Research On The Detection System Of Large Apertue Curved Optical Element And Image Processing Of Surface Defects

The surface of fused silica surface optical elements under laser irradiation is prone to defects.If not repaired in time,the size of surface defects will increase exponentially.Therefore,it is necessary to detect and repair fused silica surface optical elements off-line.Due to the large number and small size of surface defects in fused silica optical elements,a high resolution camera is required to quickly scan the defects and process the obtained defects images to obtain defect information.Based on the established multi-degree-of-freedom fused silica optical element positioning platform,this project will develop a surface optical element detection system and surface defect recognition algorithm.So it can realize accurate of surface defects on the surface organ disease,effective detection.Firstly,the overall scheme of surface defects detection system of surface optical elements of curved surface is designed and analyzed from three aspects of detection principle,scanning path and image processing.The optimal incident Angle of linear array light source,the effective length of linear array CCD and the improved LASNR threshold segmentation algorithm are designed to solve the key problems in system development.The image positioning error are analyzed,and through the experiment to calculate the number of pixels dislocation in the whole caliber splicing,proposed solutions should sex transformation matrix of single linear CCD and the Y axis Angle error introduced by the pixel displacement of machine tool,so as to realize the image full diameter without dislocation joining together,improve the positioning precision of the fault.In order to solve the problem of surface defect of curved optical element,the camera can't focus on the surface of the element in real time because of the low depth of field and the lower slope of the surface than the ambassador camera.Firstly,the ideal equation of optical element surface is obtained through the real-time range of spectral confocal distance.The error rotation matrix is calculated by the least square method,and the actual geometric equation of the surface of the surface is obtained by the error compensation of the ideal surface equation of optical element.The Z axis is controlled with the aid of the actual equation of the surface,so that the camera can focus on the camera in real time during the full-aperture scanning process,so that the image acquired by the dark field scanning is clear and reliable.In order to make the system more efficient in surface dark field scanning of surface optical elements of curved surface,the functions of field vision positioning,full-caliber image Mosaic and sub-caliber image preview were designed.The full-aperture high-efficiency acquisition of surface defects of large-aperture surface optical elements is realized.In order to solve the problem of surface defect of fused silica surface optical elements,the image background is fluctuated greatly due to the change of surface curvature,and the background is mixed with the target.Gaussian filtering and top hat transform are used to preprocess the image,and improved LASNR algorithm is used to binarize the image,so that the image defect information can be extracted quickly and effectively.In addition,the collected image of field defects was reconstructed in three dimensions to obtain the actual location,total diameter distribution and shape size of defects on surface elements.In this paper,the designed surface element dark field test system contains images,dark frame acquisition fault information accurately and effectively extract,to prevent Ming field of vision,according to the location,component surface fault information of 3D reconstruction key functions,such as size can be detected for more than 10 microns of fault information,provide technical support for the fault repair.