High-precision Vision Measurement Technology For The Dimensions And Gaps Of PCBA Parts On Smartphones

PCBA(Printed Circuit Board Assembly)components of smart phones have small size and intensive assembly.The high-precision visual measurement technology of component size and gap is one of the key technologies for precision and flexible assembly in automatic production line.Based on the actual project requirements,this paper aims to improve the dimension of PCBA parts and its gap vision measurement accuracy of smart phones.The research is carried out from four aspects: high-precision calibration of monocular vision system,component edge detection and repair,sub-pixel contour extraction,dimension and gap measurement.The main research contents include:(1)Research on high-precision calibration method of vision measuring platform.The sub-pixel corner detection algorithm is used to calculate the corner coordinates of different viewing distances and different angles.Combined with the internal and external parameter matrix of the camera,the distortion coefficient is optimized by the method of re projection.Compensating parameters are set to modify the distortion coefficient in the region of element height change,and a dynamic compensating distortion correction model of imaging system is proposed.Verification experiments show that the calibration accuracy of the system is within ±0.01 mm.(2)Research on edge detection and repair algorithm of PCBA parts.The new Canny operator PCBA component edge detection algorithm is proposed,which uses the guidance filter with "dynamic" penalty term instead of Gaussian filter of the original Canny operator,and the non-maximum suppression strategy of adaptive neighborhood window is adopted to eliminate pseudo edge pixels,and the adaptive optimal threshold segmentation algorithm of neighborhood window is put forward instead of the original dual threshold segmentation.The EPGVF-Snake model is used to repair the fracture edge.The results of the comparison experiment demonstrate that the new edge detection and repair algorithm can effectively avoid the discontinuity,non-refinement and the defects of pseudo edge;(3)Research on subpixel contour extraction and dimension measurement of PCBA components.For the detected edge pixels of parts,the adaptive step threshold is used to replace the global step threshold in the original Zernike moment,and the sub-pixel position between adjacent edge pixels is solved.Thus,a method of sub-pixel contour extraction of PCBA parts based on improved Zernike moment is proposed.On this basis,combined with the calibration parameters of the system,the measurement dimensions of the subpixel contour of the circular and rectangular parts are calculated by using the minimum circumscribed circle and minimum circumscribed distance method respectively.The experimental results verify that the visual measurement accuracy of circular and rectangular parts is within ±0.03 mm.(4)Research on PCBA parts clearance measurement method.According to the parameters calibrated by the large line-of-sight system,the position matrix of PCBA parts clearance is established,and the gap positions are measured according to the parameters of small sight distance system,and the gap measurement of circular and circular parts,circular and rectangular parts,rectangle and rectangle parts is realized according to the research results of(2)and(3).The experimental results prove that the visual measurement accuracy of the above gaps is within ±0.03 mm.