|
In modern industrial production, the visual inspection is often not the missing link. For example, the appearance of the car parts, pharmaceutical packaging, of the IC characters printed circuit board welding quality is good or bad, detection workers need to be observed by the naked eye or a combination of microscope inspection. Manual inspection of a large number of not only affects the efficiency of the plant, but also brought unreliable factors directly affect the product quality and cost. In addition, many of the detection step is not only require the detection of the appearance, and also need to accurately obtain the detection data, such as the width of the part, the hole diameter, as well as the coordinates of the reference point, and so on, the work is difficult to rely on the rapid completion of the human eye. With the development of image processing and pattern recognition techniques, a new detection method - machine vision. Machine vision system is by machine vision products (i.e. the image pickup device, stars in both CMOS and CCD) image capture, and then after the analog conversion of the number of the image is sent to the processing unit, according to the distribution of the image pixels and brightness, color, etc. information, extraction of characteristic elements, and finally determine thereby control the work of the field devices based on these characteristic elements. Embedded systems is the current rapid development of the popular technology, the solution for specific applications, and its small size, low price and simple development environment, the use of flexible, reliable operation, etc., has been widely used in various fields. China's current multiple large flat glass float glass production line in the production process will inevitably produce some defects, such as bubbles, impurities, soon due to the glass pulling speed, real-time response requirements are relatively high, manual inspection is completely unable to competent, mature solution is the use of machine vision to detect. Domestic the glass defect detection system are introduced from abroad, expensive, and difficult to maintain. There is only some design defect detection system based on PC technically is not very mature, yet reports of large-scale applications. In this context, combined with the advantages of embedded systems, this paper designed a set of embedded glass defect detection hardware system. Part of the entire hardware system including CCD imaging systems, image acquisition, image processing, and human-computer interaction systems, and do a more in-depth study on the system reliability to ensure stable operation in the harsh environment of industrial production,.
|