Research On Visual Inspection And Positioning Technology Of Alien Electronic Component Plug-in Machine

With the continuous development of society,the demand for electronic products in daily life is getting higher and higher.At present,standardized electronic components can be mounted by surface mount equipment,and the mounting efficiency and precision are getting higher and higher.However,there are still special-shaped electronic components such as shaped LEDs,relays,and plug-in switches.Due to their complicated structure and different shapes,they are basically manually inserted.Therefore,it is meaningful to study the automatic insertion technology of the special-shaped components.The plug-in machine uses the automatic plug-in to replace the traditional manual plug-in,which greatly improves the efficiency of the plug-in and reduces the loss caused by the manual error.Therefore,at home and abroad,plug-in machines have a good development prospects.This paper mainly studies the visual inspection and positioning technology of the plug-in machine,and meets the requirements of the plug-in through actual verification.According to the functions and indicators of the plug-in machine,the structure of the plug-in vision system is designed.The plug-in machine studied in this paper has two sets of vision systems.The lower vision fixed on the machine table is used to detect the electronic component pins,and the upper vision fixed on the second arm of the robot is used to position the PCB board.And according to the characteristics of the target,the selection of the visual hardware and the design of the lighting mode are completed.In-depth study of the calibration of the plug-in vision system,through the study of the eye calibration,respectively,the eye to hand and the eye in hand are calibrated.Using the nine-point calibration,the pixel coordinates of the image and the corresponding robot coordinates are acquired according to the position of the nine-square grid,and the calibration parameters of the two sets of cameras are obtained by calculation.In this paper,the image preprocessing algorithms such as image enhancement,filtering,threshold segmentation and edge detection are studied.The image processing method is selected by comparing the experimental results to analyze the advantages of the selected algorithm.A linear transformation of two gray values is also proposed.Method to enhance the contrast between the target and the background.When positioning the Mark point of PCB board,an improved Hough transform method is proposed to extract the pixel coordinates of the center point of the Mark point.In the VS2010 development environment,the software platform for implementing visual inspection and positioning system is developed,and the development of visual software is completed.The design of the human-computer interaction interface of the visual system is realized by MFC,real-time communication between the visual system and the robot controller is realized.