在贝塞尔结构光三角法表面形貌测量系统中,条纹图像预处理的效果直接影响系统的测量精度。在考虑图像固有“灰性”的基础上,利用灰色关联分析的方法对贝塞尔结构光条纹图像进行预处理,通过计算二维灰色绝对关联系数实现图像的自适应加权均值滤波。实验结果表明:相比高斯滤波、均值滤波和中值滤波而言,灰色关联分析的方法具有良好的去噪效果,同时很好地保持了图像的细节信息,有效地抑制由高斯噪声和冲激噪声组成的混合噪声。利用灰色关联分析的方法对贝塞尔结构光条纹图像进行预处理,为后续的条纹中心提取提供良好的基础。 In Bezier structured light triangulation surface topography measurement system, the effect of fringe image preprocessing directly affects the measurement accuracy of the system. Based on the gray The experimental results show that compared with Gaussian filter, mean filter and median filter, the method of gray relational analysis has a good denoising effect, while maintaining the details of the image well, effectively suppressing Gaussian noise and impulse Noise consisting of mixed noise. Gray relational analysis is used to pretreat the bezier structure light stripe image, which provides a good foundation for the subsequent extraction of the fringe center.