Optics: measuring and testing – Inspection of flaws or impurities – Surface condition
Reexamination Certificate
1999-10-15
2001-02-20
Font, Frank G. (Department: 2877)
Optics: measuring and testing
Inspection of flaws or impurities
Surface condition
C356S237100, C356S237200, C356S237300
Reexamination Certificate
active
06191850
ABSTRACT:
FIELD OF THE INVENTION
This invention relates to machine vision systems for inspecting objects, and more particularly to systems that conduct surface inspection using structured illumination.
BACKGROUND OF THE INVENTION
It is often desirable to inspect generally flat uniform surfaces for defects. Surface inspection is used in a variety of industries. The inspection process can utilize automated machine vision techniques, manual inspection techniques, in which a human operator views objects through a remote camera, or a combination of manual and automated techniques.
By way of example, the production of laminated security cards, sometimes referred to as “Smart Cards,” involves the placement of a radio frequency-generating material between two layers of plastic sheet laminate. The finished cards are used in a variety of readers in which the response of the card to a predetermined radio frequency signal, transmitted by the reader, uniquely identifies the cardholder. Because of the substantial heat and pressure used to produce such cards, they are susceptible to the formation of defects on their surfaces during manufacture. These defects, which cause problems during the printing process, may be difficult to detect during visual inspection by a human operator, or even by automated inspection. In particular, the so-called dimple defect, which exhibits a gradual waviness emanating from a point on the card surface may be quite difficult to detect, and will only show up when the card is printed upon. Other types of defects may be equally difficult to spot, particularly when viewed using a remote camera that displays the card on a monitor.
Visualization of these defects can be made simpler by the use of structured illumination, where a pattern of light and shadow projected upon the surface enhances the visibility of any defects present. An effective and reliable way to determine whether surface defects are unacceptable is to measure the structured illumination grid features against predetermined tolerance values. This generally entails the numerical quantification of the underlying defects. However, accurate and reliable translation of the visual quality of a defect into a quantitative value is often difficult to accomplish. It is, therefore, an object of this invention to provide a system and method and apparatus for quantitatively inspecting a surface for defects. This system and method should enable easier and more reliable inspection by automated machine vision systems.
SUMMARY OF THE INVENTION
This invention overcomes limitations of the prior art by providing a system and method for effectively analyzing a pattern produced on a surface using structured light to determine underlying defects in the surface.
Briefly stated, a structured light, in the form of a grid pattern, is projected onto an object surface with the individual grid elements or “features” appearing as shadow lines thereon. The projected grid is viewed by an on-axis camera that is operatively interconnected to a machine vision system that resides, typically, as software on a computer. The machine vision system conducts a two-step analysis of grid features that first compares derived scores for grid features against desired tolerances and then compares locations of grid features against a set of ideal grid feature locations. Only if the viewed grid passes both steps of the test is the surface deemed acceptable.
According to a preferred embodiment, a structured light-projecting element is positioned between a light source and the surface to be inspected. The projecting element includes a first diffuser through which light passes and a second diffuser having a desired grid pattern overlaid thereon. A beam splitter, typically in the form of a 45-degree half-mirror, reflects the projected grid pattern onto the surface. Light is reflected back from the surface with a grid pattern defined (in shadow) thereon, through the beam splitter and back toward the viewing camera along the camera axis to strike the camera lens. According to another embodiment, a collimator and polarizer can be located between the second diffuser and beam splitter, and a similar type (characteristic) of polarizer can be located at the camera lens.
The machine vision system, according to the preferred embodiment, is adapted to analyze the projection of the grid onto the surface as viewed by the camera. The projected grid is first located using, for example, a fiducial that is formed in relation to the grid, and that has a different appearance than the features of the grid for easier identification. Once the grid is located, the machine vision system uses pattern recognition to identify and score specific features of the grid. The results are used in a two-step process to analyze the grid. In the first analysis step, the scores are compared against tolerance limits. If all feature scores do not fall within these tolerance limits, then the surface is immediately rejected as defective without proceeding to the second analysis step. If tolerance limits are met in the first step, then the second step proceeds. The data derived by the machine vision software related to the grid features is ranked by row and column so as to create a mathematical array of feature locations. Based upon this array, an ideal grid is computed within the software, and the locations of the actual grid features are compared with the ideal grid locations. Deviations from the ideal locations are again matched against tolerance limits. If these limits are exceeded, then the surface is rejected, otherwise the surface is passed as acceptable.
REFERENCES:
patent: 5003166 (1991-03-01), Girod
patent: 5841530 (1998-11-01), Hewitt et al.
Cognex Corporation
Font Frank G.
Loginov William A.
Punnoose Roy M.
Weinzimmer Russ
LandOfFree
System and method for inspecting an object using structured... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with System and method for inspecting an object using structured..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and System and method for inspecting an object using structured... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2615778