending on the number of cameras, production line width and vehicle speed)
Defect Identification	Automatic sound and light alarm, interception of current defect pictures and real-time X, Y coordinate defect distribution map display
History Record	Automatically record the picture, size, distribution location and other information of each roll of glass surface defects based on the batch number

【System Functions】

　　(1) Accurate defect map (X, Y coordinates) positioning display module;

　　(2) Automatic sound and light alarm, and display the current and previous 10 defect images;

　　(3) Real-time defect information display, recording the current defect type, area (diameter), horizontal and vertical position and other information;

　　(4) Basic product information module, which records product batch number, model, specification, shift number, operator and other information (the user can modify it according to the production situation);

　　(5) Real-time meter detection and alarm information display;

　　(6) Automatic classification of defects based on characteristics such as defect area size, shape, grayscale difference;

　　(7) Automatic alarm for periodic and continuous defects;

　　(8) Automatically generate product defect records for each roll of products, and users can directly query according to the batch number;

　　(9) Automatically change rolls and automatically print a quality report of product information for each roll;

　　(10) Functions such as automatic statistics, saving, querying, calling and printing of detection history records;

　　(11) Long-term, large-capacity historical record storage;

　　(12) Automated linkage of production lines, remote technical services, etc.;

　　Note: The software functions can be customized and developed with special functions according to the different requirements of users.

【Field Application】