摘要
相移轮廓术是一种广泛使用的光学三维测量方法,其精度不仅受相位展开算法本身的影响,也受测量系统中投影仪和摄像机的非线性影响。理论上,投射更多的相移条纹可减弱非线性误差的影响,但是增加了测量时间。为了提高误差校正的效率,提出了一种基于梯形正弦相移的测量方法。该方法需要两组改进的梯形相移条纹和一幅正弦条纹。梯形条纹提供图像强度信息和条纹级次信息,图像强度信息用来求取系统的非线性响应曲线,进一步消除系统的非线性。正弦条纹经过希尔伯特变换可求得额外的条纹图像,用来计算截断相位信息。经过校正的截断相位信息,可进一步获取精度较高的三维信息。相较于先前的梯形与正弦误差校正方法,该方法的测量效率提高了28%。
The phase shifting profilometry(PSP)is a widely used optical three-dimensional measurement method.Its accuracy is affected not only by the phase unwrapping algorithm,but also by the nonlinearity of the projector and camera in the measurement system.Theoretically,the effect of nonlinear errors can be reduced by projecting more fringes,but definitely increases the measuring time.To improve the efficiency of errors correction,a measurement method based on trapezoidal plus sinusoidal phase shifting was proposed.This method required two groups of improved trapezoidal phase shifting fringes and only one sinusoidal fringe.The trapezoidal fringe provided the information of image intensity and fringe order.The image intensity information could be used to obtain the nonlinear response curve of the system and further eliminate the nonlinearity of the system.The sinusoidal fringe could obtain the additional fringe by Hilbert transform to calculate the truncated phase information,and the corrected truncated phase information could be used to obtain the three-dimensional information with higher precision.Compared with the previous trapezoidal plus sinusoidal errors correction method,the measurement efficiency of proposed method increased by 28%.
作者
张国鸣
郝志东
赵奇
张志文
蔡柏林
李兵
ZHANG Guoming;HAO Zhidong;ZHAO Qi;ZHANG Zhiwen;CAI Bolin;LI Bing(Storage and Transportation Center of Shenhua Beidian Shengli Energy Co.,Ltd.,Xilinhot 026015,China;School of Internet,Anhui University,Hefei 230039,China;Hefei Xiaobu Intelligent Technology Co.,Ltd.,Hefei 230093,China)
出处
《应用光学》
CAS
CSCD
北大核心
2022年第2期304-310,共7页
Journal of Applied Optics
基金
国家自然科学基金(61935008)。
关键词
相移轮廓术
非线性
梯形正弦相移
希尔伯特变换
phase shifting profilometry
nonlinearity
trapezoidal plus sinusoidal phase shifting
Hilbert transform