Binocular vision vibration measurement based on pixel coordinate matching of inner corner points in a chequerboard

Abstract

A binocular vision measurement system provides a simple method for obtaining three-dimensional vibration data from moving objects, which is suitable for vibration monitoring of large structures such as bridges. Aiming to address the problem that the feature selection process for binocular visual inspection affects the measurement accuracy, chequerboard feature points are selected in this paper for carrying out a visual displacement measurement method. Firstly, pixel coordinate matching of the inner corner points in the chequerboard is completed and then a binocular vision measurement system is established. The measurement results are compared with using circular feature points. Secondly, the binocular vision measurement model is applied to the vibration measurement of a cantilever beam. Using comparisons with a three-axis acceleration sensor, the effectiveness and accuracy of this method are evaluated. Finally, the method is applied to measure the vibration of the cantilever beam under different load conditions and its vibration characteristics are analysed. The results show that the accuracy of the binocular vision measurement method based on pixel coordinate matching of the inner corner points in the chequerboard is higher than that using circular feature points. From comparisons with the acceleration sensor, the measurement error of this method is found to be small. In addition, the method can effectively analyse the vibration performance of a cantilever beam under different load conditions. Therefore, this measurement method is effective and provides a theoretical basis for the identification of vibration characteristics in large engineering structures.