PC board inspecting apparatus, inspection logic setting method, and inspection logic setting apparatus

• US 20060153439A1
• Filed: 01/11/2006
• Published: 07/13/2006
• Est. Priority Date: 01/11/2005
• Status: Abandoned Application

• Current Assignee: Omron Corporation
• Original Assignee: Omron Corporation
• Inventors: Shimizu, Atsushi, Nakajima, Akira, Moriya, Toshihiro, Onishi, Takako, Wada, Hirotaka
• Application Number:
  US11/329,432
• Publication Number:
  US 20060153439A1
• Time in Patent Office: Days
• US Class Current: 382/147
• CPC Class Codes: G01N 21/95684 Patterns showing highly reflecting parts, e.g. metallic elements
• CPC Class Codes: G01R 31/71 Testing of solder joints

Abstract

A technology is provided for automatically producing an inspection logic to be used in the action of inspecting a PC board. An inspection logic setting apparatus is arranged for acquiring a plurality of first images of components to be detected by the inspection and a plurality of second images of components to be rejected by the inspection, dividing each first and second images into the plurality of blocks, calculating a color distance between the plurality of first images and the plurality of second images in each block, selecting one or more blocks from the plurality of blocks which are relatively greater in the color distance, and assigning the selected block(s) to an area condition.

Claims

1. An inspection logic setting method comprising:

acquiring a plurality of first images of components to be detected by the inspection and a plurality of second images of components to be rejected by the inspection,

dividing each of first and second images into a plurality of blocks,

calculating color distances between the first and second images in each block,

selecting one or more blocks from the plurality of blocks which are relatively greater in color distance, and

assigning the selected block(s) to an area condition.

2. An inspection logic setting method according to claim 1, further comprising:

extracting from each first image an object image determined by an area condition,

extracting from each second image a reject image determined by said area condition,

calculating a frequency distribution of one or more types of color feature values at each of the plurality of object points in the plurality of object images where each pixel in an object image represents an object point,

calculating a frequency distribution of said types of color feature values at each of the plurality of reject points in the plurality of reject images where each pixel in a reject image represents a reject point,

calculating based on the two frequency distributions ranges of said color feature values where a separation between values at each object point and values at each reject point is maximum, and

assigning the types of color feature values and the ranges of said types to a color condition.

3. An inspection logic setting method according to claim 2, wherein

the types of color feature values used in a color condition are predetermined.

4. An inspection logic setting method according to claim 2, further comprising:

making a plurality of candidates for a color condition through combining one or more types of color feature values,

calculating for each candidate ranges of said color feature values and a degree of separation where a separation between the object points and the reject points is maximum,

selecting a candidate through comparing the degrees of separation, and

assigning the types of color feature values constituting the selected candidate and the ranges of said types to a color condition.

5. An inspection logic setting method according to claim 2 further comprising:

picking up from each object image a pixel area which satisfies a color condition,

picking up from each reject image a pixel area which satisfies said color condition,

calculating a frequency distribution of one or more types of feature values in each of the pixel areas picked up from object images,

calculating a frequency distribution of said types of feature values in each of the pixel areas picked up from reject images,

calculating based on the two frequency distributions ranges of said feature values where a separation between the values in each pixel area picked up from object image and the values in each pixel area picked up from reject image is maximum, and

assigning the types of feature values and the ranges of said types to a judgment condition.

6. An inspection logic setting method according to claim 5, wherein

the types of the feature values used in a judgement condition are predetermined.

7. An inspection logic setting method according to claim 5, further comprising:

making a plurality of candidates for judgement condition through combining one or more types of feature values,

calculating for each candidate ranges of said feature values and a degree of separation where a separation between the feature values in each pixel area picked up from the object images and the feature values in each pixel area picked up from reject images is maximum,

selecting a candidate through comparing the degrees of separation, and

assigning the types of feature values constituting the selected candidate and the ranges of said types to a judgment condition.

8. An inspection logic setting method according to claim 5, further comprising:

producing a plurality of area conditions of different areas from the selected blocks, and

calculating and determining a color condition and a judgment condition in relation to each of the area conditions.

9. An inspection logic setting method according to claim 8, further comprising:

selecting two or more of the blocks which are relatively greater in the color distance,

grouping the selected plurality of blocks on the color distance, and

assigning the groups to area conditions.

10. An inspection logic setting method according to claim 5, wherein

a plurality of combinations of a color condition and a judgment condition determined in relation to an area condiion.

11. An inspection logic setting method according to claim 5, further comprising:

determining two or more combinations of area conditions and the related color conditions and judgement conditions to serve as an inspection logic for a PC board inspecting apparatus conducting inspection.

12. An inspection logic setting method according to claim 1, wherein

an inspection logic is determined for each type of components to be inspected.

13. An inspection logic setting method according to claim 1, wherein

an inspection logic is determined for each type of error.

42. A PC board inspecting apparatus comprising:

a storage for saving area conditions, color conditions and judgment conditions which have been determined by the inspection logic setting method defined in claim 4,

a projecting device for projecting different colors of light at different incident angles onto a surface mounted component on a board,

an area extracting device for extracting an inspection area determined by an area condition from an image which has been produced from the reflection of color light and picking up an area in said inspection area which satisfies an color condition, and

a judging device for examining whether said picked up area satisfies the judgment condition for judging the surface mounted state of components.

43. A PC board inspecting apparatus comprising:

a storage for saving different combinations of an area condition and the associated color conditions and judgment conditions, which have been determined by the inspection logic setting method defined in claim 11, and logical operations for calculating said combinations,

a projecting device for projecting different colors of light at different incident angles onto a surface mounted component on a board,

an area extracting device for extracting inspection areas determined by an area conditions in each combination from an image which has been produced from the reflection of color light and picking up an area in each of said inspection areas which satisfies the corresponding color condition, and

a judging device for examining and judging separately whether each picked up area in said inspection area satisfies the corresponding judgment condition and determining a final judgment for surface mounted state of components by assigning these individual judgment results to said logical operations.

14. An inspection logic setting apparatus comprising:

an image acquiring device for acquiring a plurality of first images of components to be detected by an inspection and a plurality of second images of components to be rejected by an inspection,

a color distance calculating device for dividing each of the first and second images into a plurality of blocks and calculating a color distance between first and second images in each block, and

an area condition determining device for selecting one or more blocks from the plurality of blocks which are relatively greater in color distance and assigning the selected block(s) to an area condition.

15. An inspection logic setting apparatus according to claim 14, further comprising:

a device for extracting from each first image an object image determined by an area condition,

a device for extracting from each second image a reject image determined by said area condition,

a device for calculating a frequency distribution of one or more types of color feature values at each of the plurality of object points in the plurality of object images where each pixel in an object image represents an object point,

a device for calculating a frequency distribution of said types of color feature values at each of the plurality of reject points in the plurality of reject images where each pixel in a reject image represents a reject point,

a device for calculating based on the two frequency distributions ranges of said color feature values where a separation between the values at each object point and values at each reject point is maximum, and

a device for assigning the types of color feature values and the ranges of said types to a color condition.

16. An inspection logic setting apparatus according to claim 15, wherein

types of color feature values used in a color condition are predetermined.

17. An inspection logic setting apparatus according to claim 15, further comprising:

a device for making a plurality of candidates for color condition through combining one or more types of color feature values,

a device for calculating for each candidate ranges of said types of color feature values and degrees of separation where a separation between the object points and the reject points is maximum,

a device for selecting a candidate through comparing the degrees of separation, and

a device for assigning the types of color feature values constituting the selected candidate and the ranges of said types to a color condition.

18. An inspection logic setting apparatus according to claim 15, further comprising:

a device for picking up from each object image a pixel area which satisfies a color condition,

a device for picking up from each reject image a pixel area which satisfies said color condition,

a device for calculating a frequency distribution of one or more types of feature values in each pixel area picked up from object images,

a device for calculating a frequency distribution of said types of feature values in each pixel area picked up from reject images,

a device for calculating based on the two frequency distributions ranges of said feature values where a separation between the values in each pixel area picked up from object images and the values in each pixel area picked up from reject images is maximum, and

a device for assigning the types of feature values and the ranges of said types to a judgment condition.

19. An inspection logic setting apparatus according to claim 18, wherein

the types of the feature values used in a judgement condition are predetermined.

20. An inspection logic setting apparatus according to claim 18, further comprising:

a device for making a plurality of candidates for judgement condition through combining one or more types of feature values,

a device for calculating for each candidate ranges of said feature values and a degree of separation where a separation between the values in each pixel area picked up from object images and the values in each pixel area picked up from reject images is maximum,

a device for selecting a candidate through comparing degrees of separation, and

a device for assigning the types of feature values constituting the selected candidate and the ranges of said types to a judgment condition.

21. An inspection logic setting apparatus according to claim 18, wherein

an area condition determining device produces a plurality of area conditions of different area from the selected blocks, and

a device for assigning assigns a color condition and a judgment condition in relation to each area condition.

22. An inspection logic setting apparatus according to claim 21, wherein

the area condition determining device selects two or more of the blocks which are relatively greater in color distance, and groups the selected blocks based on the color distance to determine a plurality of area conditions.

23. An inspection logic setting apparatus according to claim 18, wherein

a device for assigning assigns a plurality of combinations of a color condition and a judgment condition in relation to an area condition.

24. An inspection logic setting apparatus according to claim 18, wherein

the apparatus combines two or more area conditions, and the related color conditions and judgment conditions, and serve the combination as an inspection logic.

25. An inspection logic setting apparatus according to claim 14, wherein

the apparatus provides inspection logics for each type of components to be inspected.

26. An inspection logic setting apparatus according to claim 14, wherein

the apparatus provides inspection logics for each type of error.

27. A program for generating an inspection logic comprising the steps of instructing a data processing apparatus arranged to conduct:

acquiring a plurality of images of components to be detected by the inspection as first images,

acquiring a plurality of images of components to be rejected by the inspection as second images,

dividing each of first and second images into a plurality of blocks,

calculating a color distance between the plurality of first images and the plurality of second images in each block,

selecting one or more blocks from the plurality of blocks which are relatively greater in the color distance, and

assigning the selected block(s) to an area condition.

28. A program according to claim 27, further comprising the steps of:

extracting from each first image an object image determined by an area condition,

extracting from each second image a reject image determined by said area condition,

calculating a frequency distribution of one or more types of color feature values at each of a plurality of object point in the plurality of object image where each pixel in an object image represents an object point,

calculating a frequency distribution of said types of color feature values at each of a plurality of reject point in the plurality of reject image where each pixel in a reject image represents a reject point,

calculating based on the two frequency distributions ranges of said color feature values where a separation between the values at each object point and the values at each reject point is maximum, and

assigning the types of color feature values and the ranges of said types to a color condition.

29. A program according to claim 28, wherein

the types of color feature values used in a color condition are predetermined.

30. A program according to claim 27, further comprising the steps of:

making a plurality of candidates for a color condition through combining one or more types of color feature values,

calculating for each candidate ranges of said color feature values and a degree of separation where a separation between the object points and the reject points is maximum,

selecting a candidate through comparing the degrees of separation, and

assigning the types of color feature values constituting the selected candidate and the ranges of said types to a color condition.

31. A program according to claim 28, further comprising the steps of:

picking up from each object image a pixel area which satisfies a color condition,

picking up from each reject image a pixel area which satisfies said color condition,

calculating a frequency distribution of one or more types of feature values in each of the pixel areas picked up from object images,

calculating a frequency distribution of said types of feature values in each of the pixel areas picked up from reject images,

calculating based on the two frequency distributions ranges of said feature values where a separation between the values in each pixel area picked up from the object image and the values in each pixel area picked up from the reject image is maximum, and

assigning the types of feature values and the ranges of said types to a judgment condition.

32. A program according to claim 31, wherein

the types of feature values used in a judgement condition are predetermined.

33. A program according to claim 31, further comprising the steps of:

making a plurality of candidates for a judgement condition through combining one or more types of feature value,

calculating for each candidate ranges of said feature values and a degree of separation where a separation between the feature values in each pixel area picked up from the object images and the feature values in each pixel area picked up from reject images is maximum,

selecting a candidate through comparing the degree of separations, and

assigning the types of feature values constituting the selected candidate and the ranges of said types to a judgment condition.

34. A program according to claim 31, further comprising the steps of:

providing a plurality of area conditions of different areas from the selected blocks, and

calculating and assigning color conditions and judgment conditions in relation to each of the area conditions.

35. A program according to claim 34, further comprising the steps of:

selecting two or more of the blocks which are relatively greater in color distance,

grouping the selected plurality of blocks on the color distance, and

assigning the groups to area conditions.

36. A program according to claim 31, wherein

a plurality of combinations of color conditions and judgment conditions are calculated and determined in relation to an area condition in the assigning step.

37. A program according to claim 31, further comprising the steps of:

determining a logical combination of two or more area conditions and the related color conditions and judgement conditions, and

storing the logical combination as an inspection logic for the PC board inspecting apparatus conducting an inspection.

38. A program according to claim 27, wherein

the program determines an inspection logic for each type of components to be inspected.

39. A program according to claim 27, wherein

the program determines an inspection logic for each type of error.

40. A storage medium for saving a program defined in claim 27.

41. A PC board inspecting apparatus comprising:

a device for determing an area condition which determines inspection area and certain color and judgement conditions associated with said area conditions,

a storage device for saving area conditions, color conditions and judgement conditions,

a projecting device for projecting different colors of light at different incident angles onto a surface mounted on a component board,

an image acquiring device for acquiring an image which has been produced from the reflection of color light,

an area extracting device for extracting said inspection area determined by an area condition from an image provided by the image aquiring means,

a pickup device for picking up the area which satisfies the color condition in said inspection area, and

a judging device for examining whether said picked up area satisfies the judgement condition for judging the surface mounted state of components.