针对传统大视场立体视觉标定方法存在操作不便、精度不高、稳定性差的问题，提出基于散焦图像的大视场立体视觉标定方法.首先，在立体视觉系统和大视场测量位置之间选定一个位置作为小视场散焦位置，利用在此位置处采集的靶标图像，实现摄像机内参标定；其次，利用大视场测量位置处采集的靶标图像，实现立体视觉外参标定；最后，通过实验室标定实验和直升机桨叶运动参数测量实验进行了验证.结果表明，该方法操作方便、精度高、稳定性好，在4.6 m×2.3 m的视场范围内，测量靶标上特征点间距为505.00 mm的两个点，平均误差为0.647 mm，均方根误差为0.780 mm，能够满足直升机桨叶运动参数测量的现场标定以及测量精度要求.
A calibration method based on defocused image is proposed for stereo vision sensor with large field of view (FOV) to simplify operation process and gain accurate and steady camera calibration. First, a defocusing position is selected between the stereo vision system and the measurement position. The intrinsic parameters of cameras are obtained by using the template images at the defocusing position. Second, the external parameters of the stereo vision are obtained by using the template images at the measurement position. Finally, a laboratory calibration experiment and an experiment of measuring the helicopter blade motion parameters are conducted to verify the proposed method. Experimental results show that the method features in simple operation, high accuracy and reliability. In a field of 4.6 mm×2.3 mm, the mean error and the RMS error of a 505.00 mm distance between two feature points on the target are 0.647 mm and 0.780 mm respectively. Therefore this method can meet the requirements of onsite calibration with high-precision for the measurement of helicopter blade motion parameters.
[1] Sun J, Hossain M M, Xu C L. A novel calibration method of focused light field camera for 3-D reconstruction of flame temperature[J]. Optics Communications, 2017, 390:7-15.
[2] 王晶,苏光大. 改进的双目立体视觉正面脸合成[J]. 应用科学学报,2014, 32(1):74-78. Wang J, Su G D. Frontal face synthesis based on improved binocular stereo vision[J]. Journal of Applied Sciences, 2014, 32(1):74-78. (in Chinese)
[3] Peng J Q, Xu W F, Yuan H A. An efficient pose measurement method of a space noncooperative target based on stereo vision[J]. IEEE Access, 2017, 5:22344-22362.
[4] 欧巧凤,赵平均,熊邦书. 基于立体视觉的旋翼共锥度动态测量系统精度分析[J]. 仪器仪表学报,2015, 36(8):1692-1698. Ou Q F, Zhao P J, Xiong B S. Accuracy analysis of the measuring instrument for taper angle of running rotor blades based on stereo vision[J]. Chinese Journal of Scientific Instrument, 2015, 36(8):1692-1698. (in Chinese)
[5] Jia Z Y, Wang L L, Liu W. A field measurement method for large objects based on a multiview stereo vision system[J]. Sensors & Actuators A:Physical, 2015, 234:120-132.
[6] Xiao Z Z, Liang J, Yu D H. An accurate stereo vision system using cross-shaped target self-calibration method based on photogrammetry[J]. Optics & Lasers in Engineering, 2010, 48(12):1252-1261.
[7] Sun J H, Liu Q Z, Liu Z. A calibration method for stereo vision sensor with large FOV based on 1D targets[J]. Optics & Lasers in Engineering, 2011, 49(11):1245-1250.
[8] 刘巍,李肖,马鑫. 采用复合式靶标的近景大视场相机标定方法[J]. 红外与激光工程,2016, 45(7):230-236. Liu W, Li X, Ma X. Camera calibration method for close range large field of view camera based on compound target[J]. Infrared and Laser Engineering, 2016, 45(7):230-236. (in Chinese)
[9] Sun J H, Liu Z, Zhang G J. Camera calibration based on flexible 3D target[J]. Acta Optica Sinica, 2009, 29(12):3433-3439.
[10] Wang Z X, Wu Z Q, Zhen X J. A two-step calibration method of a large FOV binocular stereovision sensor for onsite measurement[J]. Measurement, 2015, 62:15-24.
[11] Liu Z, Li F J, Li X J. A novel and accurate calibration method for cameras with large field of view using combined small targets[J]. Measurement, 2015, 64:1-16.
[12] Yu L, Han Y Y, Nie H. A calibration method based on virtual large planar target for cameras with large FOV[J]. Optics & Lasers in Engineering, 2018, 101:67-77.
[13] Zhang Z Y. A flexible new technique for camera calibration[J]. IEEE Transactions on Pattern Analysis & Machine Intelligence, 2000, 22(11):1330-1334.
[14] Ahmed M, Farag A. Nonmetric calibration of camera lens distortion:differential methods and robust estimation[J]. IEEE Transactions on Image Processing, 2005, 14(8):1215-1230.
