Digitalized inspection of surface defects on precision component is one of the key technical problems in the field of optical manufacturing and testing. Large precision components are widely used in the fields of optical frontier and electronics industry, such as space optics, inertial confinement fusion, high power laser, micro-optics, extreme ultraviolet lithography, ultra large scale integrated circuits and so on, which demands a strict quality control on component surface. Defects are of different shapes and sizes and are randomly distributed on surfaces of tens or hundreds of millimeters, so it is challenging to achieve the submicron resolution in surface defect testing. Presently the visual method, the filter imaging method, the angular spectrum analysis method and some other methods are used in surface defects detection. However, these methods are merely experimental solutions and fail to achieve quantitative automatic testing instrumentation of high efficiency due to limited field-of-view and other reasons. The testing precision of submicron can be achieved by using scattering light caused by surface defects, so scattering testing method has a great potential in instrumentation for surface defects detection. The existing researches mainly focus on the modulation effects and damage mechanism of surface defects, yet the theoretical system of scattering testing is still incomplete, and there are still fundamental problems and technical issues to be solved and to be further studied.This dissertation mainly research on the scattering imaging technology and system for surface defects testing, including precise modeling analysis of scattering imaging using finite-difference time-domain method, the principle and scheme of microscopic scattering dark-field imaging, the calibration of the optical distortion error in the system, the research on the reverse recognition algorithm of defects detection. The research has an important significance for the implementation of precision surface defect scattering imaging detection technology. The major contents of this dissertation include:The important application significance of precision surface defects detection technology in modern optical testing and optical manufacturing is discussed, The international standard, United States standard and Chinese standard on the precision surface defects are compared, and the limitations of the existing standards are pointed out; After a review over the current surface defects detection testing technologies, especially those technologies based on the scattering mechanism, the necessity of the research on the objective, rapid, automated detection precision surface defect detection testing system and standard is put forward.The principle and scheme of the microscopic scattering dark-field imaging are introduced for the surface defects detection. The double magnification scanning method, the fast sub-aperture stitching algorithm and the digitalized calibration method are highlighted, as well as the corresponding basis is described in detail.Based on the Finite-Difference Time-Domain method, the simulation model of the surface defects scattering imaging is proposed. The influence of defects in the inertial confinement fusion system is analyzed by using the surface scattering theory. The oblique incident source of broadband spectrum is simulated and studied. The numerical method of far-field scattering imaging is also researched. The related factors to the scattering imaging, such as the incident angle of source, the receiving angle of sensor, the shape of defects, the width and depth of defects, is analyzed in detail. The simulation results provides theoretical basis for the analyses of the scattering imaging process as well as the recognition process of defects.The testing error caused by distortion in the machine vision system is analyzed. The calibration model of microscopic scattering dark-field imaging system is built. The calibration and distortion correction method is proposed for the precision measurement system as well. A pattern board is designed and fabricated by using EBE and IBE to calculate the intrinsic parameters of the system. The above theories and methods are functioned as a whole.After a study of the microscopic scattering dark-field imaging process, especially the PSF of the system, the electromagnetic simulation data base of micro-defects is established to support the testing and recognition process, which can solve the width error and provide depth information of defects. This principal of reverse recognition is proposed and simulated, which enable the system to offer the 3D characteristics of defects.The surface defects evaluation system based on the microscopic scattering dark-field imaging principle has been established. The calibration and distortion correction experiment has been carried out, and the calibration precision and distortion error has been assessed. The calibration precision can reach submicron, and the relative distortion in the edge field-of-view has been reduced from 4% to less than 0.1%. The reverse recognition experiment for width and depth of surface defects has been performed as well. The experiment results have been proved accurate by compared with those of SEM and profiler, with the precision of depth testing better than 100nm. The feasibility of the proposed system and the testing methods has been demonstrated with the experimental results, and some suggestions on the further study are given at the end. This dissertation is of great value for the performance improvement of the surface defects evaluation system based on the microscopic scattering dark-field imaging principle. |