基于IMU的3D扫描仪倾角标定方法北大核心CSCD

3D扫描仪以精度高、速度快、自动处理、稳定性好等优势在场景三维重建、智能制造、自动驾驶、虚拟现实等领域得到了广泛的应用。目前3D扫描仪设备多是安装在三脚架上,容易因为地面不平使得扫描仪的转轴和重力方向产生夹角,最终导致拍摄出的全景图和点云倾斜,成为影响三维重建精度的关键难题。针对该问题,提出一种借助惯性测量单元标定3D扫描仪倾角的方法。该方法利用惯性测量单元输出的加速度信息,通过一个面阵相机建立惯性测量单元和3D扫描仪的联系,将惯性测量单元的加速度信息转到扫描仪的转轴上。利用该信息计算出扫描仪转轴与重力方向的夹角,并将该倾角补偿到生成的全景图和点云中,校正倾斜模型。通过实验对校正前后计算的角度进行对比,实验结果证明倾角测量误差可以控制在0.5°以内。并且经过标定后,精度至少提升20%,验证了整个标定方法的可靠性和准确性。

Calibration method of 3D scanner tilt angle based on IMU

The 3D scanner has been widely used in 3D scene reconstruction, intelligent manufacturing, automatic driving, virtual reality and other fields due to its advantages of high accuracy, fast speed, automatic processing and good stability. At present, most 3D scanners are installed on tripods, which is easy to cause an angle between the rotation axis of the scanner and gravity direction due to uneven ground. Therefore, the panorama and point cloud are skewed, which is a key problem affecting the accuracy of 3D reconstruction. To solve this problem, a method to calibrate the tilt angle of 3D scanner by using inertial measurement unit (IMU) was proposed. The proposed method used the acceleration information output by IMU to establish the connection between IMU and 3D scanner through an area array camera. The acceleration information of IMU was transferred to the central axis of scanner. Using the transferred information to calculate the angle between the rotation axis of scanner and gravity direction, then compensating this tilt angle to the generated panorama and point cloud, and the skew model can be corrected. Through experiments to compare the calculated angles before and after the calibration, the experimental results show that the tilt angle measurement error can be controlled within 0.5°. And after calibration, the accuracy is increased by at least 20%, which verifies the reliability and accuracy of the entire calibration method.