Spy quantitative inspection with a machine vision light sectioning method

Describes a spy quantitative inspection system based on the principle of the line structure 3D machine vision method, also called the light sectioning method. It is intended for application to precision instruments and the inner wall surfaces of hole-like workpieces such as pipes, internal screws an...

Full description

Saved in:
Bibliographic Details
Published inMeasurement science & technology Vol. 11; no. 8; pp. 1187 - 1192
Main Authors Tu, Da-Wei, Lin, Cai-Xing
Format Journal Article
LanguageEnglish
Published Bristol IOP Publishing 01.08.2000
Institute of Physics
Subjects
Online AccessGet full text

Cover

Loading…
More Information
Summary:Describes a spy quantitative inspection system based on the principle of the line structure 3D machine vision method, also called the light sectioning method. It is intended for application to precision instruments and the inner wall surfaces of hole-like workpieces such as pipes, internal screws and axle sleeves. A light beam from a semiconductor laser diode is converged into a line shape by a cylindrical lens. A compact reflecting-refracting prism group ensures that the light is projected axially onto the inner surface. The deformed line is imaged onto a CCD sensitive area and the digitised image is captured on a computer by a 512 x 512 pixel card. Machine vision image processing methods such as thresholding, line centre detect and the least squares method have been developed for contour feature extraction and description. Two other important problems in such an inspection system are how to orientate the deep insertion optical probe and how to bring the projected line into focus. A focusing criterion based on image position deviation and a four-step orientation procedure have been devised to solve these problems. Experimental results show that the techniques are realizable and a good future for application in industry is possible. (Original abstract - amended)
Bibliography:ObjectType-Article-2
SourceType-Scholarly Journals-1
ObjectType-Feature-1
content type line 23
ISSN:0957-0233
1361-6501
DOI:10.1088/0957-0233/11/8/313