画像処理を用いたレンズ表面の欠陥検査 : 同心フレーム差分法を用いた欠陥の検出

概要

論文の詳細を見る
This paper describes the automated visual inspection system for optical transparent lenses which can recognize four kinds of surface defects. It is well known that the characteristic of transmitted light depends on the kind of surface defect of the lens. The influences of shading in image data must also be considered for image processing. Therefore, it is very difficult to develop an automated inspection system for the surface defects of lenses using image processing. If the features of surface defects can be recognized by a vision system, the nature of the defects will be clarified and the automated inspection system for the surface defects of lenses can be developed. Upon investigation, the optimum measuring condition was found to be one which can recognize four kinds of surface defects of optical lenses. Additionally, the new image processing algorithm, using the Concentric Frame Difference Method, can only recognize surface defects without the influences of shading. The kinds of surface defects can also be classified using the inspection system. The image processing algorithm, which is developed in the research, has great advantages. When using the image processing algorithm instead of the ordinary binary method, the gradation of defects is more greatly emphasized than the average image data. It is found that the visual inspection system, using the new image processing algorithm for the surface defect of lenses, is more effective in actual application.
2000-05-05

画像処理を用いたレンズ表面の欠陥検査 : 同心フレーム差分法を用いた欠陥の検出

スポンサーリンク

概要

著者

関連論文

スポンサーリンク