Correction method for geometric image distortion with application to printed circuit board inspection systems

Geometric distortion of an image is one of the most important parameters affecting the accuracy of optical inspection systems. We propose a new image distortion correction method to increase the accuracy of printed circuit board (PCB) inspection systems. Our model-free method is applied to correct the randomly distorted images that cannot be represented by a mathematical model. To reduce the correction time of the inspection system, we propose a control area-merging algorithm that minimizes the number of control areas using the quad-tree approach. We apply the proposed method to a PCB inspection system, and verify its usefulness through experiments using actual inspection images.