Newton's rings are the fringe patterns of quadratic phase, the curvature radius of optical components can beobtained from the coefficients of quadratic phase. Usually, the coordinate transformation method has been usedto the curvature radius, however, the first step of the algorithm is to find the center of the circular fringes. Inrecent years, deep learning, especially the deep convolutional neural networks (CNNs), has achieved remarkablesuccesses in object detection task. In this work, an new approach based on the Faster region-based convolutionalneural network (Faster R-CNN) is proposed to estimate the rings' center. Once the rings' center has beendetected, the squared distance from each pixel to the rings' center is calculated, the two-dimensional pattern istransformed into a one-dimensional signal by coordinate transformation, fast Fourier transform of the spectrumreveals the periodicity of the one-dimensional fringe profile, thus enabling the calculation of the unknown surfacecurvature radius. The effectiveness of this method is demonstrated by the simulation and actual images.
展开▼
