Inspection system and methods with autocompensation for edge break gauging orientation

• US 7,925,075 B2
• Filed: 05/07/2007
• Issued: 04/12/2011
• Est. Priority Date: 05/07/2007
• Status: Expired due to Fees

• Current Assignee: General Electric Company
• Original Assignee: General Electric Company
• Inventors: Jia, Ming, Song, Guiju, Ning, Yu, Ross, Joseph Benjamin, Harding, Kevin George, Isaacs, Ralph Gerald, Zheng, Jianming, Abramovich, Gil
• Primary Examiner(s): Mehta; Bhavesh M
• Assistant Examiner(s): Thomas; Mia M
• Application Number:
  US11/745,010
• Publication Number:
  US 20080281543A1
• Time in Patent Office: 1,436 Days
• Field of Search: 382/145, 382/148, 382/149, 382/199, 702/150
• US Class Current: 382/149
• CPC Class Codes: G01B 11/25 by projecting a pattern, e.g. one or more lines, moiré fringes on the object G01B11/255 takes precedence ; image analysis for depth or shape recovery G06T7/50
• CPC Class Codes: G01B 21/20 for measuring contours or curvatures, e.g. determining profile

Abstract

A method for inspecting a feature of a part is provided. The method includes obtaining a profile corresponding to the feature using a sensor and projecting the profile onto a compensation plane normal to the feature for generating an updated profile. The method also includes using the updated profile for reducing a measurement error caused by an orientation of the sensor. An inspection system is also provided. The inspection system includes a sensor configured to capture a fringe image of a feature on a part. The inspection system further includes a processor configured to process the fringe image to obtain an initial profile of the feature and to project the initial profile onto a compensation plane normal to the feature.

Claims

1. A method for inspecting a feature of a part, comprising:

capturing an image corresponding to the feature using a sensor;

extracting at least one profile from the image using a processor;

removing a plurality of data points corresponding to noise from the at least one profile to generate at least one trimmed profile using the processor;

extracting a plurality of data points corresponding to edges of the at least one trimmed profile using the processor;

fitting two edge surface planes through the plurality of data points corresponding to edges of the at least one trimmed profile using the processor;

generating a compensation plane normal to the two edge surface planes using the processor; and

projecting the at least one trimmed profile onto the compensation plane to generate an updated profile using the processor,

whereby the updated profile is used to reduce a measurement error caused by an orientation of the sensor.

2. The method of claim 1, wherein the step of removing a plurality of data points corresponding to noise from the at least one profile is performed by using one of a line fitting method and a Graham Scan method.

3. The method of claim 1, wherein the updated profile is utilized to determine a quality of the feature.

4. The method of claim 1, wherein the feature comprises an edge break in the part.

5. The method of claim 4, wherein the edge break comprises a chamfer, a bevel, a rounded edge, or a fillet.

6. A method for inspecting an edge break of a part, comprising:

obtaining an initial profile corresponding to the edge break of the part using a sensor having a plurality of different angular values; and

determining an orientation of the edge break of the part such that a cross-section of data points along the initial profile is being measured normal to a major axis of the edge break of the part, thereby reducing a measurement error due to an orientation of the sensor.

7. The method of claim 6, wherein the step of determining an orientation of the edge break of the part such that a cross-section of data points along the initial profile is being measured normal to a major axis of the edge break of the part comprises projecting the initial profile along a normal direction.

8. The method of claim 7, further comprising obtaining a plurality of profiles comprising a plurality of edges and removing a plurality of data points in the edges of the profiles corresponding to noise by performing one of a line fitting method and a Graham Scan method.

9. The method of claim 8, further comprising extracting at least ten profiles from a captured image of the initial profile obtained using the sensor.

10. The method of claim 9, further comprising extracting a plurality of data points from the at least ten profiles and fitting two edge planes through the plurality of data points.

11. The method of claim 6, further comprising determining a normal vector and projecting the initial profile along the normal vector.

12. An inspection system for inspecting a feature of a part, comprising:

a sensor configured to capture an image of the feature on the part; and

a processor configured to process the image of the feature of the part, extract at least one profile from the image of the feature of the part, remove a plurality of data points corresponding to noise from the at least one profile to generate at least one trimmed profile, extract a plurality of data points corresponding to edges of the at least one trimmed profile, fit two edge surface planes through the plurality of data points corresponding to edges of the at least one trimmed profile, generate a compensation plane normal to the two edge surface planes, and project the at least one trimmed profile onto the compensation plane to generate an updated profile, wherein the updated profile is used by the processor to reduce a measurement error caused by an orientation of the sensor.

13. The inspection system of claim 12, wherein the feature comprises an edge break on the part.

14. The inspection system of claim 13, wherein the edge break comprises a chamfer, a bevel, a rounded edge, or a fillet.

15. The inspection system of claim 12, wherein the processor is configured to remove the plurality of data points corresponding to noise from the at least one profile by performing one of a line fitting method and a Graham Scan method.