Automated coin grading system

• US 5,220,614 A
• Filed: 02/22/1991
• Issued: 06/15/1993
• Est. Priority Date: 02/22/1991
• Status: Expired due to Fees

• Current Assignee: Professional Coin Grading Service
• Original Assignee: Professional Coin Grading Service
• Inventors: Crain, Louis M.
• Primary Examiner(s): COUSO, JOSE L
• Application Number:
  US07/659,510
• Time in Patent Office: 844 Days
• Field of Search: 358/101, 358/106, 358/107, 382/1, 382/8, 382/48, 356/71
• US Class Current: 382/136
• CPC Class Codes: G06F 18/00 Pattern recognition
• CPC Class Codes: G06T 7/001 using an image reference approach
• CPC Class Codes: G06T 7/33 using feature-based methods
• CPC Class Codes: G06T 7/73 using feature-based methods
• CPC Class Codes: G07D 5/005 Testing the surface pattern, e.g. relief
• CPC Class Codes: G07D 5/10 Testing the rim, e.g. the milling of the rim

Abstract

A method and system for automated grading of coins involves alignment of a target image of the coin with a reference image by mechanically moving the coin until the target and reference images are aligned, capture of multiple images of the coin under a variety of lighting conditions, and subsequent grading using an expert script made up of individual directives which can be arranged to form different scripts for different grading standards and types of coins to be graded. The system and method provide for identification of the criteria used by an expert in grading an object such as a coin, capture of those criteria, and applying the criteria to the grading of the coin in a manner which is implementable on low-cost and widely available microcomputer hardware. The process by which the system is used to identify and test the criteria for evaluating the coin is interactive, and thus, structural changes need not be made in the system in order to grade the different types of coins or apply different coin grading standards.

Claims

1. An image capture system, comprising:

translation means for moving a first object in two dimensions, said two dimensions forming a plane;

rotation means for rotating said first object within said plane;

means including a camera for forming a first electronic image of said first object;

means for storing and retrieving a first electronic image of a reference object, said reference object being of the same type of object as said first object;

means for locating a distinct landmark on said first object;

location comparison means for comparing a location of said distinct landmark with a location of a corresponding landmark on said reference object;

orientation comparison means for comparing an orientation of said landmark with an orientation of said corresponding landmark;

translation and orientation control means for causing said translation means to move said first object a distance corresponding to a difference in locations of said first object landmark and said reference object landmark and for rotating said first object through an angle corresponding to an angular difference in orientations of said two landmarks;

means for causing said camera to form a second electronic image of said first object such that said second image of said first object is aligned with said first image of said reference object, said alignment resulting from movement of said object by said rotation means and said translation means,

wherein said translation and orientation control means further comprises means for positioning said distinct landmark on said first object in respect to said corresponding landmark on said reference object, subsequently rotating said first object about the center of said distinct landmark through a computed angular offset between a second landmark on said first object and a second landmark on said second object.

2. A system as claimed in claim 1, wherein said translation and orientation control means includes means for causing said translation and rotation to occur simultaneously.

3. A system as claimed in claim 1, wherein said camera is a video camera.

4. A system as claimed in claim 3, wherein said video camera is an RGB video camera.

5. A system as claimed in claim 3, further comprising means for inputting a video signal from said camera into a computer, said video signal inputting means including a real time image processing section having a plurality of look-up tables and a high speed frame buffer memory.

6. A system as claimed in claim 5, wherein said input means further comprises analog to digital conversion means for converting said video signal into a 480 line×

640 pixel image and for storing said image in said frame buffer memory.

7. A system as claimed in claim 6, wherein each pixel comprises eight bits of data.

8. A system as claimed in claim 1, wherein said translation means comprises an X table, a Y table, means for linearly moving said X table, and means for linearly moving said Y table in a direction perpendicular to the direction of movement of said X table.

10. A system as claimed in claim 8, wherein said rotation means comprises a rotatable table and means for rotating the table.

11. A system in claim 10, wherein said respective means for linearly moving said X table and said Y table, and said means for rotating said rotatable table, each comprises a stepper motor.

12. A system as claimed in claim 1, wherein said rotation means comprises a rotatable table and means for rotating the table.

13. A system as claimed in claim 12, wherein said means for rotating said table comprises a stepper motor.

14. A system as claimed in claim 1, wherein said rotation means comprises a table covered with means including a soft, high friction material for isolating said first object from other system elements.

15. A system as claimed in claim 1, wherein said rotation means comprises object holding means for holding a plurality of objects to be graded in sequence, one side at a time.

16. A system as claimed in claim 15, wherein said object holding means comprises a removable grading tray.

17. A system as claimed in claim 16, wherein said removable grading tray is made of a foam material.

18. A system as claimed in claim 1, wherein said rotation means comprises a piece of foam permanently mounted to a metal plate which is connected to a rotation table motor, and a second detachable piece of foam having a high coefficient of friction in respect to said first piece of foam and including means defining a surface on which objects are place for grading.

19. A system as claimed in claim 1, wherein said translation and rotation means move said first object incrementally, and the increments of movement are smaller than one pixel of either of said images.

20. A system as claimed in claim 1, wherein said translation and orientation control means comprises means for causing rotation means to rotate said object through an angle θ

equal to the difference in orientation between said first and second object landmarks as determined by said orientation comparison means, and means for causing said translation means to move said first object over mutually perpendicular distances Δ

X and Δ

Y determined by formulas

space="preserve" listing-type="equation">Δ

X=X.sub.0 cos θ

-Y.sub.0 sin θ

-X.sub.0

space="preserve" listing-type="equation">and

space="preserve" listing-type="equation">Δ

Y=X.sub.0 sin θ

+Y.sub.0 cos θ

-Y.sub.0

where X₀ and Y₀ are initial distances of the object along said mutually perpendicular axes from the axis of rotation about which said rotation means rotates said first object.

21. A system as claimed in claim 20, wherein said translation and rotation means move said first object incrementally, and the increments of movement are smaller than one pixel of either of said images.

22. A system as claimed in claim 21, wherein said translation and orientation control means includes means for controlling said movement in increments smaller than the resolution of the image by creating a bi-quadratic error surface from the correlation value of the optimum alignment pixel point and the eight surrounding pixels, and subsequently computing the location of the minimum value of a continuous version of the correlation function and thus the floating point angular offset.

23. A system as claimed in claim 1, further comprising means for repeating each of said object movements in an iterative process until alignment is achieved.

24. A system as claimed in claim 1, further comprising lighting means including a plurality of different individually controlled lights for varying an angle of illumination of said first object during image capture.

25. A system as claimed in claim 24, wherein said lighting means comprises eight different lights.

26. A system as claimed in claim 24, further comprising means including a computer for individually switching said lights on and off as a function of the type of characteristic of said first object sought by the computer for use in grading said first object, said image of said reference object being aligned with said image of said first object upon physical movement of said first object by said translation and rotation means such that reflections from said lights when capturing an image of said first object are aligned with reflections captured during capture of the reference object image.

28. A system as claimed in claim 6, wherein said first object and said reference object are coins.

29. A system as claimed in claim 8, wherein said first object and said reference object are coins.

30. A system as claimed in claim 17, wherein said first object and said reference object are coins.

31. A system as claimed in claim 19, wherein said first object and said reference object are coins.

32. A system as claimed in claim 24, wherein said first object and said reference object are coins.

33. A system as claimed in claim 26, wherein said first object and said reference object are coins.

9. A system as claimed 8, wherein each of said respective means for linearly moving said X and Y tables comprises a stepper motor.

27. An image capture system, comprising:

translation means for moving a first object in two dimensions, said two dimensions forming a plane;

rotation means for rotating said first object within said plane;

means including a camera for forming a first electronic image of said first object;

means for storing and retrieving a first electronic image of a reference object, said reference object being of the same type of object as said first object;

means for locating a distinct landmark on said first object;

location comparison means for comparing a location of said distinct landmark with a location of a corresponding landmark on said reference object;

orientation comparison means for comparing an orientation of said landmark with an orientation of said corresponding landmark;

translation and orientation control means for causing said translation means to move said first object a distance corresponding to a difference in locations of said first object landmark and said reference object landmark and for rotating said first object through an angle corresponding to an angular difference in orientations of said two landmarks;

means for causing said camera to form a second electronic image of said first object such that said second image of said first object is aligned with said first image of said reference object, said alignment resulting from movement of said object by said rotation means and said translation means.

wherein said first object and said reference object are coins.

34. A method of capturing an image of an object, comprising the steps of:

a. forming an image of a first object;

b. locating a distinct landmark in said image of said first object;

c. retrieving a stored image of a reference object of the same type as said first object;

d. comparing a location and orientation of said first object landmark with a location and orientation of the corresponding landmark on said reference object;

e. moving said first object a distance corresponding to a difference in locations of said first and reference object landmarks;

f. rotating said first object about a center of said first object landmark through an angle corresponding to an angular offset between a second landmark on said first object and a second landmark on said reference object;

g. forming a second image of said first object, said second image being aligned with said image of said reference object as a result of said movement and rotation of said first object.

35. A method as claimed in claim 34, wherein step (a) comprises the step of converting said image of a first object into a digital image.

36. A method as claimed in claim 34, wherein step (e) comprises the step of moving said first object in a first direction and a second direction perpendicular to said first direction.

37. A method as claimed in claim 34, further comprising the steps of providing a holder for a plurality of objects and capturing images of said plurality of objects in sequence, one side at a time.

38. A method as claimed in claim 34, wherein step (f) comprises the step of rotating said object through an angle θ

equal to the difference in orientation between said first and second object landmarks, and step (e) comprises the step of moving said first object over mutually perpendicular distances Δ

X and Δ

Y determined by the formulas

space="preserve" listing-type="equation">Δ

X=X.sub.0 cos θ

-Y.sub.0 sin θ

-X.sub.0

space="preserve" listing-type="equation">Δ

Y=X.sub.0 sin θ

+Y.sub.0 cos θ

-Y.sub.0

where X₀ and Y₀ are the initial distances of the object along said mutually perpendicular axes from the axis of rotation about which said first object is rotated.

39. A method as claimed in claim 34, wherein steps (e) and (f) comprises the steps of moving said object incrementally in increments of movement which are smaller than 1 pixel of said image of said first object.

40. A method as claimed in claim 39, wherein steps e and f further comprise the steps of creating a bi-quadratic error surface from the correlation value of the optimum alignment pixel point and the eight surrounding pixels, and subsequently computing the location of the minimum value of a continuous version of the correlation function and thus the floating point angular offset.

41. A method as claimed in claim 39, further comprising the steps of repeating each of said object movements in an iterative process until alignment is achieved.

42. A method as claimed in claim 38, further comprising the steps of repeating each of said object movements in an iterative process until alignment is achieved.

43. A method as claimed in claim 34, further comprising the step of varying an angle of illumination by which said object is illuminated during formation of said image by selectively switching on at least one of a plurality of different individual lights while selectively leaving at least one of said plurality of different individual lights off.

44. A method as claimed in claim 43, wherein step (g) is repeated under a plurality of different lighting conditions while said first object remains stationary.

45. A method as claimed in claim 34, further comprising the step of initializing the center of rotation around which said object is rotated by

first roughly defining the center by touching a display screen showing an object positioning table on which the object is positioned at a point which is somewhat near the center of rotation of the table;

defining a better center point by measuring the motion of a point on the screen near the center of the table as the table is rotated through 180°

.

and finally defining the precise center of rotation by measuring the motion of a point on the outside edge of the table as the table rotates through 180°

.

46. A method as claimed in claim 45, further comprising the step of aligning an X-axis of a first object positioning motor, as depicted on a display screen, with an X-axis in the screen system by

first approximately aligning the respective axes by following the motion of a point moving from one side of the screen to the other;

and

subsequently following the motion of an object positioning table on which the object is located through the full range of translation through which the table can be translated, followed by a 180°

rotation about the precise center of rotation to bring the target point back into view of the camera.

47. A method as claimed in claim 46, further comprising the step of calibrating motor translation to displayed image motion by determining the amount of pixels traversed on the screen for a given movement of the table.