Robust vision-based method for wing deflection angle measurement with defocus images

Abstract

The accuracy of existing vision-based angle measurement methods is extremely reduced when the images appear defocus. To solve this problem, a robust center location method for defocus circular targets is proposed and sub-pixel accuracy of control points can be obtained even with defocus images. Based on this location method, two vision-based angle measurement methods are proposed to measure deflection angles for fixed-axis and moving-axis wings, respectively. These methods are simple and easy to install with no need of prior information of wings’ structure or rotation axes. Simulation and real experiments show that the proposed methods have good robustness to image noise and can significantly reduce the measurement errors of defocus images. The proposed methods almost keep unchanging to the results at good focus status, and the measurement accuracy under severe defocus blur state is about 60% higher than that of the state-of-art method for comparison.