Target and method of detecting, identifying, and determining 3-D pose of the target

• US 9,002,062 B2
• Filed: 03/12/2013
• Issued: 04/07/2015
• Est. Priority Date: 10/14/2008
• Status: Expired due to Fees

• Current Assignee: Jolly Seven, Series 70 of Allied Security Trust I (Allied Security Trust)
• Original Assignee: Spot Vision LLC
• Inventors: Aller, Joshua Victor
• Primary Examiner(s): Alavi, Amir
• Application Number:
  US13/797,032
• Publication Number:
  US 20130259304A1
• Time in Patent Office: 756 Days
• Field of Search: 382/162, 382/164, 382/165, 382/168, 382/172, 382270-271, 382/103, 382/107, 382/296, 345/419, 345/633, 358/465, 358/466, 358/515, 348/158, 283/85, 386/278, 235/494
• US Class Current: 382/103
• CPC Class Codes: B42D 15/00 Printed matter of special format or style not otherwise provided for
• CPC Class Codes: G06T 2207/30204 Marker
• CPC Class Codes: G06T 7/73 using feature-based methods
• CPC Class Codes: G06V 10/245 by locating a pattern; Special marks for positioning

Abstract

We disclose a photogrammetry target that includes a background having a first color and a plurality of ovoid regions located on the background and having a second color contrasting the first color. We further disclose a method and system for detecting the target and processing image data captured from the target to discern therefrom at least one of a distance to the target, identification of the target, or pose of the target.

Claims

1. A target, comprising:

a background shape,

a plurality of foreground figures arranged in the background shape and configured to allow a transform that when applied to the background shape and at least one foreground figure converts the at least one foreground figure between a set of model space parameters and a set of rotationally non-symmetric metric space parameters assigned to the set of model space parameters.

2. The target of claim 1 wherein the plurality of foreground figures are positioned to achieve a maximum separation within the offset background shape.

3. The target of claim 1 further comprising additional features outside of the background shape configured to aid in the detection of the orientation of the target.

4. The target of claim 1 wherein the plurality of figures are ellipses configured to differentiate and identify the target, each having one or more metric space parameters from the group of size, aspect ratio, and orientation.

5. The target of claim 1 wherein the offset background shape has a background area and the plurality of foreground figures have a total area related to the background area by a predetermined total area to background area ratio.

6. The target of claim 1 wherein the plurality of foreground figures are chosen from a fixed pallet of various size and shapes configured to allow encoding of digital information without having to identify each plurality of foreground figure shape or pattern of placement of the plurality of foreground figures.

7. The target of claim 1 wherein at least two of the plurality of foreground figures have a different light intensity color or chroma.

8. The target of claim 7 wherein the each of the different light intensity colors encode substantially the same luminance light intensity and configured to allow detection by at least one of a color camera and a black and white camera.

9. The target of claim 1 wherein the metric space parameters includes a parameter from the group of fractional area of background, aspect ratio, rotational ordering, distance between center of background shape to center of respective figure, angle between respective figure and next respective figure, and alignment of spokes created between center of background and respective center of plurality of figures.

10. The target of claim 1 wherein at least one foreground figure has an eccentricity greater than one and the metric space parameters includes an angle between a major axis of the at least one foreground figure and a spoke defined from the center of the background shape to the center of the at least one foreground figure.

11. The target of claim 1 wherein the plurality of foreground figures are further arranged to identify the target and uniquely identify each of the foreground figures.

12. A target detection system, comprising:

an image receiver configured to receive an image with a set of pixels representing a target having a background and a plurality of foreground figures; and

a figure detector coupled to the image receiver including,

a schema, and

a model space convertor configured to extract a set of rotationally non-symmetric metric space parameters from the plurality of foreground figures and compare to metrics space parameters associated with at least one shape in the schema.

13. The system of claim 12 further comprising an area ratio adjustment module coupled to the schema, the area ratio adjustment module including a boundary shift compensation module configured to match an area ratio of a target shape within the set of pixels from the schema.

14. The system of claim 13 further comprising a multi-part shape detector configured to detect figures with a contouring method that extracts at least one iso-contour polygon from the set of pixels in the image provided by the image receiver.

15. The system of claim 14 wherein the contouring method uses a triangular lattice superimposed over the set of pixels of the image, the triangles of the triangular lattice having vertices on at least one of pixel centers and pixel corners.

16. The system of claim 15 wherein the at least one iso-contour polygon is extracted by selecting an intensity or chroma value within the set of pixels and linear interpolation is used between three corners of a triangle to find a directed line segment that represent the crossing of the intensity value within the triangle.

17. The system of claim 14 wherein a center of mass is determined by the multi-part shape detector and a unit circle is centered within the iso-contour polygons and a first matrix of points on the at least one iso-contour polygon is used with a second matrix of points on the unit circle to create a linear transform matrix that represents a fit between the points in the first and second matrices.

18. The system of claim 17 wherein minor and major axes of a detected ellipse is determined by extracting eigenvectors and eigenvalues from the linear transform matrix.

19. The system of claim 13 further comprising a 3-D position and orientation extractor coupled to the shape matching and marker identifier configured to determine a 3-D pose of the target.

20. The system of claim 12 further including a shape matching and marker identifier coupled to the schema and the model space convertor configured to detect an id of the target.

21. The system of claim 20, where in the target is located within a fiducial marker and the shape matching and marker identifier is further configured to detect 2-D markers in the fiducial marker outside of the target and the target is configured to encode the fiducial coding system and location of the 2-D markers.

22. One or more non-transitory computer-readable media storing computer-executable instructions, which when executed perform steps, comprising:

receiving an image with a set of pixels representing a target having a background and a plurality of foreground figures; and

detecting a figure within the image with a model space convertor by,

matching a shape and marker of the figure using a schema, and

extracting and comparing a set of rotationally non-symmetric metric space parameters from the plurality of foreground figures to metrics space parameters associated with at least one shape in the schema.

23. The computer readable media of claim 22, further comprising computer executable instructions which when executed perform the step of detecting an id of the target based on the model space convertor and the schema.

24. A target, comprising:

a background shape, and

a plurality of foreground figures arranged in the background shape and described by at least two coordinate systems including a geometric space to define a target pattern of the foreground figures and a parameter space to describe a rotationally invariant representation of the target pattern.

25. The target of claim 24 wherein none of the plurality of foreground figures are centered at or close to the center of the target.

26. The target of claim 24 wherein the plurality of foreground figures are further arranged to identify the target and uniquely identify each of the foreground figures.

27. The target of claim 26 wherein the plurality of foreground figures are arranged according to a schema of a fixed set of areas and aspect ratios and each foreground figure assigned a unique symbol and a unique element such that the schema when represented as a chart resembles a periodic table and each target is analogous to a molecule described in terms of the figure symbols and counts of the figure elements.

28. The target of claim 24 wherein the plurality of foreground figures are arranged according to a schema of consistent total foreground figure area to background area ratios.

29. The target of claim 24 wherein the foreground figures and the background shape are represented with two different polarities to substantially double the number of detectable targets.

30. The target of claim 24 where the plurality of foreground figures is configured by a randomly generated package of random figures.

31. A system for measuring position or orientation, comprising:

a plurality of targets each including a background shape, and a plurality of foreground figures arranged in the background shape and described by at least two coordinate systems including a geometric space to define a target pattern of the foreground figures and a parameter space to describe a rotationally invariant representation of the target pattern; and

a central processor coupled to a memory and a display, and an application program executing processor readable instructions from the memory, the application program configured to review an image, search for and detect one or more of the plurality of targets in the image, determine the full attitude pose of each target.

32. The system of claim 31 wherein the central processor further includes a camera and a camera lens and the application program is further configured to grab one or more video frames from the camera for the image.

33. The system of claim 32 wherein the one or more measurements are from the group consisting of distance between the camera lens and target center, angle from camera lens to target surface, the normal distance to the plane of target surface, a surface area of a shape indicated by targets, and a pose for the central processor relative to a global coordinate system.

34. The system of claim 32 wherein the application program is further configured with instructions to determine the intensity value of the video frames, detect a range of contour threshold values to search for and detect the one or more plurality of targets, and modify threshold values for each video frame to be at the center of the workable range from a previous frame.

35. The system of claim 32 wherein the central processor includes one or more non-visual sensors and the application program is further configured to combine the non-visual sensor data with the measurements from the image in order to present more meaningful information.