Process control using multiple detections

• US 6,266,436 B1
• Filed: 04/09/1999
• Issued: 07/24/2001
• Est. Priority Date: 04/09/1999
• Status: Active Grant

• Current Assignee: Kimberly-Clark Worldwide Incorporated (Kimberly-Clark Corporation)
• Original Assignee: Kimberly-Clark Worldwide Incorporated (Kimberly-Clark Corporation)
• Inventors: Giza, Robert Jeffrey, Bernhardt, Wayne Allen, Ungpiyakul, Tanakon, Bett, Thomas Arthur, Lemery, Shawn Timothy
• Primary Examiner(s): NOLAND, THOMAS
• Application Number:
  US09/289,777
• Time in Patent Office: 837 Days
• Field of Search: 382/141, 382/143, 382/111, 702/81, 702/36, 702/94, 702/82, 702/83, 702/84, 702/85, 702/123, 702/189, 702/182, 702/199, 700/28, 700/29, 700/30, 700/31, 700/32, 700/33, 700/51, 700/302, 700/90, 700/108, 700/109, 700/110, 700/117, 700/130, 700/143, 700/144
• US Class Current: 382/141
• CPC Class Codes: G06T 2207/30124 Fabrics; Textile; Paper
• CPC Class Codes: G06T 7/70 Determining position or orientation of objects or cameras camera calibration G06T7/80

Abstract

Controlling processes comprising detecting and measuring a parameter, for example presence and location of an element of a good, with at least two determinations as representations of the target parameter, transmitting signals to the computer, and processing the signals to compare the parameter to acceptable conditions. The detection can include three or more replications, optionally each for at least two parameters, optionally using at least two different methods to analyze the signals. The invention contemplates detecting and analyzing the target parameters using two or more analytical tools within the respective image to detect a given component of the product, namely two or more measurements of the parameter on a single visual image. Analytical methods can include averaging the signals, determining the number of signals of common signal duration and/or signal characteristics, computing standard deviation, modifying the signal combination to compensate for an inappropriate signal, and/or comparing the signals to a database of signal combinations. The method can automatically compute probable cause of some anomalies in the signals, develop corresponding responses, and transmit responses to process control, and thence to control devices. The methods can automatically recalibrate determinors, or automatically adjust analysis to a basis of one less determinor, and/or automatically implement back-up inspection of goods, optionally saving images for further analysis, or culling units of product. Digitized visual images represent pixels and pixel combinations. The method contemplates analyzing the pixel representations with at least two determinations of the parameter in respective at least two areas of the image, optionally for at least two parameters at respective replication sites, using software interpretation of selected areas of the visual image.

Claims

1. A method of measuring a parameter of goods being fabricated in a manufacturing operation, the method comprising:

(a) establishing a target parameter to be measured on respective units of the goods, and acceptable conditions of the target parameter;

(b) capturing a full digitized visual image of a unit of the goods being fabricated, the digitized visual image representing pixels and pixel combinations in the visual image;

(c) in the captured full digitized visual image, analyzing the digital pixel combination representations in at least first and second areas of the image, which respective areas of the image are specified to indicate, collectively and in combination, a common acceptable condition of the target parameter, and thereby generating respective first and second replicate determination signals representative of the target parameter;

(d) analyzing the determination signals in combination, for conformity of the established target parameter to the established acceptable conditions utilizing respective analysis methods; and

(e) processing the determination signals such that if a given one of the determination signals is concluded to be inappropriate, modifying the signal combination to thereby compensate for the inappropriate signal.

2. A method as in claim 1, including analyzing pixel combination representations in at least first and second areas of the image and thereby generating respective first and second combination determination signals, for at least first and second parameters.

3. A method as in claim 1, including processing the determination signals so as to use first and second different analytical methods to analyze the determination signals representative of the respective first and second parameters.

4. A method as in claim 1, including analyzing the pixel combination representations with respective at least first, second, and third separate and distinct replications of determinations of the condition of the target parameter in respective at least first, second, and third areas of the image.

5. A method as in claim 4, processing of the determination signals comprising computing an average of the signals.

6. A method as in claim 4, processing of the determination signals comprising computing a standard deviation based on the determination signals.

7. A method as in claim 4 wherein, when analysis detects inappropriate input from one of the selected areas of the image, automatically adjusting the analysis to a basis of analyzing one less area.

8. A method as in claim 1, including analyzing the pixel combination representations in respective at least first, second, and third areas of the image for at least first and second parameters at respective replication sites on the goods.

9. A method as in claim 8, including processing the determination signals from the respective first and second parameters so as to use first and second different analytical methods to analyze the determination signals representative of the respective first and second parameters.

10. A method as in claim 1, processing of the determination signals comprising determining the number of signals of common or nearly common signal characteristics.

11. A method as in claim 1, including comparing the signal combination to a database of known and/or expected signal combinations, and based on the comparison, developing a conclusion as to the probable cause of any anomaly in the signal combination, and developing a corresponding response to the signal combination.

12. A method as in claim 11, including transmitting the response as a control signal to a process controller controlling the manufacturing operation.

13. A method as in claim 1, including comparing the signal combination to a database of known and/or expected signal combinations, including a historical probability of the occurrence of respective ones of the combinations, and based on the comparison, developing a conclusion as to the probable cause of any anomaly in the signal combination, and developing a corresponding response to the signal combination.

14. A method as in claim 1, the multiple analyses of the pixel combination representations comprising respective multiple determinations using software interpretation of selected areas of the full digitized visual image.

15. A method as in claim 1, the method further comprising so analyzing each of the goods being fabricated in the manufacturing operation.

16. A method of measuring the location of an element on an absorbent article being fabricated in a manufacturing operation, the method comprising:

(a) establishing an acceptable location for the element on the absorbent article;

(b) capturing a full digitized visual image of the absorbent article, the full digitized visual image representing pixels and pixel combinations in the visual image;

(c) in the captured full digitized visual image, analyzing the digital pixel combination representations in at least first and second areas of the image, which respective areas of the image are specified to indicate, collectively and in combination, a common acceptable location of the element, and thereby generating respective first and second replicate determination signals representative of the location of the element on the product;

(d) analyzing the determination signals in combination, for conformity of the location of the element to the established acceptable locations utilizing respective analysis methods; and

(e) processing the determination signals such that if a given one of the determination signals is concluded to be inappropriate, modifying the signal combination to thereby compensate for the inappropriate signal.

17. A method as in claim 16, including analyzing pixel combination representations in at least first and second areas of the image and thereby generating respective first and second combination determination signals, for at least the above-recited element location, and for a second parameter.

18. A method as in claim 17, including processing the determination signals so as to use first and second different analytical methods to analyze the determination signals representative of the respective location, and the second parameter.

19. A method as in claim 16, including analyzing the pixel combination representations with respective at least first, second, and third separate and distinct replications of determinations of the location of the element in respective at least first, second, and third areas of the image.

20. A method as in claim 19, processing of the determination signals comprising computing an average of the signals.

21. A method as in claim 19 wherein, when the analysis detects inappropriate input from one of the above areas of the image, automatically adjusting the analysis to a basis of analyzing one less area.

22. A method as in claim 16, including analyzing the pixel combination representations in respective at least first, second, and third areas of the image for at least the above-recited location, and a second parameter, at respective replication sites on the goods.

23. A method as in claim 22, including processing the determination signals from the respective location, and the second parameter, so as to use first and second different analytical methods to analyze the determination signals representative of the respective location, and the second parameter.

24. A method as in claim 16, processing of the determination signals comprising determining the number of signals of common or nearly common signal characteristics.

25. A method as in claim 16, processing of the determination signals comprising computing a standard deviation based on the determination signals.

26. A method as in claim 16, including comparing the signal combination to a database of known and/or expected signal combinations, and based on the comparison, developing a conclusion as to the probable cause of any anomaly in the signal combination, and developing a corresponding response to the signal combination.

27. A method as in claim 26, including transmitting the response as a control signal to a process controller controlling the manufacturing operation.

28. A method as in claim 16, including comparing the signal combination to a database of known and/or expected signal combinations, including a historical probability of the occurrence of respective ones of the combinations in such absorbent articles, and based on the comparison, developing a conclusion as to the probable cause of any anomaly in the signal combination, and developing a corresponding response to the signal combination.

29. A method as in claim 16, the multiple analyses of the pixel combination representations comprising respective multiple determinations using software interpretation of selected areas of the full digitized visual image.

30. A method as in claim 16, the method further comprising so analyzing each of the absorbent articles fabricated in the manufacturing operation.