Measurement of dimensions of solid objects from two-dimensional image(s)

• US 6,995,762 B1
• Filed: 09/13/2002
• Issued: 02/07/2006
• Est. Priority Date: 09/13/2001
• Status: Expired due to Term

• Current Assignee: Symbol Technologies, LLC (Zebra Technologies Corporation)
• Original Assignee: Symbol Technologies, Inc. (Zebra Technologies Corporation)
• Inventors: He, Duanfeng, Joseph, Eugene, Hatton, Edward, Lu, Kefei, Pavlidis, Theo
• Primary Examiner(s): Nguyen, Phu K.
• Application Number:
  US10/243,396
• Time in Patent Office: 1,243 Days
• Field of Search: 345/419, 345/418, 345/619, 345/617, 382/106, 348/46, 348/47, 348/140
• US Class Current: 345/419
• CPC Class Codes: G06T 17/00 Three dimensional [3D] modelling, e.g. data description of 3D objects
• CPC Class Codes: G06T 7/50 Depth or shape recovery

Abstract

The present invention facilitates solid object reconstruction from a two-dimensional image. If an object is of known and regular shape, information about the object can be extracted from at least one view by utilizing appropriate constraints and measuring a distance between a camera and the object and/or by estimating a scale factor between a camera image and a real world image. The same device can perform both the image capture and the distance measurement or the scaling factor estimation. The following processes can be performed for object identification: parameter estimation; image enhancement; detection of line segments; aggregation of short line segments into segments; detection of proximity clusters of segments; estimation of a convex hull of at least one cluster; derivation of an object outline from the convex hull; combination of the object outline, shape constraints, and distance value.

Claims

1. An image processing system comprising:

an image capture component that receives at least one view of a two-dimensional image, the image capture component employing two laser beams to create bright spots on the object to calculate the scale factor;

an image analyzer that analyzes the captured image, the image analyzer including at least one of a distance analyzer and scaling component employed in connection with generating data for object construction; and

an object construction component that receives the analyzed image and generates a three-dimensional object relating to the captured two-dimensional image.

2. The system of claim 1 locating edges of an image in connection with locating corners of the image.

3. The image processing system of claim 1, further comprising a data store to store image related data.

4. The image processing system of claim 1, further comprising a processing component that receives the generated three-dimensional object and employs the three-dimensional object in connection with determining measurement of the object.

5. A camera employing the image processing system of claim 1.

6. A portable computing devices employing the image processing system of claim 1.

7. A cellular telephone employing the image processing system of claim 1.

8. An inventory monitoring device employing the image processing system of claim 1.

9. A method for object reconstruction comprising:

capturing an image, the image being a single two-dimensional view of a solid object;

employing a subset of total viewable corners of the solid object in connection with the object reconstruction;

detecting a plurality of line segments of the captured image;

identifying an outline of the object; and

forming a three-dimensional model of the object.

10. The method of claim 9, further comprising automatically extracting at least one parameter for at least one of smoothing out noise and enhancing contrast.

11. The method of claim 9, wherein the detecting the plurality of line segments is performed over a plurality of blocks of the image.

12. The method of claim 11, further comprising identifying a pixel for each of the plurality of blocks where an image gradient is at a maximum value and each of the plurality of blocks is divided into two regions by a line perpendicular to the direction of the gradient and passing through the identified pixel.

13. The method of claim 12, further comprising estimating an average brightness and a corresponding square root average error for each of the regions for determining whether a line segment is significant.

14. The method of claim 9, further comprising detecting at least one proximity cluster, the proximity cluster comprising a set of line segments.

15. The method of claim 9, further comprising estimating a convex hull.

16. The method of claim 15, wherein the outline of the object is derived from the convex hull.

17. The method of claim 9, further comprising estimating a dimension for a three dimension rectangular box.

18. The method of claim 9, wherein the object is a three dimension box.

19. The method of claim 18, further comprising employing a least squared method for reconciling locations of corner points of the box.

20. The method of claim 9, further comprising determining parameters for the three-dimensional model via an orthographic solution.

21. The method of claim 9, further comprising determining parameters for the three-dimensional model via a linear solution.

22. The method of claim 9, further comprising determining parameters for the three-dimensional model via a combined orthographic and linear solution.

23. The method of claim 9, further comprising determining parameters for the three-dimensional model via an iterative solution.

24. The method of claim 9, further comprising determining parameters for the three-dimensional model via an unbiased iterative solution.

25. The method of claim 9, further comprising estimating a scale factor between a camera image and the object.

26. The method of claim 25, further comprising employing two laser beams to create bright spots on the object to calculate the scale factor.

27. The method of claim 9 being employed in a parcel slipping business.

28. The method of claim 9 further comprising locating one surface of the solid object and determining object makeup via employment of the one surface.

29. A method for object reconstruction comprising:

capturing at least two images taken with differing amount of integration time respectively;

employing a darker image of the at least two images to determine a laser spot and at least one surface of an object under measurement; and

employing a brighter image of the at least two images to determine another surface of the object.

30. The method of claim 29, wherein the at least two images are taken within a time period less than or equal to 40 ms.

31. A method for object reconstruction comprising:

capturing an image, the image being a single two-dimensional view of a solid object;

employing a subset of total viewable corners of the solid object in connection with the object reconstruction;

creating a sub-sampled version of the image;

enhancing the sub-sampled version of the image;

detecting a plurality of line segments of the sub-sampled version of the image, such that the line segments are scaled to original image dimensions;

aggregating the plurality of line segments of the captured image;

identifying an outline of the object; and

forming a three-dimensional model of the object.

32. An image processing system comprising:

an image capturing means for receiving at least one view of a two-dimensional image, the image capturing means employing two laser beams to create bright spots on the object to calculate the scale factor;

an image analyzing means for analyzing the captured image, the image analyzer means including at least one of a distance analyzer means and scaling component means employed in connection with generating data for object construction; and

means for receiving the analyzed image and generating a three-dimensional object relating to the captured two-dimensional image.

33. The system of claim 32, further comprising means for storing image related data and means for processing the image related data.