Web inspection method and device

• US 6,950,547 B2
• Filed: 02/12/2001
• Issued: 09/27/2005
• Est. Priority Date: 02/12/2001
• Status: Active Grant

• Current Assignee: 3M Innovative Properties Company (3M Company)
• Original Assignee: 3M Innovative Properties Company (3M Company)
• Inventors: Yu, Xi, Peick, Matthew P., Wageman, Steven R., Xu, Wenyuan, Skeps, Carl J., Masterman, James A., Floeder, Steven P.
• Primary Examiner(s): Bali, Vikkram
• Application Number:
  US09/781,372
• Publication Number:
  US 20020110269A1
• Time in Patent Office: 1,688 Days
• Field of Search: 382/141, 382143-151, 382/134, 348/86, 348/87, 348/88, 348/126, 348/127, 348/128, 348/129, 348/125, 356/237.1, 356/238.1, 356/239.3, 356/237.5, 700/95, 435/39, 250/459.1
• US Class Current: 382/143
• CPC Class Codes: G01N 21/89 in moving material, e.g. running paper or textiles G01N21/90, G01N21/91, G01N21/94 take precedence

Abstract

An imaging device for sequentially imaging a portion of a continuously moving web to provide a digital data stream which is then analyzed by a single computer without the used of dedicated signal processing hardware. Techniques for operating on the data stream from an imaging device are disclosed, particularly including operations based on blob information stored in terms of starting position and segment run lengths in a crossweb direction. These allow definitions of blobs to be accumulated in a line-by-line fashion, and allow classes of defects commonly found in continuous web manufacturing to be identified with far less computing power than was previously required. In particular, in the challenging application of inspecting flexible circuits, data rates in excess of 10 mega-pixels/second are achieved and successfully processed.

Claims

1. A method of inspecting a continuously moving web, comprising:

a) imaging a sequential portion of the continuously moving web to provide a digital data stream, wherein the digital data stream corresponding to each sequential portion describes pixels in an X domain corresponding to their position accross the web,

b) binarizing said digital data stream,

c) forming a blob list from the data stream by;

c1) determining collections of pixels connected to each other in the X domain so as to define segments, and

c2) resolving line to line whether connections exist between segments in a Y domain corresponding to the direction of web movement;

wherein the determining step and the resolving step are accomplished in a single iteration, and

d) analyzing blobs on the blob list to identify defects, wherein c) and d) occur in a single computer.

2. A method according to claim 1, further comprises saving in turn a list of the segments in each sequential line as a comparison list, and wherein

the resolving comprises comparing the segment list for the current line with the comparison list.

3. A method according to claim 1, wherein a filter is applied to the digital data stream in the single computer prior to forming the blob list.

4. A method according to claim 1, further comprising communicating between the single computer and a process control system.

5. A method according to claim 1, further comprising marking identified defects on the continuously moving web.

6. A method according to claim 5, wherein said marking occurs through ink deposition, paint deposition, laser tagging, label application, hole punching, physical deformation, magnetic pulsing or combinations thereof.

7. A method according to claim 5, wherein said marking occurs substantially near the point of occurrence of the defect.

8. A method according to claim 1, wherein said web is selected from metals, paper, polymeric films, wovens, non-wovens, glass or combinations thereof.

9. A method according to claim 8, wherein one or more coatings or one or more patterns are applied to said web.

10. A method according to claim 9, wherein said continuously moving web is a flexible circuit web.

11. A method according to claim 1, wherein said imaging occurs through reflected light transmitted light or transflected light.

12. A method according to claim 1, wherein multiple imaging sources are utilized.

13. A method according to claim 1, wherein said binarizing includes adaptive thresholding or multiple value thresholding.

14. A method according to claim 1, wherein said data stream is at least 10 mega-pixels/second.

15. A method according to claim 1, further comprising classifying defects into specific categories.

16. A method comprising:

a) imaging a sequential portion of the continuously moving web to provide a digital data stream,

b) forming a blob list from the data stream, and

c) analyzing blobs on the blob list to identify defects, wherein b) and c) occur in a single computer,

wherein the web is a patterned web, and further comprising binarizing the digital data stream prior to forming the blob list, the binarizing comprising;

identifying at least one sequential portion having substantially the entire range of optical properties characteristic of the web;

identifying the pixel values corresponding to local maxima and minima;

defining a range bounded by the lowest value among the pixel values identified as local maxima and the highest value among the pixel values identified as local minima;

calculating a threshold value within the range; and

comparing at least a portion of the digital data stream to the threshold value.

17. A method of inspecting continuously moving articles on a web, comprising analyzing blobs formed from a continuous digital data stream of at least 10 mega-pixels/second imaged from at least a portion of a continuously moving article to identify defects on the articles, wherein the blobs are formed and analyzed in a single computer, and further wherein the digital data stream describes pixels in an X domain corresponding to their position accross the web, the method further comprising:

a) binarizing said digital data stream; and

b) forming the blobs from the digital data stream by determining collections of pixels connected to each other in the X domain so as to define segments, and resolving line to line whether connections exist between segments in a Y domain corresponding to the direction of web movement, wherein the determining and the resolving are accomplished in a single iteration.

18. A method for inspecting continuously moving webs having a repeating pattern, the method comprising:

a) imaging sequential portions of the continuously moving web to provide a digital data stream, wherein the digital data stream corresponding to each sequential portion describes pixels in an X domain corresponding to their position accross the web,

b) identifying instances of the repeating pattern,

c) forming a blob list representative of each instance of the repeating pattern from the data stream, wherein the blob list includes information on the lengths of collections of pixels connected to each other in the X domain, and

d) analyzing blobs on the blob list to identify defects, wherein c) and d) occur in a single computer and the analyzing step comprises;

caculating information on the lengths of colletions of pixels connected to each other in a Y domain corresponding to the direction of web movement;

modifying the lengths of the collections of pixels in at least one of the X domain, the Y domain or both domains, by a first predetermined number;

preparing a new blob list based on the modified lengths; and

comparing the number of blobs on the new blob list against a second predetermined number.

19. A method according to 18, wherein the number of blobs on the blob list for each instance of the repeating pattern is compared against a predetermined number.

20. A method according to 18, wherein positional and geometric properties of each blob is compared against a corresponding blob in a reference blob list.

21. A method according to 18, wherein said imaging occurs through reflected light, transmitted light or transflected light.

22. A method according to claim 18, wherein the data stream is utilized to find individual patterns on said web without external synchronization.

23. A method according to claim 18, further comprising binarizing of said digital data stream prior to forming said blob list.

24. A method according to claim 23, wherein said binarizing occurs using adaptive thresholding or multiple value thresholding.

25. A method according to claim 18, wherein said continuously moving web is a flexible circuit web.

26. A method according to claim 25, wherein said defects include one or more of shorts, opens, lead reductions, space reductions, substrate defects, pattern misregistration, bent leads, covercoat defects, lamination defects, stains, or debris.

27. A method according to claim 18, further comprising marking one or more defects on said continuously moving web.

28. A method according to claim 18, wherein said marking occurs substantially near the point of occurrence of the defect.

29. A method according to claim 18, wherein said computer communicates with a process control system that controlls said continuously moving web.

30. A method according to claim 18, wherein said imaging device is spatially synchronized to the continuously moving web.

31. A method according to claim 18, wherein multiple imaging sources are utilized.

32. A method according to claim 18, further comprising classifying defects into specific categories.

33. A device for inspecting a continuously moving web, comprising

(a) An imaging device for sequentially imaging a portion of a continuously moving web to provide a digital data stream;

and

(b) A single computer capable of forming a blob list from the data stream and analyzing the blob list in order to identify defects in at least a portion of said continuously moving web,

wherein the digital data stream corresponding to each sequential portion describes pixels in an X domain corresponding to their position across the web, and

wherein the blob list includes information on the lengths of collections of pixels connected to each other in the X domain, and further wherein the computer analyzes the blob list by;

calculating information on the lengths of collections of pixels connected to each other in a Y domain corresponding to the direction of web movement;

modifying the lengths of the collecions of pixels in at least one of the X domain, the Y domain, or both domains by a first predetermined number;

preparing a new blob list based on the modified lengths; and

comparing the number of blobs on the new blob list against a second predetermined number.

34. A device according to claim 33, further comprising a process control system in communication with the single computer.

35. A device according to claim 33, further comprising a marking system for marking identified defects on the continuously moving web.

36. A device according to claim 33, wherein said imaging device is a line scan camera.

37. A device according to claim 33, wherein said imaging device utilizes optical assemblies which utilize reflected light, transmitted light or transflected light.

38. A device according to claim 33, wherein multiple imaging devices are utilized.

39. A device for inspecting flexible circuits, comprising

(a) An imaging device for sequentially imaging a portion of a continuously moving flexible circuit web to provide a digital data stream;

and

(b) A single computer capable of forming a blob list from the data stream and analyzing the blob list in order to identify defects in at least a portion of said continuously moving flexible circuit web,

wherein the digital data stream corresponding to each sequential portion describes pixels in an X domain corresponding to their position accross the web, and

wherein the blob list includes information on the lengths of collections of pixels connected to each other in the X domain, and further wherein the computer analyzes the blob list by;

calculating information on the lengths of collections of pixels connected to each other in a Y domain corresponding to the direction of web movement;

modifying the lengths of the collections of pixels in at least one of the X domain, the Y domain, or both domains, by a first predetermined number;

preparing a new blob list based on the modified lengths; and

comparing the number of blobs on the new blob list against a second predetermined number.

40. A device according to claim 39, further comprising a process control system in communication with the single computer.

41. A device according to claim 39, further comprising a marking system for marking identified defects on the continuously moving web.

42. A device according to claim 39, wherein said imaging device is a line scan camera.

43. A device according to claim 39, wherein said imaging device utilizes optical assemblies which utilize reflected light, transmitted light or transflected light.

44. A device according to claim 39, wherein multiple imaging devices are utilized.

45. A method of inspecting a flexible circuit web, comprising analyzing blobs formed from a continuous digital data stream of at least 10 mega-pixels/second imaged from at least a portion of a flexible circuit web to identify defects on the flexible circuit web, wherein the blobs are formed and analyzed in a single computer, and further wherein the digital data stream describes pixels in an X domain corresponding to their position accross the web, the method further comprising:

a) binarizing said digital data stream; and

b) forming the blobs from the digital data stream by determining collections of pixels connected to each other in the X domain so as to define segments, and resolving line to line whether connections exist between segments in a Y domain corresponding to the direction of web movement, wherein the determining and the resolving are accomplished in a single iteration.

46. A method comprising:

imaging a continuously moving web to provide a digital data stream, wherein the digital data stream describes pixels in an X domain corresponding to their position across the web;

forming a data structure from the data stream, wherein the data structure includes a set of objects, each object describing a set of pixels within the digital data stream that each have binary values that satisfy a connection threshold; and

analyzing the objects of the data structure to identify defects within the web;

wherein forming the data structure comprises;

determining sets of pixels that satisfy a pixel connection threshold in the X domain so as to define segments;

resolving line to line whether connections exist between segments in a Y domain corresponding to the direction of web movement; and

storing information within one of the objects of the data structure to describe the sets of pixels upon resolving the connections.

47. A method according to claim 46, wherein the objects store data that describe at least a start position and an end position for the corresponding set of connected pixels within the digital data stream.