System and method for detecting defects in a surface of a workpiece

• US 6,266,138 B1
• Filed: 10/12/1999
• Issued: 07/24/2001
• Est. Priority Date: 10/12/1999
• Status: Expired due to Term

• Current Assignee: Perceptron Incorporated (Atlas Copco AB)
• Original Assignee: Perceptron Incorporated (Atlas Copco AB)
• Inventors: Keshavmurthy, Shyam P.
• Primary Examiner(s): Pham, Hoa Q.
• Application Number:
  US09/415,930
• Time in Patent Office: 651 Days
• Field of Search: 356/237.2, 356/237.1, 356/371, 356/445, 356/446, 356/448, 356/394, 382/141, 382/168, 382/257, 382/107, 250/462.1, 250/463.1, 348/128, 348/92, 348/94, 348/125, 348/126
• US Class Current: 356/237.2
• CPC Class Codes: G01B 11/303 using photoelectric detection means
• CPC Class Codes: G01N 21/88 Investigating the presence of flaws or contamination
• CPC Class Codes: G01N 21/9515 Objects of complex shape, e.g. examined with use of a surface follower device measuring contours and curvatures G01B11/24
• CPC Class Codes: G06T 2207/10016 Video; Image sequence
• CPC Class Codes: G06T 2207/30156 Vehicle coating
• CPC Class Codes: G06T 2207/30164 Workpiece; Machine component
• CPC Class Codes: G06T 7/0006 using a design-rule based approach

Abstract

A surface inspection system is provided for detecting defects on a surface of a workpiece. The surface inspection system includes a diffused light source for emitting an elongated line of light onto the surface of the workpiece, a movable member for translating the workpiece in relation to the light source, an imaging device positioned at a vantage point such that the line of light is within its field of observation for capturing two or more sets of image data representative of a portion of the surface of the workpiece, and a data structure for storing model data, where the model data is indicative of the spatial relationship between the surface of the workpiece and the observation plane of the imaging device. The surface inspection system further includes an anomaly detection module adapted to receive the image data from the imaging device and for identifying at least one potential surface defect in the image data, and a defect tracking module connected to the data structure and the anomaly detection module for tracking the potential surface defect from the first set of image data to the second set of image data by using said model data, thereby assessing if the potential surface defect constitutes a defect in the surface of the workpiece.

Claims

1. A surface inspection system for detecting defects on a surface of a workpiece, comprising:

a diffused light source for emitting an elongated line of light onto the surface of the workpiece;

a movable member for translating the workpiece in relation to said light source;

an imaging device for capturing two or more sets of image data representative of a portion of the surface of the workpiece, said imaging device having a field of observation and being positionable at a vantage point such that the line of light is within the field of observation;

a data structure for storing model data indicative of a spatial relationship between said portion of the surface of the workpiece and the observation plane of said imaging device and;

an anomaly detection subsystem adapted to receive at least a first set of image data and a second set of image data from said imaging device, said anomaly detection module identifying at least one potential surface defect in said first set of image data and said second set of image data; and

a tracking subsystem connected to said data structure and said anomaly detection module for tracking the potential surface defect from said first set of image data to said second set of image data using said model data, thereby assessing if the potential surface defect constitutes a defect in the surface of the workpiece.

2. The surface inspection system of claim 1 wherein said tracking subsystem operative to transform the velocity of the workpiece into a velocity at which a potential surface defect is moving in relation to said imaging device through the use of said model data, and to predict a location of the potential surface defect from said first set of image data to said second set of image data through the use of said velocity.

3. The surface inspection system of claim 1 wherein the workpiece is modeled using a plurality of facets, where each facet represents a portion of the surface of the workpiece, and said model data is indicative of the spatial relationship between each facet of the workpiece and the observation plane of said imaging device.

4. The surface inspection system of claim 3 wherein said tracking subsystem operative to access model data which corresponds to the portion of the surface of the workpiece being inspected by the surface inspection system.

5. The surface inspection system of claim 1 further comprising at least two diffused light sources for emitting at least two elongated lines of light onto the surface of the workpiece.

6. The surface inspection system of claim 5 being operative to adjust an intensity of said light sources based on the color of the surface of the workpiece.

7. The surface inspection system of claim 1 wherein said anomaly detection subsystem includes:

a subtraction module receiving said first set of image data and said second set of image data and operative to generate image data indicative of the differences between said first set of image data and said second set of image data; and

a thresholding module receiving the image data from said subtraction module and operative to generate a binary representation of said image data.

8. The surface inspection system of claim 1 wherein said anomaly detection subsystem further includes:

a edge detection module receiving said first set of image data and said second set of image data and operative to identify the edges of the line of light in each set of image data;

a second thresholding module receiving said first and second sets of image data from said edge detection module and operative to generate a binary representation for each set of image data;

a dilation module receiving said first and second sets of image data from said second thresholding module and operative to thicken the edges of the line of light in each set of image data; and

a mask formation module receiving said first and second sets of image data from said dilation module, said mask formation module operative to combine said first set of image data with said second set of image data into a resultant set of image data using an OR operation and to invert the resultant image data using a NOT operation, thereby forming a mask for filtering out a portion of the noise in the image data.

9. The surface inspection system of claim 1 wherein said anomaly detection subsystem further includes:

an anomaly identification module receiving image data from said thresholding module and a mask from said mask formation module and operative to apply said mask to said image data using an AND operation, thereby filtering out a portion of the noise in the image data.

10. The surface inspection system of claim 1 wherein said movable member is further defined as a conveyor system.

11. The surface inspection system of claim 1 wherein said workpiece is further defined as a vehicle body for a motor vehicle and the surface inspection system operates within a motor vehicle manufacturing system.

12. A method of detecting defects in a surface of a workpiece using a surface inspection system, comprising the steps of:

emitting a line of light onto the surface of the workpiece;

passing the surface of the workpiece through the line of light;

capturing a first set of image data representative of a portion of the surface of the workpiece using an imaging device, said imaging device having a plane of observation and being positionable at a vantage point such that the line of light is within its plane of observation;

providing model data indicative of the spatial relationship between said portion of the surface of the workpiece and the observation plane of said imaging device;

capturing a second set of image data representative of said portion of the surface of the workpiece using said imaging device;

identifying at least one potential surface defect in said second set of image data;

capturing a third set of image data representative of said portion of the surface of the workpiece using said imaging device; and

tracking the potential surface defect from said second set of image data to said third set of image data using said model data, thereby assessing if the potential surface defect constitutes a defect in the surface of the workpiece.

13. The method of claim 12 wherein the step of emitting an elongated line of light further comprises using at least two diffused light sources to emit at least two elongated lines of light onto the surface of the workpiece.

14. The method of claim 13 further comprising the step of adjusting an intensity of said light sources based on the color of the surface of the workpiece.

15. The method of claim 12 wherein the step of identifying at least one potential surface defect further comprises:

subtracting said first set of image data from said second set of image data to form an intermediary set of image data, the intermediary set of image data indicative of potential surface defects in at least said second set of image data.

16. The method of claim 15 further comprises generating a binary representation of the intermediary set of image data.

17. The method of claim 15 further comprises the step of filtering out at least a portion of noise in said first and second set of image data, where the noise is caused by a variation in the viscosity of the paint on the surface of the workpiece.

18. The method of claim 17 wherein the step of filtering out at least a portion of noise further comprises:

providing a copy of said first set and said second set of image data;

identifying the edges of the line of light in each copy of image data;

combining the copy of said first set of image data with the copy of said second set of image data into a resultant set of image data by using an OR operation;

inverting said resultant set of image data using a NOT operation, thereby forming a mask for filtering out a portion of the noise; and

applying said mask to said intermediary set of image data using an AND operation, thereby filtering out a portion of the noise in the image data.

19. The method of claim 12 further comprising:

modeling the workpiece using a plurality of facets, where each facet represents a portion of the surface of the workpiece;

providing model data indicative of the spatial relationship between each facet of the workpiece and the observation plane of said imaging device; and

tracking the potential surface defect using model data which corresponds to the portion of the surface of the workpiece being inspected by the surface inspection system.

20. The method of claim 12 wherein the step of tracking the potential surface defect further comprises:

determining a first location of the potential surface defect in a reference frame for said imaging device, where the first location derived from said second set of image data;

determining a velocity for the workpiece;

transforming the velocity of the workpiece into a velocity at which potential surface defects are moving in relation to the said imaging device through the use of said model data;

predicting a location of the potential surface defect in said third set of image data by using the velocity for potential surface defects;

determining a second location of the potential surface defect in the reference frame of said imaging device, where the second location is derived from said third set of image data; and

comparing the predicted location of the potential surface defect with said second location of the potential surface defect, thereby tracking the potential surface defect.

21. A method of detecting defects in a surface of a workpiece, comprising the steps of:

emitting a line of light onto the surface of the workpiece;

passing the surface of the workpiece through the line of light;

capturing a first set of image data representative of a portion of the surface of the workpiece using an imaging device, said imaging device having a plane of observation and being positionable at a vantage point such that the line of light is within its plane of observation;

identifying at least one potential surface defect in said first set of image data;

determining a velocity for the workpiece;

transforming the velocity of the workpiece into a velocity at which potential surface defects are moving in relation to said imaging device;

capturing a second set of image data representative of said portion of the surface of the workpiece using said imaging device; and

tracking the potential surface defect from said first set of image data to said second set of image data using the velocity at which potential surface defects are moving in relation to said imaging device, thereby assessing if the potential surface defect constitutes a defect in the surface of the workpiece.

22. The method of claim 21 wherein the step of emitting an elongated line of light further comprises using at least two diffused light sources to emit at least two elongated lines of light onto the surface of the workpiece.

23. The method of claim 22 further comprising the step of adjusting an intensity of said light sources based on the color of the surface of the workpiece.

24. The method of claim 21 wherein the step of identifying at least one potential surface defect further comprises:

subtracting said first set of image data from said second set of image data to form an intermediary set of image data, the intermediary set of image data indicative of potential surface defects in at least said second set of image data.

25. The method of claim 24 further comprises generating a binary representation of the intermediary set of image data.

26. The method of claim 21 further comprises the step of filtering out at least a portion of noise in said first and second set of image data prior to tracking the potential surface defect, where the noise is caused by a variation in the viscosity of the paint on the surface of the workpiece.

27. The method of claim 26 wherein the step of filtering out at least a portion of noise further comprises:

providing a copy of said first set and said second set of image data;

identifying the edges of the line of light in each copy of image data;

combining the copy of said first set of image data with the copy of said second set of image data into a resultant set of image data by using an OR operation;

inverting said resultant set of image data using a NOT operation, thereby forming a mask for filtering out a portion of the noise; and

applying said mask to said intermediary set of image data using an AND operation, thereby filtering out a portion of the noise in the image data.

28. The method of claim 21 further comprising:

modeling the workpiece using a plurality of facets, where each facet represents a portion of the surface of the workpiece;

providing model data indicative of the spatial relationship between each facet of the workpiece and the observation plane of said imaging device; and

tracking the potential surface defect using model data which corresponds to the portion of the surface of the workpiece being inspected by the surface inspection system.