Machine vision systems and methods for morphological transformation of an image with zero or other uniform offsets

• US 6,236,769 B1
• Filed: 01/28/1998
• Issued: 05/22/2001
• Est. Priority Date: 01/28/1998
• Status: Expired due to Fees

• Current Assignee: Cognex Corporation
• Original Assignee: Cognex Corporation
• Inventors: Desai, Ranjit P.
• Primary Examiner(s): COUSO, YON JUNG
• Application Number:
  US09/014,773
• Time in Patent Office: 1,210 Days
• Field of Search: 382/308, 382/257, 382/274, 382/273, 382/205, 382/153, 382/256, 382/258, 382/259, 382/260, 382/261, 382/265
• US Class Current: 382/308
• CPC Class Codes: G06T 5/30 Erosion or dilatation, e.g. thinning

Abstract

The invention provides machine vision systems for morphological transformation of a source image e.g., adapted for use with zero or uniform offsets. The systems have application, for example, in image erosion or dilation of images of varied image quality. Systems according to the invention compare each pixel in a first row (or line) of the image with a corresponding pixel in a second row of the image. In each comparison, the system identifies the pixel intensity of selected rank. Where the system effects a dilation-type transformation, for example, the comparison seeks a maximum of the compared pixels. For an erosion-type transformation, the comparison seeks a minimum. The pixel intensity value of selected rank (e.g., minimum or maximum) determined from each comparison is retained and, in turn, compared with the intensity values of the corresponding pixels in every other row in the image. These maximum/minimum values are, in turn, compared with the each other and, more particularly, with the values of selected rank (e.g., maximum or minimum) for the neighboring columns. The intensity value of selected rank from among each set of compared values is retained in the transformation image, e.g., in the row and column position corresponding to the center of neighborhood.

Claims

1. A machine vision method for morphological transformation of a source image, the method comprising the steps of

A. for each of selected columns of pixels in the source image, determining and retaining a value of selected rank as between an intensity of a pixel in that column in a first row of that image and intensities of pixels in that column in one or more neighboring rows of that image, so as to determine a “

column winner”

for each selected column,

B. for each of the selected columns in the source image, determining a value of selected rank as between the column winner determined in step (A) for that column and the column winners retained in step (A) for one or more columns-neighboring that column,

C. retaining in a transformation image at least selected values of selected rank determined in step (B).

2. A method according to claim 1, adapted for dilation of the source image, where the selected rank in steps (A) and (B) is a maximum.

3. A method according to claim 1, adapted for erosion of the source image, where the selected rank in steps (A) and (B) is a minimum.

4. A method according to claim 1, wherein step (A) comprises

(i) loading the selected columns of pixels of the first row of the source image into corresponding columns in a first data store,

(ii) loading the selected columns of pixels of a second row of pixels of the source image into corresponding columns in a second data store,

(iii) for each of the selected columns, determining a value of selected rank by comparing with one another intensities of pixels in that column in the first and second data stores, and loading the value of selected rank back into that column of the first data store.

5. A method according to claim 4, wherein step (A) further comprises

(i) loading the selected columns of pixels of each of one or more additional rows of the source image into corresponding columns in the second data store,

(ii) for each of the selected columns, determining a value of selected rank by comparing with one another intensities of pixels in that column in the first and second data stores, and loading the value of selected rank back into that column of the first data store.

6. A method according to claim 5, wherein step (B) comprises, for each of the selected columns, determining the value of selected rank by comparing the value loaded into that column of the first data store in step 5(ii) with the values loaded into adjacent columns of that data store in that same step.

7. A method according to claim 1, comprising the step of adding an offset to a value of selected rank prior to retaining it in step (C).

8. A method according to any of claims 1 and 4, comprising the step of executing steps (A)-(C) to determine and retain in the transformed image values of selected rank for the selected columns of a first set of rows in the source image, while substantially concurrently executing steps (A)-(C) to determine and retain in the transformed image values of selected rank for those columns of a second set of rows in the source image, wherein the first and second sets may include one or more rows in common.

9. A method according to claim 8, adapted for morphological transformation of a source image having at least six rows of pixels, where the transformation uses a neighborhood having a height of 3 rows, the method comprising the steps of

(i) executing steps (A)-(C) to retain in the transformed image values of selected rank from the selected columns of second, third and fourth rows of the source image, while substantially concurrently executing steps (A)-(C) to retain in the transformed image values of selected rank from the selected columns of third, fourth, and fifth rows of the source image,

(ii) executing steps (A)-(C) to retain in the transformed image values of selected rank from the selected columns of first, second, and third rows of the source image, while substantially concurrently executing steps (A)-(C) to retain in the transformed image values of selected rank from the selected columns of fourth, fifth and sixth rows of the source image.

24. A digital data processing system according to claim 8, adapted for morphological transformation of a source image having at least six rows of pixels, where the transformation uses a neighborhood having a height of 3 rows, the method comprising the steps of

(i) executing steps (A)-(C) to retain in the transformed image values of selected rank from the selected columns of second, third and fourth rows of the source image, while substantially concurrently executing steps (A)-(C) to retain in the transformed image values of selected rank from the selected columns of third, fourth, and fifth rows of the source image,

(ii) executing steps (A)-(C) to retain in the transformed image values of selected rank from the selected columns of first, second, and third rows of the source image, while substantially concurrently executing steps (A)-(C) to retain in the transformed image values of selected rank from the selected columns of fourth, fifth and sixth rows of the source image.

10. An article of manufacture comprising a computer usable medium embodying program code for causing a digital data processor to carry out a method for morphological transformation of a source image, the method comprising the steps of

A. for each of selected columns of pixels in the source image, determining and retaining a value of selected rank as between an intensity of a pixel in that column in a first row of that image and intensities of pixels in that column in one or more neighboring rows of that image, so as to determine a “

column winner”

for each selected column,

B. for each of the selected columns in the source image, determining a value of selected rank as between the column winner determined in step (A) for that column and the column winners retained in step (A) for one or more columns neighboring that column,

C. retaining in a transformation image at least selected values of selected rank determined in step (B).

11. An article according to claim 10, wherein the method is adapted for dilation of the source image, where the selected rank is a maximum.

12. An article according to claim 10, wherein the method is adapted for erosion of the source image, where the selected rank is a minimum.

13. An article according to claim 10, wherein step (A) comprises

(i) loading the selected columns of pixels of the first row of the source image into corresponding columns in a first data store,

(ii) loading the selected columns of pixels of a second row of pixels of the source image into corresponding columns in a second data store,

(iii) for each of the selected columns, determining a value of selected rank by comparing with one another intensities of pixels in that column in the first and second data stores, and loading the value of selected rank back into that column of the first data store.

14. An article according to claim 10, wherein the method comprises the step of adding an offset to a value of selected rank prior to retaining it in step (C).

15. An article according to claim 10, wherein the method comprises the step of executing steps (A)-(C) to determine and retain in the transformed image values of selected rank for respective ones of at least selected columns of a first set of rows in the source image, while substantially concurrently executing steps (A)-(C) to determine and retain in the transformed image values of selected rank for those columns of a second set of rows in the source image, wherein the first and second sets may include one or more rows in common.

16. A digital data processing system carrying out a method for morphological transformation of a source image, the method comprising the steps of

A. for each of selected columns of pixels in the source image, determining and retaining a value of selected rank as between an intensity of a pixel in that column in a first row of that image and intensities of pixels in that column in one or more neighboring rows of that image, so as to determine a “

column winner”

for each selected column,

B. for each of the selected columns in the source image, determining a value of selected rank as between the column winner determined in step (A) for that column and the column winners retained in step (A) for one or more columns neighboring that column,

C. retaining in a transformation image at least selected values of selected rank determined in step (B).

17. A digital data processing system according to claim 16, adapted for dilation of the source image, where the selected rank in steps (A) and (B) is a maximum.

18. A digital data processing system according to claim 16, adapted for erosion of the source image, where the selected rank in steps (A) and (B) is a minimum.

19. A digital data processing system according to claim 16, wherein step (A) comprises

(i) loading the selected columns of pixels of the first row of the source image into corresponding columns in a first data store,

(ii) loading the selected columns of pixels of a second row of pixels of the source image into corresponding columns in a second data store,

(iii) for each of the selected columns, determining a value of selected rank by comparing an intensity of a pixel in that column in the first data store with an intensity of a pixel in that column in the second data store, and loading the value of selected rank back into that column of the first data store.

20. A digital data processing system according to claim 19, wherein step (A) further comprises

(i) loading at least selected columns of pixels of one or more additional rows of the source image into corresponding columns in the second data store,

(ii) for each of the selected columns, determining a value of selected rank by comparing an intensity of a pixel in that column in the first data store with an intensity of a pixel in that column in the second data store, and loading the value of selected rank back into that column of the first data store.

21. A digital data processing system according to claim 20, wherein step (B) comprises, for each of the selected columns, determining the value of selected rank by comparing the value loaded into that column of the first data store in step 5(ii) with the values loaded into adjacent columns of that data store in that same step.

22. A digital data processing system according to claim 16, comprising the step of adding an offset to a value of selected rank prior to retaining it in step (C).

23. A digital data processing system according to any of claims 16 and 19, comprising the step of executing steps (A)-(C) to determine and retain in the transformed image values of selected rank for the selected columns of a first set of rows in the source image, while substantially concurrently executing steps (A)-(C) to determine and retain in the transformed image values of selected rank for those columns of a second set of rows in the source image, wherein the first and second sets may include one or more rows in common.