Abstract:

We developed an integrated laser-soldering system which combined a computer vision. Based on the analysis of experimental data, it is found the len distortion and assembly angular errors of X-Y stage to play a distinct role in introducing positioning errors. For the sake of satisfying positioning accuracy of end-effector and precision specification of packaging technique, an algorithm based on straight line method was built to abate distortion of image, and a compensation algorithm based on computer vision for positioning calibration of the non-vertical X-Y positioning stage was also developed. Finally, experimental results and comparative data to evaluate the performance of these approaches were reported.

Key words: laser-soldering, image distortion, computer vision, calibration, compensation

摘要：

构建了一个基于机器视觉辅助定位的激光植球系统。通过试验发现，整个系统的定位误差主要由机器视觉中透镜畸变和X-Y定位平台的装配误差引起。对于透镜畸变，引入了基于直线方法的图像畸变校正算法进行校正；对于非垂直的X-Y定位平台，通过基于机器视觉的软件补偿算法来校正X-Y定位平台的装配误差。最后，给出了试验对比数据和试验结果，验证了整个系统的标定效果。

关键词: 激光植球, 图像畸变, 机器视觉, 标定, 补偿