结构光测量中相位误差的过补偿与欠补偿校正

针对现有结构光测量中采用的相位误差补偿算法存在的相位误差过补偿或欠补偿问题,提出了一种新的相位补偿误差校正算法。推导了环境光下四步相移的相位误差数学模型,解释了相位误差过补偿、欠补偿的产生原因;通过数学推导获得相位误差的解析表达式,提出了相位误差过补偿、欠补偿的校正算法。该方法通过向标定平面投射4步相移图像、16步相移图像与黑白图像获得相位误差系数;然后在8种不同环境光条件下重复这一步骤获得多组系数;最后运用参数拟合法获得相位误差数学模型的具体表达式。实验结果表明:无论在黑暗环境还是光环境下,利用该修正方法进行相位误差补偿后均可使相位精度达到0.002rad,比进行相位误差补偿前提高了8.6倍左右,比查找表(LUT)提高了2.5倍左右。该算法精度高,速度快,能有效解决相位误差过补偿、欠补偿的问题。

Correction of phase error overcompensation and under-compensation in structured light measurement

In structured light measurement, the current phase error compensation algorithms always lead to phase error overcompensation or under-compensation as the ambient light is varied. Therefore, a new correction method for the phase compensation errors was proposed. Based on analyzing the impact of ambient light on phase errors of four-step phase-shifting method, the reasons of phase error overcompensation and under-compensation were illuminated, the expression of the phase errors was deduced and a method to correct the phase compensation error was proposed. To compensate phase error accurately, a group of four-step phase-shifting images, sixteen-step phase-shifting images, black and white images were projected to the calibration plane, then the operation was repeated in eight different ambient light conditions to obtain several groups of coefficients. Subsequently, the analytical expression was obtained by a curve fitting. Experimental results show that the accuracy of phase error compensation is 0.002 rad no matter under dark or light environments when the method is applied, which is about 9.6 times higher than that without phase error compensation and about 3.5 times higher than that with Look-Up-Table(LUT) compensation only. This method is characterized by higher accuracy and reliability and can be used to compensate the phase errors in varied ambient lights accurately.