SYSTEM AND METHOD FOR ESTIMATING THE HEIGHT OF AN OBJECT USING TOMOSYNTHESIS-LIKE TECHNIQUES

• US 20110193953A1
• Filed: 02/05/2010
• Published: 08/11/2011
• Est. Priority Date: 02/05/2010
• Status: Active Grant

• Current Assignee: Applied Vision Corporation (Antares Vision S.p.A.)
• Original Assignee: Applied Vision Company, LLC
• Inventors: Sones, Richard A., Sebeny, Carl E.
• Granted Patent: US 8,508,591 B2
• Time in Patent Office: Days
• US Class Current: 348/135
• CPC Class Codes: G06T 2207/10112 Digital tomosynthesis [DTS]
• CPC Class Codes: G06T 2207/30108 Industrial image inspection
• CPC Class Codes: G06T 2207/30141 Printed circuit board [PCB]
• CPC Class Codes: G06T 7/571 from focus
• CPC Class Codes: H04N 7/18 Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast

Abstract

Systems and methods to estimate the height profile of an object using tomosynthesis-like techniques. A plurality of raw images of an object to be characterized are acquired, where the plurality of raw images are representative of a plurality of spatial shifts of an imaging device relative to the object to be characterized. The raw images are processed to generate composite images, where each composite image corresponds to a unique image shift between spatially adjacent raw images. A volatility parameter value is calculated within a neighborhood of a same image pixel location for each composite image. The composite image having the largest volatility parameter value for the image pixel location is determined. A unique image shift, corresponding to the composite image having the largest volatility parameter value, is transformed into a height value representative of a height dimension of the image pixel location.

Claims

1. A method to estimate the height profile of an object, said method comprising:

acquiring a plurality of raw images of an object to be characterized using at least one imaging device, where said plurality of raw images are representative of a plurality of spatial shifts of said at least one imaging device relative to said object to be characterized;

processing said plurality of raw images to generate a plurality of composite images, where each composite image of said plurality of composite images corresponds to a unique image shift between spatially adjacent raw images and is made up of a plurality of image pixel locations;

calculating a volatility parameter value within a neighborhood of at least one same image pixel location of said plurality of image pixel locations for each of said plurality of composite images;

determining which composite image of said plurality of composite images has a largest volatility parameter value for said at least one same image pixel location; and

transforming said unique image shift, corresponding to said composite image having said largest volatility parameter value, into a height value representative of a height dimension of said at least one same image pixel location.

2. The method of claim 1 further comprising spatially translating said at least one imaging device relative to said object to be characterized during acquisition of said plurality of raw images.

3. The method of claim 1 further comprising spatially translating said object to be characterized relative to said at least one imaging device during acquisition of said plurality of raw images.

4. The method of claim 1 further comprising generating a height value for each image pixel location of said plurality of image pixel locations to form a height profile image of said object to be characterized.

5. The method of claim 4 further comprising displaying said height profile image.

6. The method of claim 1 further comprising filtering said volatility parameter value across said plurality of composite images for said at least one same image pixel location before determining which composite image of said plurality of composite images has a largest volatility parameter value for said at least one same image pixel location.

7. The method of claim 1 wherein said unique image shift is a fraction of a distance between two adjacent image pixel locations.

8. The method of claim 1 wherein said processing said plurality of raw images includes using a sub-pixel location interpolation technique to generate said plurality of composite images.

9. The method of claim 4 further comprising comparing said height profile image to at least one reference image to find any defects in said object to be characterized.

10. A system to estimate the height profile of an object, said system comprising:

means for acquiring a plurality of raw images of an object to be characterized, where each of said plurality of raw images is acquired at a unique spatial location relative to said object to be characterized;

means for processing said plurality of raw images to generate a plurality of composite images, where each composite image of said plurality of composite images corresponds to a unique image shift between spatially adjacent raw images and is made up of a plurality of image pixel locations;

means for calculating a volatility parameter value within a neighborhood of at least one same image pixel location of said plurality of image pixel locations for each of said plurality of composite images;

means for determining which composite image of said plurality of composite images has a largest volatility parameter value for said at least one same image pixel location; and

means for transforming said unique image shift, corresponding to said composite image having said largest volatility parameter value, into a height value representative of a height dimension of said at least one same image pixel location.

11. The system of claim 10 further comprising means for spatially translating said means for acquiring relative to said object to be characterized during acquisition of said plurality of raw images.

12. The system of claim 10 further comprising means for spatially translating said object to be characterized relative to said means for acquiring during acquisition of said plurality of raw images.

13. The system of claim 10 further comprising means for generating a height value for each image pixel location of said plurality of image pixel locations to form a height profile image of said object to be characterized.

14. The system of claim 13 further comprising means for displaying said height profile image.

15. The system of claim 10 further comprising means for filtering said volatility parameter across said plurality of composite images for said at least one same image pixel location before determining which composite image of said plurality of composite images has a largest volatility parameter value for said at least one same image pixel location.

16. The system of claim 13 further comprising means for comparing said height profile image to at least one reference image to find any defects in said object to be characterized.

17. An apparatus to estimate the height profile of an object, said apparatus comprising:

at least one imaging device capable of acquiring each of a plurality of raw images of an object to be characterized at a unique spatial relation to said object to be characterized; and

a vision engine operatively connected to said at least one imaging device and capable of;

processing said plurality of raw images to generate a plurality of composite images, where each composite image of said plurality of composite images corresponds to a unique image shift between spatially adjacent raw images and is made up of a plurality of image pixel locations;

calculating a volatility parameter value within a neighborhood of at least one same image pixel location of said plurality of image pixel locations for each of said plurality of composite images,

determining which composite image of said plurality of composite images has a largest volatility parameter value for said at least one same image pixel location, and

transforming said unique image shift, corresponding to said composite image having said largest volatility parameter value, into a height value representative of a height dimension of said at least one same image pixel location.

18. The system of claim 17 further comprising a spatial translation mechanism operatively connected to said at least one imaging device and to said vision engine, and capable of spatially translating said at least one imaging device relative to said object to be characterized.

19. The system of claim 17 further comprising a spatial translation mechanism capable of spatially translating said object to be characterized relative to said at least one imaging device.

20. The system of claim 17 wherein said vision engine is further capable of generating a height value for each image pixel location of said plurality of image pixel locations to form a height profile image of said object to be characterized.

21. The system of claim 17 wherein said vision engine is further capable of filtering said volatility parameter value across said plurality of composite images for said at least one same image pixel location before determining which composite image of said plurality of composite images has a largest volatility parameter value for said at least one same image pixel location.

22. The system of claim 20 further comprising a display operatively connected to said vision engine and capable of displaying said height profile image.

23. The system of claim 20 wherein said vision engine is further capable of comparing said height profile image to at least one reference image to find any defects in said object to be characterized.

24. A method to estimate the dimensions of a cuboid, said method comprising:

(a) acquiring a plurality of raw images of a cuboid using at least one imaging device, where said plurality of raw images are representative of a plurality of spatial shifts of said at least one imaging device relative to said cuboid;

(b) processing said plurality of raw images to generate a composite image, where said composite image corresponds to a unique image shift between spatially adjacent raw images and is made up of a plurality of image pixel locations;

(c) analyzing said composite image in an attempt to find upper-most edges of said cuboid in said composite image;

(d) repeating steps (b) and (c) using a different unique image shift until said upper-most edges of said cuboid are found; and

(e) transforming said unique image shift, corresponding to said composite image in which said upper-most edges of said cuboid have been found, into an estimated height of said cuboid.

25. The method of claim 24 further comprising spatially translating said at least one imaging device relative to said cuboid during acquisition of said plurality of raw images.

26. The method of claim 24 further comprising spatially translating said cuboid relative to said at least one imaging device during acquisition of said plurality of raw images.

27. The method of claim 24 further comprising estimating a length and a width of said cuboid using at least said estimated height of said cuboid and a similar triangles technique.

28. The method of claim 27 further comprising estimating a volume of said cuboid using said estimated height, said estimated length, and said estimated width of said cuboid.

29. A system to estimate the dimensions of a cuboid, said system comprising:

means for acquiring a plurality of raw images of a cuboid, where each of said plurality of raw images is acquired at a unique spatial location relative to said cuboid;

means for processing said plurality of raw images to generate a composite image, where said composite image corresponds to a unique image shift between spatially adjacent raw images and is made up of a plurality of image pixel locations;

means for analyzing said composite image in an attempt to find upper-most edges of said cuboid in said composite image; and

means for transforming said unique image shift, corresponding to said composite image in which said upper-most edges of said cuboid have been found, into an estimated height of said cuboid.

30. The system of claim 29 further comprising means for spatially translating said means for acquiring relative to said cuboid during acquisition of said plurality of raw images.

31. The system of claim 29 further comprising means for spatially translating said cuboid relative to said means for acquiring during acquisition of said plurality of raw images.

32. The system of claim 29 further comprising means for estimating a length and a width of said cuboid using at least said estimated height of said cuboid and a similar triangles technique.

33. The system of claim 32 further comprising means for estimating a volume of said cuboid using said estimated height, said estimated length, and said estimated width of said cuboid.

34. An apparatus to estimate the dimensions of a cuboid, said apparatus comprising:

at least one imaging device capable of acquiring each of a plurality of raw images of a cuboid at a unique spatial relation to said cuboid; and

a vision engine operatively connected to said at least one imaging device and capable of;

(a) processing said plurality of raw images to generate a composite image, where said composite image corresponds to a unique image shift between spatially adjacent raw images and is made up of a plurality of image pixel locations,

(b) analyzing said composite image in an attempt to find upper-most edges of said cuboid in said composite image,

(c) repeating steps (a) and (b) using a different unique image shift until said upper-most edges of said cuboid are found, and

(d) transforming said unique image shift, corresponding to said composite image in which said upper-most edges of said cuboid have been found, into an estimated height of said cuboid.

35. The apparatus of claim 34 further comprising a spatial translation mechanism operatively connected to said at least one imaging device and to said vision engine, and capable of spatially translating said at least one imaging device relative to said cuboid.

36. The apparatus of claim 34 further comprising a spatial translation mechanism capable of spatially translating said cuboid relative to said at least one imaging device.

37. The apparatus of claim 34 wherein said vision engine is further capable of:

estimating a length and a width of said cuboid using at least said estimated height of said cuboid and a similar triangles technique; and

estimating a volume of said cuboid using said estimated height, said estimated length, and said estimated width of said cuboid.