Structure-guided automatic alignment for image processing

• US 20030026473A1
• Filed: 06/13/2001
• Published: 02/06/2003
• Est. Priority Date: 06/13/2001
• Status: Active Grant

• Current Assignee: Leica Microsystems CMS GmbH (Danaher Corporation)
• Original Assignee: Shih-Jong J. Lee
• Inventors: Oh, Seho, Lee, Shih-Jong J.
• Granted Patent: US 6,829,382 B2
• Time in Patent Office: Days
• US Class Current: 382/152
• CPC Class Codes: G06T 2207/30164 Workpiece; Machine component
• CPC Class Codes: G06T 7/0002 Inspection of images, e.g. flaw detection
• CPC Class Codes: G06T 7/12 Edge-based segmentation
• CPC Class Codes: G06T 7/33 using feature-based methods

Abstract

When application domain structure information is erroneously encoded into parameters for image processing and measurements the accuracy of the result can degrade. A structure-guided automatic alignment system for image processing receives an image input and application domain structure input and automatically creates an estimated structure output having improved alignment. Measurement and image processing robustness are improved.

Claims

1. A structure-guided automatic alignment system for image processing comprising:

(a) means to receive an image input;

(b) means to receive an application domain structure input;

(c) a structure estimation module having an estimated structure output processes the image input and the application domain structure input.

2. The system of claim 1 further comprises an alignment decision module that receives the estimated structure and the application domain structure inputs and having an alignment parameter output.

3. The system of claim 1 further comprises a structure alignment module that receives the alignment parameter input and the application domain structure input and having a structure alignment output.

4. The system of claim 1 wherein the application domain structure is specified using a caliper method.

5. The caliper method of claim 4 further comprises a directional box caliper.

6. The caliper method of claim 4 further comprises a circle caliper.

7. The caliper method of claim 4 further comprises an arc caliper.

8. The system of claim 1 wherein the structure estimation module comprises

(a) a preprocessing module that receives the image input having a pre-processed image output;

(b) a distributed estimation module that receives the pre-processed image and application domain structure and having an estimated structure output.

9. The structure estimation module of claim 8 wherein the preprocessing module performs edge detection.

10. The structure estimation module of claim 8 wherein the distributed estimation module further comprises

(a) a sub-region generation module having a sub-regions output;

(b) a robust structure-guided estimation module that receives a sub-regions output having an estimated structure output.

11. A structure estimation module comprising

(a) means to receive an image input;

(b) means to receive an application domain structure input;

(c) a preprocessing module that receives the image input having a pre-processed image output;

(d) a distributed estimation module that receives the pre-processed image and application domain structure inputs having an estimated structure output.

12. The structure estimation module of claim 11 wherein the preprocessing module performs a feature extraction such as edge detection.

13. The structure estimation module of claim 11 wherein the distributed estimation module comprises

(c) a sub-region generation module having a sub-region output;

(d) a robust structure-guided estimation module that receives the sub-region output having a feature parameter output.

14. The distributed estimation module of claim 13 wherein the robust structure-guided estimation module receives a box caliper input.

15. The distributed estimation module of claim 13 wherein the robust structure-guided estimation module receives a circle caliper input.

16. The distributed estimation module of claim 13 wherein the robust structure-guided estimation module receives an arc caliper input

17. The box caliper robust structure-guided estimation module of claim 14 comprises:

(a) means to perform row-wise robust feature direction estimation and weight adjustment having an adjusted weight output;

(b) means to perform column-wise robust feature direction estimation and weight adjustment having an adjusted weight output;

(c) means to perform overall robust feature direction estimation and weight adjustment having a feature direction estimation result.

18. The circle caliper robust structure-guided estimation module of claim 15 comprises:

(a) means to perform radial-wise robust center estimation and weight adjustment having adjusted weight output;

(b) means to perform angular-wise robust center estimation and weight adjustment having adjusted weight output;

(c) means to perform overall robust center estimation and weight adjustment having center estimation output.

19. The arc caliper robust structure-guided estimation module of claim 16 comprises:

(d) means to perform radial-wise robust center estimation and weight adjustment having adjusted weight output;

(e) means to perform angular-wise robust center estimation and weight adjustment having adjusted weight output;

(f) means to perform overall robust center estimation and weight adjustment having center estimation output.

20. A robust feature direction estimation and weight adjustment method for a group of box caliper sub-regions comprises:

(a) for each sub-region, estimate the feature direction using line estimation without constraints;

(b) for a group of sub-regions, estimate the feature direction using line estimation constrained by a parallelism relation;

(c) compare the sub-region feature direction estimated in step (a), with the group direction estimated in step (b) and adjust weight for the sub-region based on an error function.

(d) update and output the group of sub-regions box caliper estimate of the feature direction

23. The method of claim 20 further comprising a step for re-alignment of the box caliper direction vector responsive to the group of sub-regions estimate of the feature direction output.

21. A robust feature direction estimation and weight adjustment method for a group of circle caliper sub-regions comprises:

(a) for each sub-region, estimate the center of a circular arc without constraints;

(b) for a group of sub-regions, estimate the center of circular arcs constrained by the same center point;

(c) compare the sub-region center estimated in step (a), with the group center estimated in step (b) and adjust the weight for the sub-region based on an error function.

(d) update and output the group of sub-regions center

24. The method of claim 21 further comprising a step for re-alignment of the circle caliper center location responsive to the output of the group of sub-regions estimate of the center of circular arcs output.

22. A robust feature direction estimation and weight adjustment method for a group of arc caliper sub-regions comprises the following steps:

(a) for each sub-region, estimate the center of a circular arc without constraints;

(b) for a group of sub-regions, estimate the center of circular arcs constrained by the same center point;

(c) compare the sub-region center estimated in step (a), with the group center estimated in step (b) and adjust weight for the sub-region based on an error function;

(d) update and output the group of sub-regions estimate for the center of circular arcs.

25. The method of claim 22 further comprising a step for re-alignment of the arc caliper center location responsive to the output of the group of sub-regions estimate of the center of circular arcs output.