System and method for dimensioning objects using stereoscopic imaging

• US 8,781,159 B2
• Filed: 04/08/2011
• Issued: 07/15/2014
• Est. Priority Date: 05/13/2009
• Status: Active Grant

• Current Assignee: Applied Vision Corporation (Antares Vision S.p.A.)
• Original Assignee: Applied Vision Corporation (Antares Vision S.p.A.)
• Inventors: Sones, Richard A., Brumbaugh, Kris, Wilson, Ryan
• Primary Examiner(s): DULANEY, KATHLEEN YUAN
• Application Number:
  US13/082,702
• Publication Number:
  US 20110188741A1
• Time in Patent Office: 1,194 Days
• Field of Search: 382/101, 382/154, 356/601, 356/625, 235/454, 235/440
• US Class Current: 382/101
• CPC Class Codes: G01B 11/02 for measuring length, width or thickness G01B11/08 takes precedence
• CPC Class Codes: G06F 18/00 Pattern recognition
• CPC Class Codes: G06T 2207/30204 Marker
• CPC Class Codes: G06T 7/62 of area, perimeter, diameter or volume
• CPC Class Codes: G06T 7/73 using feature-based methods

Abstract

A method and configuration to estimate the dimensions of a cuboid. The configuration includes two image acquisition units offset from each other with at least one of the units positioned at a defined acquisition height above a background surface. Image processing techniques are used to extract a perimeter of a top surface of the cuboid, placed on the background surface, from pairs of acquired images. A height estimation technique, which corrects for spatial drift of the configuration, is used to calculate an absolute height of the cuboid. The absolute height of the cuboid is used, along with the extracted perimeter of the top surface of the cuboid, to calculate an absolute length and an absolute width of the cuboid. The height, length, and width may be used to calculate an estimated volume of the cuboid.

Claims

1. An image acquisition and processing configuration to estimate the dimensions of a package, said configuration comprising:

a primary image acquisition unit positioned at a primary acquisition location;

a secondary image acquisition unit positioned at a secondary acquisition location which is offset from said primary acquisition location by a defined offset distance;

a platform having a substantially horizontal top surface positioned substantially directly beneath said primary image acquisition unit such that said primary image acquisition unit is at a defined image acquisition height above said top surface of said platform;

a fiducial marker positioned on said top surface of said platform; and

a vision engine operatively connected to said image acquisition units and configured for;

triggering said image acquisition units to acquire at least background images of said top surface of said platform and package images of a top surface of a package placed on said top surface of said platform, wherein at least one of said images is acquired using said primary image acquisition unit and contains an imaged representation of said fiducial marker, and wherein at least one other of said images is acquired using said secondary image acquisition unit and contains an imaged representation of said fiducial marker,

receiving said acquired images from said image acquisition units,

processing two of said received images, each of said two images having therein an imaged representation of said fiducial marker, to determine a fiducial image shift accounting for spatial drift of said image acquisition and processing configuration, and

processing said received images using said fiducial image shift to estimate at least a height dimension of said package.

2. The configuration of claim 1 wherein said platform comprises a scale capable of weighing said package placed on said top surface of said platform.

3. The configuration of claim 1 further comprising a scale having said platform resting thereon and capable of weighing said package placed on said top surface of said platform.

4. The configuration of claim 1 wherein each of said image acquisition units comprises a camera.

5. The configuration of claim 1 further comprising at least one illuminator positioned to illuminate said top surface of said platform and said top surface of said package placed on said platform, wherein said at least one illuminator is operatively connected to said vision engine and is capable of being turned on by said vision engine.

6. The configuration of claim 5 wherein said at least one illuminator and said image acquisition units operate in an infrared range of wavelengths.

7. The configuration of claim 5 wherein said at least one illuminator and said image acquisition units operate in an ultraviolet range of wavelengths.

8. The configuration of claim 5 wherein said at least one illuminator and said image acquisition units operate in a visible range of wavelengths.

9. The configuration of claim 5 further comprising a linear polarizer operatively connected to said at least one illuminator.

10. The configuration of claim 9 further comprising a cross-polarizer operatively connected to said primary image acquisition unit.

11. The configuration of claim 9 further comprising a cross-polarizer operatively connected to said secondary image acquisition unit.

12. The configuration of claim 1 wherein said configuration is capable of accommodating cuboid-shaped packages ranging in dimensions between 2 inches by 2 inches by one inch and 36 inches by 24 inches by 26 inches.

13. The configuration of claim 1 wherein said processing performed by said vision engine to estimate at least a height dimension of said package using said fiducial image shift includes:

correcting each of said acquired images for optical distortion to form a corrected primary background image, a corrected secondary background image, a corrected primary package image, and a corrected secondary package image;

performing edge enhancement and edge extraction on each of said corrected images to form a primary background edge image, a secondary background edge image, a primary package edge image, and a secondary package edge image;

subtracting said primary background edge image from said primary package edge image to form a primary subtraction image;

subtracting said secondary background edge image from said secondary package edge image to form a secondary subtraction image;

finding a best-fit rectangle within said primary subtraction image to extract a primary rectangle representing said top surface of said package;

finding a best-fit rectangle within said secondary subtraction image to extract a secondary rectangle representing said top surface of said package;

spatially correlating at least one common feature of said primary rectangle and said secondary rectangle;

calculating a spatial image shift between said primary rectangle and said secondary rectangle using said common correlated feature;

subtracting said fiducial image shift from said spatial image shift to form a corrected spatial image shift;

calculating an absolute height of said package by applying a height estimation technique using at least said corrected spatial image shift and said image acquisition height; and

calculating an absolute length and an absolute width of said package using said calculated absolute height of said package and at least one of said primary rectangle and said secondary rectangle.

14. The configuration of claim 13 wherein the processing performed by the vision engine further includes calculating an estimated volume of said package using said absolute length, said absolute width, and said absolute height.

15. The configuration of claim 13 wherein said processing performed by said vision engine further includes binarizing said primary subtraction image and said secondary subtraction image before finding said best-fit rectangles.

16. The configuration of claim 1 further comprising an optical bandpass filter operatively connected to said primary image acquisition unit to reject ambient light.

17. The configuration of claim 1 further comprising an optical bandpass filter operatively connected to said secondary image acquisition unit to reject ambient light.

18. The configuration of claim 1 wherein said processing performed by said vision engine to determine a fiducial image shift accounting for spatial drift of said image acquisition and processing configuration includes:

performing a centroiding operation on each of said two images to determine a center of said fiducial marker in each of said two images; and

comparing said centers to determine the fiducial image shift of the fiducial marker between the two images at least along a direction between the acquisition positions, wherein said fiducial image shift is used to correct an estimate of a height of the package.

19. The configuration of claim 1 wherein said processing performed by said vision engine to determine a fiducial image shift accounting for spatial drift of said image acquisition and processing configuration includes performing a correlation operation on said two images to determine the fiducial image shift of the fiducial marker between the at least two images at least along a direction between the acquisition positions, wherein said fiducial image shift is used to correct an estimate of a height of a cuboid structure.