Research On Visual Positioning And Detection Technology Of Multi-pin Chip Mount

Chip is the carrier of integrated circuit,which can be turned into microelectronic components with practical value through surface mount technology,and has a very wide range of applications in civil and military fields.With the continuous improvement of chip mount performance index,the requirements of surface mount technology are also increasing,and the development of online positioning recognition and detection technology in the field of machine vision has also received extensive attention and applications.The research object of this paper is multi-pin chip,mainly research on the visual positioning technology of chip placement based on template matching,and realizes the accurate positioning and identification of target chip in the case of block and stack.Researching on pin defect and size detection algorithm of multi-pin chips to achieve accurate measurement of chip pin number,pitch and size.The main research contents are as follows:(1)Conduct a demand analysis for the visual positioning and inspection of multi-pin chips,and build an experimental platform for the visual positioning and inspection of chip mounting.On the basis of this experimental platform,the selection design of camera,lens and light source is carried out,and the algorithm design of chip placement positioning and detection is initially completed.(2)Researching calibration method for the extraction of the center point of the circle,and compare the theoretical and experimental analysis of the traditional checkerboard corner point calibration and the center of the center point calibration.Based on the established visual positioning and detection experimental platform,through the correction template made,the corner and circle center extraction experiments were carried out under the same environmental conditions.The experiment showed that the circle center extraction algorithm based on the circular calibration plate has strong anti-noise ability,the algorithm is simple and fast,and the calibration accuracy is higher,which meets the requirements of system positioning and detection accuracy.(3)In the visual positioning of multi-pin chips,the positioning algorithm based on template matching is studied.Three types of template matching methods based on gray characteristics,correlation characteristics and shape characteristics are compared and analyzed.Through positioning experiments under complex lighting conditions,different types of chips and stacking of obstacles,the shape-based template matching positioning effect is obtained optimal.This method uses the generalized Hough transform algorithm to reduce the sensitivity of the edge contour to light,and can accurately locate the qualified chip in complex situations such as occlusion or incompleteness of the target chip;create a template based on the shape and contour of the target chip to reduce the impact of noise,can effectively improve the operationefficiency.(4)In the visual inspection of multi-pin chips,it mainly completes the coordinate extraction of the center position of the chip,the angular offset,and the detection of pin defects and dimensions.For size measurement,Canny edge detection operator is used for Gaussian fitting to obtain the edge contour of the target chip,and the minimum external rectangle is created to obtain the posture information of the chip;for the detection principle of the number of chip pins,the edge to gray projection method is used to realize the chip accurate measurement of lead spacing and external dimensions,measurement error can reach 0.01 mm.