Optical Components Detection Based On DSP And Image Processing

With the development of the science and technology, optical componets play a very important part on the industry production, science tecnology, military and our life. The procession of the optical componets includs cutting, milling molding, coarse grinding, fine grinding, polishing and etc. The machining procession will makes the suface of the componets smooth and tidy, but may also produce some flaws and defects inevitably on the elements, such as scratches, edge brokens, pittings, bubbles, and points breaking and etc. The flaws and defects not only affects the appearance of the element, but also affecting the functons of the system and othe components which will lead to a fatal issue. Due to some defects such as scratches or broken edge have obvious features of geometry on height and width, the detection of the defects and damages is mainly using visual method currently. Although the human eyes identifination is simple and has no aspects of the equipment, labor-intensive, inefficient, large erros and long testing period are the major drawback. While detection of the subtle defects and damage aways need complex and sophisticated laboratory equipments and other conditons, both the detection and procession is separated from etch other, these are inconsistent with the line detection, real-time control on modern indusrual requirements.Against of the background of the Related research in the laboratory, Proposing the image processing method to detect the optical element which can improve the detection efficiency without reducing the effrect. This paper deals with the study of the image detection system structure. Using DSP to collect image date and transmist to the host PC, in the host PC, mosaic the image and process the results and statistic the date to juge the elements’completation. In the image stitching part, extracting the characters to fuse image, and in the detection part, conrasting the mosaic image with the standard image to judge the components completion by statistical information. Use the image processing algorithms to process the image to ger thr purpose of the detection ultimately.The paper introduced the related technology about DSP hardware, image throry and algorithms which is used for system development. Propossing future research content and focus which will produce practical application in industrial production and lay the foundation of the science and theory.