Optics: measuring and testing – Inspection of flaws or impurities – Transparent or translucent material
Reexamination Certificate
1999-07-06
2003-01-21
Stafira, Michael P. (Department: 2877)
Optics: measuring and testing
Inspection of flaws or impurities
Transparent or translucent material
C356S239700
Reexamination Certificate
active
06509967
ABSTRACT:
DESCRIPTION
1. Technical Field
The invention relates to a method for determining optical errors, in particular in the refractive power, in large-area panes composed of a transparent material such as glass, by evaluation of the observed image, comprising the steps of projecting of a pattern composed of regular sequences, with the sequences comprising at least two different light intensities, and arrangement of the pane in the beam path of the projection.
2. Prior Art
EP-A-0 416 302 describes such a method. In this method, an illuminated flat grid is imaged via an objective on a reference grid, a pane to be checked is arranged in the beam path between the flat grid and the reference grid, and the superimposed image composed of the image from the flat grid and the reference grid is investigated; in this case, the flat grid is imaged on a reference grid whose surface size is smaller than that of the pane to be examined, and the superimposed image is recorded a plurality of times by a video camera in order subsequently to be evaluated using a phase-shift method, in which the recorded brightness distribution is used as a measure of the refractive power of the pane.
An extremely high level of complexity is required to carry out this method. In practice, the flat grid is generally a cruciform grid which is formed from alternating opaque strips and transparent strips, with the transparent strips being exactly as wide as the opaque strips, and with two such strip patterns being superimposed offset through 90° with respect to one another. The pane to be examined is arranged in the beam path and is imaged on the reference grid, which is a linear grid, and likewise comprises transparent and opaque lines, to be precise with the same width ratio as the flat grid. A first practical problem arises when the lines of the two grids coincide exactly; in order to avoid this situation, lines which are semi-opaque and semi-permeable are in practice generally used as the flat grid rather than opaque lines. This reduces the contrast.
In order to allow evaluation of the resulting moiré image using the phase-shift method, the superimposed image of the flat grid which is imaged on the reference grid has to be imaged three times. In this case, the reference grid must be shifted twice, in each case by one third of the width of a line pair, so that, overall, at least three records are required in order to determine the phase shift in one dimension (for example the horizontal) of the pane. If it also desired to determine the refractive power of the pane in a dimension running at right angles to this (for example the vertical), three records of the resultant moiré image must be made once again, from a second reference grid. Thus, overall, six records are required in order to allow the refractive power of the pane to be determined using the phase-shift.
The apparatus for carrying out the method is correspondingly complex, since shifting the reference grid by one third of the width of a line pair on the reference grid must be carried out really precisely. The same is true for the second record, at right angles to the first, since it is difficult to avoid an undesirable moiré effect occurring due to positioning errors. A large amount of time is required to measure the refractive power, owing to the complex handling.
In addition, the known method can lead to undesirable moiré fringes on the “pixel period” of the camera (see column 4, lines 36-40) if the illumination pattern or the reference pattern make up a multiple of one period of the pixels. Measures therefore have to be taken to avoid the occurrence of this constellation, since these undesirable additional moiré images corrupt the evaluation of the image of the pane.
EP-A-0 559 524 describes another method, namely for testing the transparency in particular of laminated glass after the initial assembly process and before autoclaving, that is to say at a time at which the initial assembly, or the interlayer, as a rule has a milky color which impedes light transmission. This transmission method uses a light source arranged on one side (underneath) the initial assembly and a camera on the other side of the initial assembly in order to monitor the test image produced. The test image produced by the light source and projected is a line pattern comprising a small number of lines. A mean value from all the observed values is used as the basis for deciding whether a laminated glass pane is “good” or “poor”. No specific imaging rule for the lines on the camera and its pixels is proposed. It is also impossible to detect errors in the refractive power, small inclusions or the like, since they have only a minimal effect on the measured mean value over the entire observed image.
The mathematical derivatives of angles originating from measured moiré images as well as a summary of the various moiré techniques are given in the article by Selb, M., and Höfler, H. in “Vision & Voice Magazine”, Volume 4, (1990), No. 2, pages 145-151. This article also deals with high-resolution moiré topography measurements by gratings imaged directly onto a CCD chip, that is to say with single-stage imaging without a reference grating.
DESCRIPTION OF THE INVENTION
The object of the invention is to specify a method as claimed in the preamble of claim 1, using which optical errors in at least one dimension of a pane can be determined without a reference grid.
This object is achieved by the features in the descriptive steps of the claims, including imaging the pattern onto a camera, with a sequence of the pattern in each case being imaged onto a number of adjacently arranged pixels of the camera, and the number being an integer multiple of the sequence.
A sequence of the pattern can be defined by a periodic sequence of two or more light intensities. In the simplest case, this is a sequence in which light and dark strips, preferably of identical width, alternate with one another and form a light/dark sequence. However, it is also possible for the sequence to be composed of three, four or more strips, which have a regular sequence with intensity minima and maxima that are always equidistant.
In order to produce these sequences it is, on the one hand, possible to produce the light intensities by the local light permeabilities of a physical grid, by means of a light source arranged behind the grid. Where the grid is opaque, the light intensity is zero and the point is dark; where the grid is completely transparent, the light intensity assumes a maximum. The use of a physical grid as in the prior art, that is to say comparable to a large screen or filter, is adequate for sharp light/dark sequences. However, semi-transparent filters must be provided to produce sequences with strips of different brightness, which filters would possibly have to have three or more different light permeabilities, reproduced very precisely.
A light wall is preferably provided in an apparatus for carrying out the method, which light wall is used in the method according to the invention for projecting a pattern with regular sequences, and can be used instead of a light source with a grid. The light wall is expediently composed of a large number of individual LEDs which can be actuated as required individually, in blocks or in lines and columns in order either to illuminate or not to illuminate in accordance with a light/dark profile, or in order to emit different intensities as a function of a suitable characteristic. Similar light walls are used, for example, as display panels in sports stadiums. It is self-evident that the apparatus for determining optical errors, comprising a light wall composed of a plurality of light areas which can be actuated individually as a flat grid which is projected onto a pane whose refractive power is intended to be determined also works when it is used with a reference grid from the prior art. It is self-evident that, in principle, the light/dark sequence can be displayed with either or the two grids. It is furthermore self-evident that the sequences can also be enlarged via lenses, before the
Niepel Christian
Pingel Ulrich
Innomess Gelsellschaft fur Messtechnik mbH
Mersereau C. G.
Nikolai & Mersereau , P.A.
Stafira Michael P.
LandOfFree
Method for detecting optical errors in large surface panels does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method for detecting optical errors in large surface panels, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method for detecting optical errors in large surface panels will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3066903