A method and system for scatter detection of surface defects of a curved optical element

Abstract

The application provides a curved surface optical element surface defect scattering detection method and system, which comprises the following steps: setting a pre-focusing incident light beam in a detection light path, so that the incident light beam forms a virtual object point behind the curved surface optical element; adjusting the distance from the virtual object point to the vertex of the curved surface to satisfy a conjugate relationship with the distance from the detector to the vertex of the curved surface; discretizing the illuminated local curved surface region into an array of micro surface elements; calculating the projection of the unit vector of the scattered light and the unit vector of the incident light on the local tangent plane of each micro surface element to establish a local spatial frequency mapping relationship; performing surface-detector geometric modeling, micro surface BRDF calculation, radiation transmission and irradiance calculation of the scattered light field, constructing a curved surface scattering physical model and reconstructing the irradiance distribution on the detector target surface; performing total integral scattering processing on the irradiance distribution to identify defects, extracting the power spectral density of the defect region to obtain five frequency domain characteristic values, and inputting the five frequency domain characteristic values into a machine learning classifier for defect classification.

Images