点光源非定域干涉同心圆环半径提取算法研究

Research on extraction algorithms of concentric circle radius in non localized interferences of point light source

目的:解决点光源非定域干涉圆环图像中圆环半径精确提取问题。方法:提出了一种基于霍夫变换的形态学中心快速定位算法。首先读入图像进行预处理,然后对干涉同心圆环进行霍夫变换提取圆心,最后使用形态学中心快速定位算法获得各圆环的半径值,并分别通过仿真图片和实验图片的处理验证了该算法的可行性和准确性。结果:算法提取存在-8.58μm的系统误差。补偿该误差后,对仿真圆环的识别误差在1μm以下。对于实验图片,同时使用了游标卡尺对图像样本进行测量,且与算法得到的各环参数进行比对,其数据差值均在1μm以下。结论:算法能够满足光学干涉测量所需的精度。

Aims:This paper aims to solve the problem of accurately extracting the radius of a point light source in non-localied interferometric circular ring images.Methods:A fast morphological center localization algorithm based on Hough transform was proposed.Firstly,the image was read in for preprocessing.Then,the center of the interference concentric ring was extracted using Hough transform.Finally,a morphological center fast localization algorithm was used to obtain the radius values of each ring.The feasibility and accuracy of the algorithm were verified through simulation and experimental image processing.Results:There was a system error of-8.58μm in the algorithm extraction.After compensating for this error,the recognition error of the simulated ring was below 1μm.For the experimental images,a vernier caliper was used to measure the image samples.The experimental images were compared with the results obtained by the algorithm.The differences were all below 1μm.Conclusions:The algorithm can meet the accuracy required for optical interferometry measurement.