Enhancing dynamics measurement from moving cameras through sensor-fusion motion compensation approaches

The use of cameras has gained popularity in the engineering world due to their ease of use and non-contact nature. The combined use of cameras and unmanned aerial vehicles (UAVs) allows performing complex acquisition in hard-to-reach locations. However, due to the motion of the UAV, measurements can be inaccurate. This study focuses on the mitigation of UAV-induced motion, to enhance the measurement precision for structural dynamic assessment by proposing a combination of sensor-based and algorithm-based camera motion compensation approaches. The sensor-based approach relies on the use of a novel system integrating an Inertial Measurement Unit and two laser distance sensors to account for the low-frequency components of the motion. An Extended Kalman Filter algorithm is then implemented to improve the accuracy of five of the six degrees of freedom of motion. Laboratory experiments were performed to compare the displacement measured with the moving camera post-processed using the proposed method against a reference stationary camera. The results of the experiments showed that the proposed motion-correction method provides displacements that are in good agreement with the stationary camera and show a significant reduction of the induced motion. Further developed, this technique can be used in various applications where motion-corrected data must be obtained for accurate assessment of the dynamic properties of the targeted system.