Calibration of computer vision positioning system for MEMS wire bonder

In this paper, a calibration method to improve the accuracy of MEMS Wire Bonder is presented. With the requirements of high accuracy and high speed in wire bonding, vision system has become one of important factors which influence the precision of MEMS packaging. It can be used as a feedback of control system to fulfill the requirement of accurate locating. A positioning system using computer vision technology has been designed to obtain high locating accuracy during wire bonding. The calibration of this system is the assurance to get distorting parameters and the relationship of all coordinate systems. Usually, Zhang's camera calibration method is widely used to get high accurate camera calibration results. But this method was only applied in normal imaging system. In this paper, the method is improved to be applied in micro-imaging system used for MEMS Wire Bonder. Combining the camera pinhole model and the Gauss lens law, the micro-imaging camera model is established. Certain smaller rotation angles are adopted to get acceptable calibration results. Simulations and experiments are performed to validate the proposed method with satisfied results.