基于图像处理的牛顿环应力测量方法

Measurement Method of Newton's Ring Stress Based on Image Processing

根据牛顿环受力变形导致干涉图像发生微小变化的特点,提出了一种基于图像处理的牛顿环应力测量方法。首先采用自行设计的可安装在读数显微镜上的CMOS图像采集装置代替人眼作为接收器采集牛顿环图像;然后利用直方图均衡化与高斯高通滤波算法对牛顿环图像进行增强处理,增加其对比度,通过SUSAN算子与霍夫变换有效地对牛顿环图像的边缘不连续圆进行补足处理,对处理的图像做垂直投影,实现在不断施加压力情况下牛顿环半径的测量;最后通过实验验证了应用牛顿环应力变形图像测玻璃体应力的可行性,应用图像处理测得透镜曲率半径的相对误差缩小到0.2%~6.5%范围内,测得应力的相对误差缩小到0.7%~9.9%范围内,应用图像处理测量玻璃体应力,误差减小且方便。

According to the tiny change characteristics of the interference image,which was caused by deformation of Newton＇s ring under stress,a measurement method of Newton＇s ring stress based on image processing was proposed.First,a CMOS image capture device was installed on the reading microscope in replace of human eyes as a receptor.Then,the image of Newton＇s ring is processed to improve its contrast by the algorithm of histogram equalization and Gaussian high-pass filtering. It can make up the discontinuity circle of Newton＇s ring image edge and process vertical projection of the image to measure the parameter of Newton＇s ring under the condition of constant pressure by SUSAN operator and Hough transform. Finally,the application,that the vitreous stress measurement by the deformation of Newton ring image,was completed for verification. The relative error of the curvature radius of lens and the measured stress decreases to the range from 0. 7% to 9. 9% because of the application of image processing. The using of image processing to measure vitreous stress can reduce the error and be more convenient.