Method for detecting optical errors in large surface panels

• US 6,509,967 B1
• Filed: 07/06/1999
• Issued: 01/21/2003
• Est. Priority Date: 10/18/1996
• Status: Expired due to Term

• Current Assignee: ISRA Surface Vision GmbH (Atlas Copco AB)
• Original Assignee: Innomess Gelsellschaft Fur Messtechnik mbH
• Inventors: Pingel, Ulrich, Niepel, Christian
• Primary Examiner(s): STAFIRA, MICHAEL PATRICK
• Application Number:
  US09/284,437
• Time in Patent Office: 1,295 Days
• Field of Search: 356/230.1, 356/239.2, 356/239.7, 356/237.2, 356/376, 250/562, 250/563, 250/572
• US Class Current: 356/239.1
• CPC Class Codes: G01N 2021/9586 Windscreens
• CPC Class Codes: G01N 21/455 Schlieren methods, e.g. for gradient index determination; Shadowgraph
• CPC Class Codes: G01N 21/958 Inspecting transparent materials or objects, e.g. windscreens for conveyed flat sheet or rod G01N21/896

Abstract

The invention relates to a method for detecting optical errors, especially in refractive power, in large surface panels made of transparent material such as glass, by evaluating an observed image. The steps of the inventive method include: projection of a defined model consisting of regular sequences, whereby the sequences includes two varying light intensities and arrangement of the panel in the optical path of projection. The inventive method enables detection of optical errors in at least one dimension of a panel without a reference screen by projecting the model onto a camera, whereby a model sequence is respectively represented on a certain number of adjacent camera pixels and the number is an integral multiple of the sequence.

Claims

1. 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 an observed image, comprising the steps:

(a) Aiming of a defined pattern composed of regular sequences of a light intensity and a dark intensity towards a camera, thus defining a beam path;

(b) arranging the pane in the beam path of the pattern;

(c) imaging of the pattern beyond the pane on the camera; and

(d) arranging the pattern such that each observed sequence of the pattern is exactly imaged on an integer number of adjacent pixels of the camera, wherein said integer number is selected from the group of integer numbers consisting of two, three, four and five.

2. 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 an observed image, comprising the steps:

(a) aiming of a defined pattern composed of regular sequences towards a camera, thus defining a beam path, with the sequences comprising at least two different light intensities;

(b) arranging of the pane in the beam path of the pattern;

(c) imaging of the pattern beyond the pane on the camera with a sequence of the pattern in each case being imaged on a number of adjacently arranged pixels in the camera, said number being an integer multiple of said sequence;

(d) wherein a record for the evaluation of a moiré

image is based on the pixels of the camera.

3. The method as claimed in claim 2 wherein the sequence of the pattern comprises at least two light intensities, and wherein the integer multiple is at least three.

4. The method as claimed in claim 2 wherein a grid is provided for producing the pattern and wherein the grid comprises equidistant opaque and transparent strips in order to produce a regular light/dark pattern.

5. The method as claimed in claim 2 wherein a grid is provided for producing the pattern and wherein the grid comprises transparent squares which are bounded by opaque lines having the same width as the squares, so that essentially one quarter of the grid is transparent and wherein the camera is one of a line-scan camera and a matrix camera.

6. The method as claimed in claim 5 wherein a two-dimensional sequence of the pattern comprises a transparent square and two crossing opaque lines, with the surface area of the crossing lines being coincident with the transparent square.

7. The method as claimed in claim 6 wherein the two-dimensional sequence is imaged on respectively adjacent n×

n pixels of a matrix camera, where n is an integer multiple of the sequence greater than two.

8. The method as claimed in claim 2 wherein a grid is provided for producing the pattern and wherein the grid is formed as a checkerboard pattern composed of opaque and transparent congruent squares and wherein the camera is one selected from the group consisting of line-scan cameras and matrix cameras.

9. The method as claimed in claim 8 wherein a square surface area comprising two pairs of light and dark squares, which are arranged diagonally offset with one another, are imaged on respectively adjacent n×

n pixels of a matrix camera, where n is an integer greater than two.

10. The method as claimed in claim 2 wherein a sequence comprises at least three different light intensities, wherein the integer multiple is at least two and wherein the camera is one selected from the group consisting of line-scan cameras and matrix cameras.

11. The method as claimed in claim 2 wherein a grid is provided for producing the pattern and wherein the grid comprises a strip pattern of different light permeabilities.

12. The method as claimed in claim 11 wherein the grid comprises a second strip pattern of different light permeabilities at right angles to the strip pattern so that the superimposition of the strip patterns produces a pattern of rectangles of different light permeabilities, with at least three different light permeabilities overall.

13. The method according to claim 10 wherein two records for the evaluation of a moiré

image are based on the pixels of the camera.

14. The method as claimed in claim 2 wherein the pattern comprises a large number of strips which can be actuated selectively in such a manner that each strip appears either light and dark, wherein each n-th strip in the sequence is illuminated at the same time, where n is at least equal to three and wherein n adjacent strips are imaged successively on exactly an integer multiple of one pixel of the camera.

15. The method as claimed in claim 2 wherein the camera is one selected from the group consisting of line-scan cameras and matrix cameras which is displaced in order to scan a plurality of sequences.

16. The method as claimed in claim 2 comprising an evaluation step in which a change in the refractive power with respect to an illumination pattern in the pane is determined at the corresponding point on the basis of a phase shift which occurs on the pixels of the camera.

17. The method as claimed in claim 16 wherein the evaluation step comprises the formation of a difference in the phase profile of the image with a light-refracting pane to a state without light refraction.

18. The method as claimed in claim 16 wherein the refractive power is determined by the differentiation of the angle through which the light beam is deflected in the pane.

19. The method as claimed in claim 2 wherein the pane is one of curved glass and plastic.

20. The method as claimed in claim 2 wherein the camera is a color camera and wherein the evaluation of the same image is carried out with respect to the different recorded colors.

21. 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 an observed image, comprising the steps:

(a) aiming of a defined pattern composed of regular sequences towards a camera, thus defining a beam path, with the sequences comprising at least two different light intensities;

(b) arranging of the pane in the beam path of the pattern; and

(c) imaging of the pattern beyond the pane on the camera, with a sequence of the pattern in each case being imaged on a number of adjacently arranged pixels in the camera;

wherein a grid comprising a strip pattern of different light permeabilities is provided for producing the pattern; and

wherein the light permeability of the grid varies as a function of the wavelength of the light.

22. 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 an observed image, comprising the steps:

(a) aiming of a defined pattern composed of regular sequences towards a camera, thus defining a beam path, with the sequences comprising at least two different light intensities;

(b) arranging of the pane in the beam path of the pattern;

(c) imaging of the pattern beyond the pane on the camera, with a sequence of the pattern in each case being imaged on a number of adjacently arranged pixels in the camera; and

(d) selectively actuating lines and columns of a matrix of LEDs, wherein a projection of the LEDs is aimed directly to the pane.

23. The method according to claim 22, wherein the LEDs of the matrix are actuated to define on pattern selected from the group consisting of strips and checkerboard.

24. A method for determining the refractive power in large-area panes composed of a transparent material such as glass, by evaluation of an observed image, comprising the steps:

(a) aiming of a defined pattern composed of regular sequences towards a camera selected from the group consisting of line-scan cameras and matrix cameras, thus defining a beam path, each sequence comprising at least two different light intensities;

(b) arranging of the pane in the beam path of the pattern;

(c) imaging of the pattern beyond the pane on the camera;

(d) recording the pattern with the camera; and

(e) evaluating variations in a Moire image generated by said defined pattern and on the pixels of the camera for determining local distortion angles in the pane.