Research On Optical Surface Rickets Detection System And Image Processing Technology

The quality of optical components on the surface will produce huge impact to the whole optical system. Any point mishandling is likely to cause surface flaws in element production process. So the study of flaws detection becomes very meaningful to optical components.In order for the surface flaws further study, this paper designs a set of optical components surface flaws detection system and sets up a digital CCD image acquisition and computer processing system based upon the principle of micro scattering dark field imaging. According to the scattering properties of the optical components surface flaws light and using the principle of imaging to detect flaws, the system uses the laser irradiation sub-region component surface and captures flaws image by a digital CCD acquisition, then sent to the computer for storage. On the image stitching algorithm, this paper proposes an improved classification algorithm of multi-level and sub-region based on feature classification. The algorithm handles aperture flaws image according to feature and splices the smaller image blocks into larger tiles until a full aperture image for different mosaic method. It is similar to the process of regional growth. Then the system processes flaws image enhancement, filtering and binarization pretreatment by using digital image processing technology. Finally it extracts the flaws image feature and locates pixel positioning. So detecting flaws comes to an end on the entire image.The system has large detection range, good stitching effect and high degree of automation compared with the traditional method of measurement. Experimental results show that the system can efficiently detect flaws of optical components surface.