Research And Design Of Intelligent Beer Bottle Damage Detection System Based On Machine Vision

China is a big country of production and consumption of beer drinks.The rapid development of the beer industry urgently requires to refine and improve the existing beer production line,in order to meet the high standards of modern industrial production lines. The mainstream beer bottle fouling detection approach is artificial light inspection,particularly susceptible to subjective factors,so the accuracy and detection speed can not achieve a high level, and it is hard to ensure the quality of the filling index bottles. Eventually it will cause losses to consumers and businesses.There is an urgent need to improve the existing artificial light inspection defacement detection methods. We use high-precision detection of new technologies to automate and improve the detection system of defaced beer bottles.In this paper, we studied the bottles defacement detection system based on machine vision technology.By a CCD camera, a reasonable choice of light, bottle beer bottle image acquisition, positioning and other processing, to finish the task of high-speed detection of empty bottles..The whole system mainly consists of transfer function from the front of the module into the bottle detection module, mechanical and electrical control module, visual detection module and unqualified hit functional modules and other components..By studying beer bottle detection technology index, the mechanical structure of the experiment, electrical control, image acquisition devices and software designed to detect bottles edge defect detection and image pre-processing programs to achieve beer bottles on the assembly line for image acquisition and subsequent processing.Image pre-processing is the basis for image analysis and identifying defects. Use the method of median filter to smooth the images of all parts of beer bottle.Edge defect detection is a key step in image processing, mainly to complete edge location and defect detection work.After a variety of edge detection algorithms for analysis, including the definition of type, scope operator template comparison and detection features, and detection results do horizontal comparison, we use the Canny operator to complete the image edge defect detection.MATLAB is a powerful mathematical software involves of analysis mathematics, algorithm development and system simulation,It has a complete image processing functions, friendly platform and programming environment, with good visualization image processing, etc..It is easy to be learned and mastered.This paper, based on MATLAB software development platform, the development of edge detection GUI, using the least squares fitting circle, chain code tracking, improved hough transform like the bottle, the bottom position.The principle and method of defect detection. Finally, in the form of experimental test data table shows that, after the actual industrial production line testing, access to more satisfactory results.