作者：陈志祥, 须颖, 石锦洋

作者单位：广东工业大学机电工程学院，广东 广州 510006

关键词：工业CT;几何校正;校正模体;图像重建

摘要：

工业计算机断层成像技术（computer tomography，CT）系统几何参数不准确会影响重建图像的质量和尺寸测量的精度。为校正工业CT几何位姿失配参数，提出基于标准球板模体的几何校正方法。首先，利用球板在两个不同位置的投影图求解出射线源到探测器的距离以及射线源到运动平台的距离。再在360°范围内间隔一定角度均匀采集球板投影信息，拟合球心投影轨迹，解析求解探测器偏角η、$\varphi $和探测器中心点坐标。最后根据重建出的体素模型的球心距测量结果，利用牛顿迭代法，以工业CT尺寸测量最大允许误差为约束条件迭代求解探测器偏角θ。实验结果显示：该文几何校正方法可以求解出所有工业CT几何位姿失配参数。参数校正后，提升图像质量，消除重建图像伪影，且工业CT球心距测量最大误差由44.8 μm降低到16.3 μm，表明该几何校正方法可以满足实际工业CT系统校正的要求。

Geometric parameter calibration of industrial CT system
CHEN Zhixiang, XU Ying, SHI Jinyang
School of Mechanical and Electrical Engineering, Guangdong University of Technology, Guangzhou 510006, China
Abstract: Incorrect geometric parameters of the industrial CT (computer tomography) system will affect the quality of the reconstructed image and the accuracy of dimensional measurement. In order to calibrate geometric mismatching parameters of industrial CT, a geometric calibration method based on the standard spherical plate phantom was proposed. Firstly, the distance from the ray source to the detector and the distance from the ray source to the rotating platform were calculated by using the projection of the spherical plate at two different positions. Then, the projection information of the spherical plate was uniformly collected at a certain angle interval within 360° range, fitted the sphere center projection trajectory, and the detector deflection angle η, $\varphi $ and the detector center point coordinates were calculated analytically. Finally, based on the spherical-center distance measurement results of the voxel model, the Newton iteration method was used to solved the detector deflection angle θ by taking the Maximum Permissible Error of the industrial CT dimensional measurement as the constraint condition. The experimental results showed that the geometric correction method could solved all the geometric position mismatch parameters of industrial CT. After the parameters were calibrated, the image quality was improved, and the reconstructed image artifacts were eliminated. The maximum error of industrial CT spherical distance measurement was reduced from 44.8 μm to 16.3 μm, indicated that the geometric calibration method could met the requirements of actual industrial CT system calibration.
Keywords: industrial computer tomography;geometry calibration;calibration phantom;image reconstruction
2021, 47(12):23-28  收稿日期: 2020-11-04;收到修改稿日期: 2021-01-09
基金项目:
作者简介: 陈志祥（1996-），男，江西南昌市人，硕士研究生，专业方向为工业CT图像处理
参考文献
[1] 张朝宗. 工业CT技术和原理[M]. 北京: 科学出版社, 2009.
[2] FELDKAMP L A, DAVIS L C, KRESS J W. Practical cone-beam algorithm[J]. Journal of the Optical Society of America A, 1984, 1(6): 612-619
[3] NOO F, CLACKDOYLE R, MENNESSIER C, et al. Analytic method based on identification of ellipse parameters for scanner calibration in cone-beam tomography[J]. Physics in Medicine and Biology, 2000, 45(11): 3489-3508
[4] JACOBSON M W, KETCHA M D, CAPOSTAGNO S, et al. A line fiducial method for geometric calibration of cone-beam CT systems with diverse scan trajectories[J]. Physics in Medicine and Biology, 2018, 63(2): 025030
[5] ZHAO J T, HU X D, ZOU J, et al. Geometric parameters estimation and calibration in cone-beam micro-CT[J]. Sensors, 2015, 15(9): 22811-22825
[6] 周凌宏, 李翰威, 徐圆, 等. 锥束CT圆轨道扫描的几何校正[J]. 光学精密工程, 2014, 22(10): 2847-2854
[7] SMEKAL L V, KACHELRIE M, STEPINA E, et al. Geometric misalignment and calibration in cone-beam tomography[J]. Medical Physics, 2004, 31(12): 3242
[8] CHO Y, MOSELEY D J, SIEWERDSEN J H, et al. Accurate technique for complete geometric calibration of cone-beam computed tomography systems[J]. Medical Physics, 2005, 32(4): 968-983
[9] XIANG Q, WANG J, CAI Y F. A geometric calibration method for cone beam CT system[C]//Eighth International Conference on Digital Image Processing (ICDIP 2016). International Society for Optics and Photonics, 2016.
[10] PATEL V, CHITYALA R N, HOFFMANN K R, et al. Self-calibration of a cone-beam micro-CT system[J]. Medical Physics, 2008, 36(1): 48-58
[11] 张才鑫, 黄魁东, 陈华. 基于标定量具的快速锥束CT几何校准方法[J]. 仪器仪表学报, 2019, 40(1): 175-182
[12] ZHOU K, HUANG Y X, MENG X X, et al. A new method for cone-beam computed tomography geometric parameters estimation[J]. Journal of Computer Assisted Tomography, 2016, 40(4): 639-648
[13] LI G, LUO S H, YOU C Y, et al. A novel calibration method incorporating non-linear optimization and ball-bearing markers for cone-beam CT with a parameterized trajectory[J]. Medical Physics, 2019, 46(1): 152-164
[14] LI H C, HONG W, LIU Y, et al. A novel method of micro-tomography geometric angle calibration with random phantom[J]. Journal of X-ray science and technology, 2017, 25(4): 641-652
[15] SUN Y, HOU Y, ZHAO F Y, et al. A calibration method for misaligned scanner geometry in cone-beam computed tomography[J]. NDT and E International, 2006, 39(6): 499-513