Machine vision methods using feedback to determine an orientation, pixel width and pixel height of a field of view

• US 5,978,080 A
• Filed: 09/25/1997
• Issued: 11/02/1999
• Est. Priority Date: 09/25/1997
• Status: Expired due to Term

• Current Assignee: Cognex Corporation
• Original Assignee: Cognex Corporation
• Inventors: Michael, David J., Wallack, Aaron S.
• Primary Examiner(s): Font, Frank G.
• Assistant Examiner(s): STAFIRA, MICHAEL PATRICK
• Application Number:
  US08/938,443
• Time in Patent Office: 768 Days
• Field of Search: 356/243, 356/375, 356/376, 356/388, 356/390, 358/106, 358/107, 358/101, 324/158 F, 382/8, 382/26, 382/18, 382/45, 382/152, 382/154, 364/474.34, 364/559, 364/560, 364/570, 364/571.08, 364/579, 364/580, 345/156, 345/158, 348/135, 348/136, 348/142, 348/143
• US Class Current: 356/243.1
• CPC Class Codes: G01B 11/002 for measuring two or more coordinates

Abstract

The invention provides machine vision methods and apparatus for determining a calibration relationship between the imaging reference frame of an image acquisition device (e.g., a camera) and the motion reference frame of a moveable object (e.g., a motion stage) by positioning the object so that a fiducial thereon (e.g., a calibration mark) lies at a first fiducial position in the field of view and recording the corresponding position of the object in the motion reference frame. The object is then repositioned so that the fiducial moves to a second fiducial position in the field of view. Again, the corresponding position of the object in the motion reference frame is recorded. Likewise, the object is moved to a third fiducial position in the field of view, and its corresponding position in the motion reference frame is recorded. A calibration relationship between the motion reference frame of the object and the imaging reference frame of the camera as a function of the first, second and third fiducial positions and as a function of the first, second and third object positions.

Claims

1. A machine vision method of determining a calibration relationship between a motion reference frame of a moveable object and an imaging reference frame of an imaging device that acquires an image of the object, the method comprising the steps of:

A. positioning the object so that a fiducial thereon is at a first specified position with respect to the imaging reference frame and recording a position of the object with respect to the motion reference frame, that recorded position being referred to as a first object position;

B. positioning the object so that a fiducial thereon is at a second specified position with respect to the imaging reference frame and recording a position of the object with respect to the motion reference frame, that recorded position being referred to as a second object position;

C. positioning the object so that a fiducial thereon is at a third specified position with respect to the imaging reference frame and recording a position of the object with respect to the motion reference frame, that recorded position being referred to as a third object position; and

D. determining a calibration relationship between the reference frame of motion of the object and the reference frame of the camera as a function of respective motion reference frame coordinates of the first, second, and third object positions.

2. A method according to claim 1, wherein the second and third specified positions are substantially equidistant from the first specified position with respect to the imaging reference frame.

3. A method according to claim 2, wherein the second and third specified positions are substantially a unit distance from the first specified position with respect to the imaging reference frame.

4. A method according to claim 1, wherein line segments connecting the first specified position and each of the two specified positions are perpendicular to one another.

5. A method according to claim 1, wherein

i. the first, second and third specified positions lie substantially on axes of the imaging reference frame;

ii. the first specified position substantially lying on a center of the imaging reference frame; and

iii. the second and third specified positions substantially lying on respective axes of the imaging reference frame.

6. A method according to claim 5, wherein the second and third specified positions each lie substantially k units from the first specified position with respect to the imaging reference frame.

7. A method according to claim 6, step (D) includes the step of determining τ

, an orientation of the imaging reference frame with respect to the motion reference frame, based substantially on a mathematical relation;

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

=atan2 (m.sub.y1 -m.sub.y0,m.sub.x1 -m.sub.x0)

where

(m_x0,m_y0) are the motion reference frame coordinates of the first object position;

(m_x1,m_y1) are the motion reference frame coordinates of the second object position;

atan2 ( ) is an arctangent function.

8. A method according to claim 6, step (D) includes the step of determining at least one of a pixel width, w, and a pixel height, h, of the image frame with respect to the motion reference frame in accord with a mathematical relation:

space="preserve" listing-type="equation">w=sqrt ((m.sub.y1 -m.sub.y0).sup.2 +(m.sub.x1 -m.sub.x0).sup.2)/abs (i.sub.x1 -i.sub.x0)

space="preserve" listing-type="equation">h=sqrt ((m.sub.y2 -m.sub.y0).sup.2 +(m.sub.x2 -m.sub.x0).sup.2)/abs (i.sub.y2 -i.sub.y0)

where,

w is the unit height of the imaging reference frame vis-a-vis the motion reference frame;

h is the unit height of the imaging reference frame vis-a-vis the motion reference frame;

(m_x0,m_y0) are the motion reference frame coordinates of the first object position;

(m_x1,m_y1) are the motion reference frame coordinates of the second object position;

(m_x2,m_y2) are the motion reference frame coordinates of the third object position;

(i_x0, i_y0) are the image coordinates of the first object position;

(i_x1, i_y1) are the image coordinates of the second object position;

(i_x2, i_y2) are the image coordinates of the second object position;

sqrt ( ) is a square root function; and

abs ( ) is an absolute value function.

9. A machine vision method of determining a calibration relationship between a motion reference frame of a moveable object and an imaging reference frame of an imaging device that acquires an image of the object, the method comprising the steps of:

A. positioning the object so that a fiducial thereon moves to each of three specified positions with respect to the imaging reference frame and recording, for each of those specified positions, a corresponding position of the object with respect to the motion reference frame;

wherein the specified positions lie substantially on axes of the imaging reference frame; and

wherein a first specified position substantially lies on a center of the imaging reference frame; and

wherein second and third specified positions substantially lie on respective axes of the imaging reference frame;

B. determining a calibration relationship between the reference frame of motion of the object and the reference frame of the camera in accord with any of the following mathematical relations;

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

=atan2 (m.sub.y1 -m.sub.y0,m.sub.x1 -m.sub.x0)

space="preserve" listing-type="equation">w=sqrt ((m.sub.y1 -m.sub.y0).sup.2 +(m.sub.x1 -m.sub.x0).sup.2)/abs (i.sub.x1 -i.sub.x0)

space="preserve" listing-type="equation">h=sqrt ((m.sub.y2 -m.sub.y0).sup.2 +(m.sub.x2 -m.sub.x0).sup.2)/abs (i.sub.y2 -i.sub.y0)

where

τ

is an angular orientation of the imaging reference frame with respect to the motion reference frame;

h is a pixel height in the imaging reference frame with respect to the motion reference frame;

w is a pixel width in the imaging reference frame with respect to the motion reference frame;

(m_x0,m_y0) are the motion reference frame coordinates of the first object position;

(m_x1,m_y1) are the motion reference frame coordinates of the second object position;

(m_x2,m_y2) are the motion reference frame coordinates of the third object position;

atan2 ( ) is an arctangent function;

(i_x0, i_y0) are the image coordinates of the first object position;

(i_x1, i_y1) are the image coordinates of the second object position;

(i_x2, i_y2) are the image coordinates of the second object position;

sqrt ( ) is a square root function; and

abs ( ) is an absolute value function.

10. A machine vision apparatus for determining a calibration relationship between a motion reference frame of moveable object and an imaging reference frame of an imaging device that acquires an image of the object, the apparatus comprising:

A. first positioning means, coupled to the object and to the imaging device, for positioning the object so that a fiducial thereon is at a first specified position with respect to the imaging reference frame and recording a position of the object with respect to the motion reference frame, that recorded position being referred to as a first object position;

B. second positioning means, coupled to the object and to the imaging device, for positioning the object so that a fiducial thereon is at a second specified position with respect to the imaging reference frame and recording a position of the object with respect to the motion reference frame, that recorded position being referred to as a second object position;

C. third positioning means, coupled to the object and to the imaging device, positioning the object so that a fiducial thereon is at a third specified position with respect to the imaging reference frame and recording a position of the object with respect to the motion reference frame, that recorded position being referred to as a third object position; and

D. analysis means, coupled to the first, second and third positioning means, for determining a calibration relationship between the reference frame of motion of the object and the reference frame of the camera as a function of respective motion reference frame coordinates of the first, second, and third object positions.

11. An apparatus according to claim 10, wherein the second and third specified positions are substantially equidistant from the first specified position with respect to the imaging reference frame.

12. An apparatus according to claim 11, wherein the second and third specified positions are substantially a unit distance from the first specified position with respect to the imaging reference frame.

13. An apparatus according to claim 10, wherein line segments connecting the first specified position and each of the two specified positions are perpendicular to one another.

14. An apparatus according to claim 10, wherein

i. the first, second and third specified positions lie substantially on axes of the imaging reference frame;

ii. the first specified position substantially lying on a center of the imaging reference frame; and

iii. the second and third specified positions substantially lying on respective axes of the imaging reference frame.

15. An apparatus according to claim 14, wherein the second and third specified positions each lie substantially k units from the first specified position with respect to the imaging reference frame.

16. An apparatus according to claim 15, wherein the analysis means determines τ

, an orientation of the imaging reference frame with respect to the motion reference frame, based substantially on a mathematical relation;

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

=atan2 (m.sub.y1 -m.sub.y0,m.sub.x1 -m.sub.x0)

where

(m_x0,m_y0) are the motion reference frame coordinates of the first object position;

(m_x1,m_y1) are the motion reference frame coordinates of the second object position;

atan2 ( ) is an arctangent function.

17. An apparatus according to claim 15, wherein the analysis means determines at least one of a pixel width, w, and a pixel height, h, of the image frame with respect to the motion reference frame in accord with a mathematical relation:

space="preserve" listing-type="equation">w=sqrt ((m.sub.y1 -m.sub.y0).sup.2 +(m.sub.x1 -m.sub.x0).sup.2)/abs (i.sub.x1 -i.sub.x0)

space="preserve" listing-type="equation">h=sqrt ((m.sub.y2 -m.sub.y0).sup.2 +(m.sub.x2 -m.sub.x0).sup.2)/abs (i.sub.y2 -i.sub.y0)

where,

w is the unit height of the imaging reference frame vis-a-vis the motion reference frame;

h is the unit height of the imaging reference frame vis-a-vis the motion reference frame;

(m_x0,m_y0) are the motion reference frame coordinates of the first object position;

(m_x1,m_y1) are the motion reference frame coordinates of the second object position;

(m_x2,m_y2) are the motion reference frame coordinates of the third object position;

(i_x0, i_y0) are the image coordinates of the first object position;

(i_x1, i_y1) are the image coordinates of the second object position;

(i_x2, i_y2) are the image coordinates of the second object position;

sqrt ( ) is a square root function; and

abs ( ) is an absolute value function.

18. A machine vision apparatus for determining a calibration relationship between a motion reference frame of a moveable object and an imaging reference frame of an image acquisition device that acquires an image of the object, the apparatus comprising:

A. positioning means, coupled to the object and to the image acquisition device, for positioning the object so that a fiducial thereon moves to each of three specified positions with respect to the imaging reference frame and recording, for each of those specified positions, a corresponding position of the object with respect to the motion reference frame;

wherein the specified positions lie substantially on axes of the imaging reference frame; and

wherein a first specified position substantially lies on a center of the imaging reference frame; and

wherein second and third specified positions substantially lie on respective axes of the imaging reference frame;

B. analysis means for determining a calibration relationship between the reference frame of motion of the object and the reference frame of the camera in accord with any of the following mathematical relations;

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

=atan2 (m.sub.y1 -m.sub.y0,m.sub.x1 -m.sub.x0)

space="preserve" listing-type="equation">w=sqrt ((m.sub.y1 -m.sub.y0).sup.2 +(m.sub.x1 -m.sub.x0).sup.2)/abs (i.sub.x1 -i.sub.x0)

space="preserve" listing-type="equation">h=sqrt ((m.sub.y2 -m.sub.y0).sup.2 +(m.sub.x2 -m.sub.x0).sup.2)/abs (i.sub.y2 -i.sub.y0)

where

τ

is an angular orientation of the imaging reference frame with respect to the motion reference frame;

h is a pixel height in the imaging reference frame with respect to the motion reference frame;

w is a pixel width in the imaging reference frame with respect to the motion reference frame;

(m_x0,m_y0) are the motion reference frame coordinates of the first object position;

(m_x1,m_y1) are the motion reference frame coordinates of the second object position;

(m_x2,m_y2) are the motion reference frame coordinates of the third object position;

atan2 ( ) is an arctangent function;

(i_x0, i_y0) are the image coordinates of the first object position;

(i_x1, i_y1) are the image coordinates of the second object position;

(i_x2, i_y2) are the image coordinates of the second object position;

sqrt ( ) is a square root function; and

abs ( ) is an absolute value function.

19. An article of manufacture comprising a computer usable medium embodying program code for causing a digital data processor to carry out a method of determining a calibration relationship between a motion reference frame of a moveable object and an imaging reference frame of an imaging device that acquires an image of the object, the method comprising the steps of:

A. positioning the object so that a fiducial thereon is at a first specified position with respect to the imaging reference frame and recording a position of the object with respect to the motion reference frame, that recorded position being referred to as a first object position;

B. positioning the object so that a fiducial thereon is at a second specified position with respect to the imaging reference frame and recording a position of the object with respect to the motion reference frame, that recorded position being referred to as a second object position;

C. positioning the object so that a fiducial thereon is at a third specified position with respect to the imaging reference frame and recording a position of the object with respect to the motion reference frame, that recorded position being referred to as a third object position; and

D. determining a calibration relationship between the reference frame of motion of the object and the reference frame of the camera as a function of respective motion reference frame coordinates of the first, second, and third object positions.

20. An article of manufacture according to claim 19, wherein the second and third specified positions are substantially equidistant from the first specified position with respect to the imaging reference frame.

21. An article of manufacture according to claim 20, wherein the second and third specified positions are substantially a unit distance from the first specified position with respect to the imaging reference frame.

22. An article of manufacture according to claim 19, wherein line segments connecting the first specified position and each of the two specified positions are perpendicular to one another.

23. An article of manufacture according to claim 19, wherein

i. the first, second and third specified positions lie substantially on axes of the imaging reference frame;

ii. the first specified position substantially lying on a center of the imaging reference frame; and

iii. the second and third specified positions substantially lying on respective axes of the imaging reference frame.

24. An article of manufacture according to claim 23, wherein the second and third specified positions each lie substantially k units from the first specified position with respect to the imaging reference frame.

25. An article of manufacture according to claim 24, step (D) includes the step of determining τ

, an orientation of the imaging reference frame with respect to the motion reference frame, based substantially on a mathematical relation;

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

=atan2 (m.sub.y1 -m.sub.y0,m.sub.x1 -m.sub.x0)

where

(m_x0,m_y0) are the motion reference frame coordinates of the first object position;

(m_x1,m_y1) are the motion reference frame coordinates of the second object position;

atan2 ( ) is an arctangent function.

26. An article of manufacture according to claim 24, step (D) includes the step of determining at least one of a pixel width, w, and a pixel height, h, of the image frame with respect to the motion reference frame in accord with a mathematical relation:

space="preserve" listing-type="equation">w=sqrt ((m.sub.y1 -m.sub.y0).sup.2 +(m.sub.x1 -m.sub.x0).sup.2)/abs (i.sub.x1 -i.sub.x0)

space="preserve" listing-type="equation">h=sqrt ((m.sub.y2 -m.sub.y0).sup.2 +(m.sub.x2 -m.sub.x0).sup.2)/abs (i.sub.y2 -i.sub.y0)

where,

w is the unit height of the imaging reference frame vis-a-vis the motion reference frame;

h is the unit height of the imaging reference frame vis-a-vis the motion reference frame;

(m_x0,m_y0) are the motion reference frame coordinates of the first object position;

(m_x1,m_y1) are the motion reference frame coordinates of the second object position;

(m_x2,m_y2) are the motion reference frame coordinates of the third object position;

(i_x0, i_y0) are the image coordinates of the first object position;

(i_x1, i_y1) are the image coordinates of the second object position;

(i_x2, i_y2) are the image coordinates of the second object position;

sqrt ( ) is a square root function; and

abs ( ) is an absolute value function.

27. An article of manufacture of determining a calibration relationship between a motion reference frame of a moveable object and an imaging reference frame of an imaging device that acquires an image of the object, the method comprising the steps of:

A. positioning the object so that a fiducial thereon moves to each of three specified positions with respect to the imaging reference frame and recording, for each of those specified positions, a corresponding position of the object with respect to the motion reference frame;

wherein the specified positions lie substantially on axes of the imaging reference frame; and

wherein a first specified position substantially lies on a center of the imaging reference frame; and

wherein second and third specified positions substantially lie on respective axes of the imaging reference frame;

B. determining a calibration relationship between the reference frame of motion of the object and the reference frame of the camera in accord with any of the following mathematical relations;

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

=atan2 (m.sub.y1 -m.sub.y0,m.sub.x1 -m.sub.x0)

space="preserve" listing-type="equation">w=sqrt ((m.sub.y1 -m.sub.y0).sup.2 +(m.sub.x1 -m.sub.x0).sup.2)/abs (i.sub.x1 -i.sub.x0)

space="preserve" listing-type="equation">h=sqrt ((m.sub.y2 -m.sub.y0).sup.2 +(m.sub.x2 -m.sub.x0).sup.2)/abs (i.sub.y2 -i.sub.y0)

where

τ

is an angular orientation of the imaging reference frame with respect to the motion reference frame;

h is a pixel height in the imaging reference frame with respect to the motion reference frame;

w is a pixel width in the imaging reference frame with respect to the motion reference frame;

(m_x0,m_y0) are the motion reference frame coordinates of the first object position;

(m_x1,m_y1) are the motion reference frame coordinates of the second object position;

(m_x2,m_y2) are the motion reference frame coordinates of the third object position;

atan2 ( ) is an arctangent function;

(i_x0, i_y0) are the image coordinates of the first object position;

(i_x1, i_y1) are the image coordinates of the second object position;

(i_x2, i_y2) are the image coordinates of the second object position;

sqrt ( ) is a square root function; and

abs ( ) is an absolute value function.