The Research On Automatic Coating Vision Positioning And Detection System Of High Power LED Module

At present,LED industry is in a high-speed growth period relying on the high efficiency and energy saving of LED devices.Due to the technology blockade of foreign countries,the upstream technology of LED industry in China is still controlled by most multinational enterprises.Although the number of patent applications in the middle and lower reaches has a large advantage,there is still no obvious technical advantage.LED packaging industry is very important to determine the performance of LED products and the quality of light.At present,the more mature coating process is still not fully automated,positioning is not accurate enough,and the nozzle of the spray gun is occasionally blocked and can not be detected in time,which can not better control the quality of the product.Therefore,for LED coating system,it is necessary to improve the automation and intelligence of the system.By introducing the vision system into the automatic coating system,the coating module feeding process can be positioned more quickly and accurately.At the same time,the visual system can also complete the preliminary detection of the nozzle blockage of the spray gun and the distribution uniformity of the phosphor after coating,so as to improve the accuracy and intelligence of the LED coating system.Therefore,the main work of this paper is as follows:(1)Analyze the LED coating process and determine the overall function of the system based on the project requirements,select and test the hardware and software required by the vision system.(2)Develop a vision system for the selected camera,and realize some visual functions,such as image saving,reading,cutting,image display,acquisition and transmission of different regions of interest,coordinate display,etc.(3)The selected camera is calibrated and the image distortion generated by the lens is corrected.At the same time,through the black-and-white chessboard combined with corner detection algorithm,the number of pixels per unit length in the world coordinate system under the working height can be obtained quickly and accurately,which is convenient for the subsequent positioning and detection tasks..(4)This paper proposes a method to locate the module in the feeding process by using photoelectric switch,small cylinder,camera and Hough transform in image processing,so as to realize the accurate matching between the fixture and the module to be coated.In the camera positioning scheme,the performance is compared with that of the template matching scheme to get a better positioning effect.(5)A machine vision algorithm is designed to carry out preliminary detection on COB module after coating to ensure that nozzle is not blocked and affect the quality of subsequent coated products.When the nozzle is not blocked,the least square method is used to fit the data to predict whether the nozzle of the spray gun is likely to be blocked.The uniformity of coating was evaluated according to the characteristics of phosphor distribution in COB module after coating.