Method and apparatus for correcting distortion in an imaging system

• US 5,748,768 A
• Filed: 10/07/1994
• Issued: 05/05/1998
• Est. Priority Date: 10/30/1992
• Status: Expired due to Term

• Current Assignee: Kabushiki Kaisha Toshiba (Toshiba Corporation), Toshiba Corporation
• Original Assignee: Kabushiki Kaisha Toshiba (Toshiba Corporation)
• Inventors: Drawert, Bruce M., Sivers, Erlvada A., Silver, Michael D.
• Primary Examiner(s): Mancuso, Joseph
• Assistant Examiner(s): Patel, Jayanti K.
• Application Number:
  US08/322,024
• Time in Patent Office: 1,306 Days
• Field of Search: 382/128, 382/130, 382/131, 382/132, 382/205, 382/298, 382/299, 382/300, 364/413.23, 364/413.14, 364/413.18, 364/413.19
• US Class Current: 382/130
• CPC Class Codes: G06T 5/80 Geometric correction

Abstract

A method and apparatus to calibrate and correct magnetic and geometrical distortions in an imaging system, specifically a computed tomography (CT) system, includes a distortion measuring object which is placed on the surface of an image intensifier. An image frame of the distortion measuring object is generated, and a correction table is generated to correct the image frame. Once the correction table has been generated using the distortion measuring object, subsequent image frames are corrected using the correction values in the correction tables. A method and apparatus to calibrate and correct offsets, veiling glare and scatter in an imaging system, specifically a computed tomography system. An image frame is generated and areas not directly impinged by direct radiation are used to correct areas of the detector that are directly impinged. A method and apparatus for correcting and calibrating an imaging system, specifically a computed tomography system. An image frame is corrected for distortions and for the effects of offsets, veiling glare and scatter.

Claims

1. A system for correcting spatial distortions in an imaging system for imaging an object under test, the imaging system including a distortion measuring object and having a plurality of reference point attenuators at predetermined spaced locations in said distortion measuring object, a background measuring object, said correcting system comprising:

means for generating a first image frame of said distortion measuring object with said plurality of reference point attenuators at said locations, said first image frame having representations of said reference point attenuators, said first image frame including spatial distortion;

means for generating a second image frame of said background measuring object;

means for subtracting the two generated image frames to produce a subtracted image frame including spatial distortion; and

means for removing said spatial distortion, including;

means for automatically detecting approximate two-dimensional coordinates of said reference point attenuator representations in said subtracted image frame;

means for obtaining a plurality of smooth curves in two approximately orthogonal directions having curvatures fitted to the reference point attenuator representations;

means for determining refined two-dimensional coordinates in said subtracted image frame by locating intersections of said plurality of smooth curves;

means for generating a correction table including a two-dimensional correction coordinate corresponding to each refined two-dimensional coordinate in said subtracted image frame; and

means for correcting said distortion of an image of the object under test in accordance with said correction coordinates in the correction table, and for providing a corrected image of the object under test.

2. An system according to claim 1, wherein said plurality of smooth curves is a first plurality of smooth curves, each said smooth curve parameterized by coefficients;

wherein said removing distortion means further includes;

means for fitting a second plurality of coefficient curves to the respective coefficients of said first plurality of smooth curves; and

means for determining the two-dimensional coordinates between the refined reference point attenuator representations from said second plurality of coefficient curves.

3. An system according to claim 1, wherein said plurality of reference point attenuators comprises a pattern of radiopaque objects.

4. An system according to claim 3, wherein said radiopaque objects of said distortion measuring object comprise metallic spheres.

5. The system according to claim 1, wherein said means for generating image frames of said distortion measuring object and said background measuring object includes:

means for irradiating said objects with x rays from the imaging system;

means for detecting said x rays passing through said objects; and

means for generating image data values, corresponding to said detected x rays of each of said image frames.

6. An system according to claim 1, wherein the spatial distortions are corrected by determining the correct data value for each two-dimensional refined coordinate in accordance with the corresponding two-dimensional correction coordinate.

7. An system according to claim 6, wherein each data value for the corrected image is determined according to the relationship:

##EQU11## Where;

OUTPUT(X,Y)=Intensity value at column X, row Y for corrected image;

INPUT(ix,iy)=Intensity value at column ix, row iy for original distorted image;

X=Table-value(X,Y,1) (floating point);

Y=Table-value(X,Y,2) (floating point);

ix=integer part of x;

fx=fractional part of x;

iy=integer part of y; and

fy=fractional part of y.

8. A system for correcting spatial distortions in an imaging system for imaging an object under test, said correcting system comprising:

a distortion measuring object having a plurality of reference point attenuators at predetermined spaced locations in said distortion measuring object;

means for generating an image frame of said distortion measuring object with said plurality of reference point attenuators in said locations, said image frame having representations of said reference point attenuators, said image frame including spatial distortions;

a background measuring object;

means for generating an image frame of said background measuring object;

means for subtracting the two generated image frames to produce a subtracted image frame including spatial distortion; and

means for removing said spatial distortion, including;

means for automatically detecting approximate two-dimensional coordinates of said reference point attenuator representations in said subtracted image frame;

means for fitting a plurality of smooth curves in two approximately orthogonal directions having curvatures corresponding to the reference point attenuator representations;

means for determining refined two-dimensional coordinates in said subtracted image frame by locating intersections of said plurality of smooth curves;

means for generating a correction table including a two-dimensional correction coordinate corresponding to each refined two-dimensional coordinate in said subtracted image frame; and

means for correcting said distortion of an image of the object under test in accordance with said correction coordinates in the correction table, and for providing a corrected image of the object under test.

9. A system for correcting spatial distortions in an imaging system for imaging an object under test, the imaging system including a distortion measuring object having a plurality of reference point attenuators at predetermined spaced locations in said distortion measuring object, a background measuring object, means for generating x rays and for detecting the generated x rays, means for generating a first image frame of said distortion measuring object with said plurality of reference point attenuators at said locations having representations of said reference point attenuators and including spatial distortion, means for generating a second image frame of said background measuring object, and means for storing said image frames, said correcting system comprising:

means for retrieving said image frames;

means for subtracting said image frames to produce a subtracted image frame including spatial distortion;

means for automatically determining approximate two-dimensional coordinates of said reference point attenuator representations in said subtracted image frame;

means for fitting a plurality of smooth curves to said two-dimensional coordinates of the reference point attenuator representations, each said smooth curve being parameterized by coefficients;

means for determining refined two-dimensional coordinates in said subtracted image frame by locating intersections of said smooth curves;

means for fitting a plurality of smooth curves to respective coefficients of the curves fit to said two-dimensional coordinates of said points;

means for determining dense points between the two-dimensional coordinates of the reference point attenuator representations;

means for generating a correction table including a two-dimensional correction coordinate corresponding to each refined two-dimensional coordinate in the image frame; and

means for correcting the distortion in an image frame of an object under test in accordance with said correction coordinates in the correction table, and providing for a corrected image of the object under test.

10. A method for correcting spatial distortions in an imaging system for imaging an object under test, the imaging system including a distortion measuring object having a plurality of reference point attenuators at predetermined spaced locations in said distortion measuring object, a background measuring object, means for generating x rays and for detecting the generated x rays, means for generating a first image frame of said distortion measuring object with said plurality of reference point attenuators at said locations having representations of said reference point attenuators and including spatial distortion, means for generating a second image frame of said background measuring object, and means for storing the image frames, said correcting method including the steps of:

retrieving the image frames;

subtracting said image frames to produce a subtracted image frame including spatial distortion;

automatically determining approximate two-dimensional coordinates of said reference point attenuator representations in said image frame;

fitting a plurality of smooth curves to said two-dimensional coordinates of the reference point attenuator representations, each said smooth curve being parameterized by coefficients;

determining refined two-dimensional coordinates in said subtracted image frame by locating intersections of said smooth curves;

fitting a plurality of smooth curves to respective coefficients of the curves fit to said two-dimensional coordinates of said points;

determining dense points between the two-dimensional coordinates of the reference point attenuator representations;

generating a correction table including a two-dimensional correction coordinate corresponding to each refined two-dimensional coordinate in the image frame; and

correcting the distortion in an image frame of an object under test in accordance with said correction coordinates in the correction table, and providing for a corrected image of the object under test.

11. A method for correcting spatial distortions in an imaging system for imaging an object under test, the imaging system including a distortion measuring object having a plurality of reference point attenuators at predetermined spaced locations in said distortion measuring object and a background measuring object, said correcting method including the steps of:

generating a first image frame of said distortion measuring object with said plurality of reference point attenuators at said locations, said first image frame having representations of said reference point attenuators, said first image frame including spatial distortion;

generating a second image frame of said background measuring object;

subtracting the two generated image frames to produce a subtracted image frame including spatial distortion; and

removing said spatial distortion, including;

automatically detecting approximate two-dimensional coordinates of said reference point attenuator representations in said subtracted image frame;

fitting a plurality of smooth curves in two approximately orthogonal directions having curvatures corresponding to the reference point attenuator representations;

determining refined two-dimensional coordinates in said subtracted image frame by locating intersections of said plurality of smooth curves;

generating a correction table including a two-dimensional correction coordinate corresponding to each refined two-dimensional coordinate in said subtracted image frame; and

correcting said distortion of an image of the object under test in accordance with said correction coordinates in the correction table, and providing a corrected image of the object under test.

12. The method according to claim 11, wherein said step of generating image frames of said distortion measuring object and background measuring object includes:

irradiating said objects with x rays from the imaging system;

detecting said x rays passing through said objects; and

generating of image data values, corresponding to said detected x rays of each of said image frames.

13. The method according to claim 11, wherein the spatial distortions are corrected by determining the correct data value for each two-dimensional refined coordinate in accordance with the corresponding two-dimensional correction coordinate.

14. The method according to claim 13, wherein each data value for the corrected image is determined according to the relationship:

##EQU12## Where;

OUTPUT(X,Y)=Intensity value at column X, row Y for corrected image

INPUT(ix,iy)=Intensity value at column ix, row iy for original distorted image;

X=Table-value(X,Y,1) (floating point);

Y=Table-value(X,Y,2) (floating point);

ix=integer part of x;

fx=fractional part of x;

iy=integer part of y; and

fy=fractional part of y.

15. An extraneous signals reduction system for reducing the effects of extraneous signals, including offsets, veiling glare, and/or scatter, in an imaging system, the imaging system including means for generating a beam of x rays having a predetermined beam shape and means for detecting x rays of the x-ray beam, the beam having a cross section projected on the detecting means to define a directly irradiated area of the detecting means and at least one non-irradiated area of the detecting means adjacent to the directly irradiated area, and means for generating an image frame representative of the detected x rays of the x-ray beam, the image frame having columns and rows of data values corresponding to at least one of detected x rays and extraneous signals, said extraneous signals reduction system comprises:

means for defining a first area of the image frame corresponding to the directly irradiated area of the detecting means;

means for defining at least one second area of the image frame corresponding to the at least one non-irradiated area of the detecting means; and

means for correcting the data values in the first area of the image frame in accordance with the data values from the at least one second area of the image frame.

16. The system according to claim 15, comprising:

means for locating each row in a region of interest of the image frame containing at least one data value above a predetermined threshold value and corresponding to the first area of the image frame and for locating each row in a region of interest of the image frame containing all data values below said threshold value and corresponding to the at least one second area of the image frame;

means for summing the data values in each column for the rows containing at least one data value above said threshold value to provide an image sum and for summing the data values in each column for the rows containing all data values below said threshold value to provide a extraneous signals sum; and

said means for correcting includes means for adjusting the data values in locations above the threshold value in the region of interest as a function of the image sum and the extraneous signals sum.

17. The system according to claim 15, wherein the rows with data values above a predetermined threshold value are located in a single contiguous image area in a region of interest of the image frame and corresponding to the first area of the image frame and the rows with values below said threshold value are located in a first contiguous extraneous signal area and a second contiguous extraneous signal area in a region of interest of the image frame directly adjacent to the image area and respectively on opposite sides thereof and corresponding to the at least one second area of the image frame;

wherein the means for defining a first area of the image frame includes means for determining the first area as the image area of the image frame; and

wherein the means for defining at least one second area includes;

means for determining the second area as the first extraneous signal area of the image frame; and

means for determining a third area as the second extraneous signal area of the image frame.

18. The system according to claim 17, wherein the means for removing the effects of extraneous signals from the image frame is preceded by means for removing distortion of the image wherein said means for removing distortion comprises:

means for generating a first image frame of a distortion measuring object having a plurality of reference point attenuators at predetermined space locations, said first image frame having representations of said reference point attenuators, said first image frame including spatial distortions;

means for generating a second image frame of a background measuring object;

means for subtracting the two generated image frames to produce a subtracted image frame including spatial distortion; and

means for removing said spatial distortion, including;

means for automatically detecting approximate two-dimensional coordinates of said reference point attenuator representations in said subtracted image frame;

means for fitting a plurality of smooth curves in two approximately orthogonal directions having curvatures corresponding to the reference point attenuator representations;

means for determining refined two-dimensional coordinates in said image frame by locating intersections of said plurality of smooth curves;

means for generating a correction table including a two-dimensional correction coordinate corresponding to each refined two-dimensional coordinate in the image frame; and

means for correcting said distortion of an image of the object under test in accordance with said correction coordinates in the correction table, and for providing a corrected image frame of the object under test.

19. The system according to claim 17, comprising:

means for locating each row in a region of interest of the image frame containing at least one data value above a predetermined threshold value and corresponding to the first area of the image frame and for locating each row in a region of interest of the image frame containing all data values below the certain threshold value and corresponding to the at least one second area of the image frame;

means for summing the data values in the first area of the image frame to provide an image sum and for summing the data values in each of the second and third areas to provide two extraneous signal sums; and

said means for correcting includes means for adjusting the data values in locations above the threshold value in the region of interest as a function of the image sum and the extraneous signals sums, and

wherein the extraneous signal sums are determined according to the following relationships;

##EQU13## where S₁ (i,k) is extraneous signal sum of the second area;

S₃ (i,k) is extraneous signal sum of the third area;

R (i,j,k) labels the data values of each generated image frame;

i labels the image frame;

j labels each row of data values;

k labels each column of data values;

j₁ is the row at one edge of the second area;

j₂ is the row dividing the first and second areas;

j₃ is the row dividing the first and the third area; and

j₄ is the row at the other edge of the third area.

20. The system according to claim 19, wherein the image sum is determined according to the following relationship:

##EQU14## where S₂ (i,k) is image sum;

R (i,j,k) labels the data values of each generated image frame;

i labels the image frame;

j labels each row of data values;

k labels each column of data values;

j₂ is the row at one edge of the first area; and

j₃ is the row at the other edge of the first area.

21. The system according to claim 20, wherein the imaging system includes means for generating a video signal which has a front porch region, a video region, and a back porch region;

and wherein the noise reduction system determines the adjusted data values in the first area according to the following relationship;

##EQU15## where R'"'"'(i,k) labels the corrected data values in the first area of image frame;

i labels the image frame;

j labels the image line;

k labels the image column;

S_I (i,k) is sum of data values in the first area;

S₁ (i,k) is sum of data values in the second area;

S₂ (i, k) is sum of data values in the third area;

S₁ °

(i) is the average of the front porch region of the video signal for each row of the second area;

S₂ °

(i) is the average of the front porch region of the video signal for each row in the third area;

j₁ is the row at one edge of the second area;

j₂ is the row dividing the first and second areas;

j₃ is the row dividing the first and the third area; and

j₄ is the row at the other edge of the third area.

22. The system according to claim 17, wherein sequential column numbers and sequential row numbers are respectively assigned to the columns and rows of data values, wherein said means for defining a first area further includes:

means for identifying a plurality of rows in each column of said image frame with a data value above said threshold value;

means for identifying in each column the largest and the smallest row number from said plurality of rows in each column and for defining said first area as a function of the largest and smallest row number.

23. The system according to claim 22, wherein:

said means for determining said second area includes means for defining said second area as a function of the smallest row number in each column decremented by a first predetermined number of rows; and

said means for determining said third area includes means for defining said third area as a function of the largest row number in each column incremented by a second predetermined number of rows.

24. The system according to claim 15, wherein the imaging system is for imaging an object under test, the imaging system further including a distortion measuring object having a plurality of reference point attenuators at predetermined spaced locations in said distortion measuring object and a background measuring object, said extraneous signals reduction system further including:

means for generating a first image frame of said background measuring object with said plurality of reference point attenuators in said locations, said first image frame having representations of said reference point attenuators, said first image frame including spatial distortions;

means for generating a second image frame of said background measuring object;

means for subtracting the two generated image frames to produce a subtracted image frame including spatial distortion;

means for removing said spatial distortion, including;

means for automatically detecting approximate two-dimensional coordinates of said reference point attenuator representations in said subtracted image frame;

means for fitting a plurality of smooth curves in two approximately orthogonal directions having curvatures corresponding to the reference point attenuator representations;

means for determining refined two-dimensional coordinates in said image frame by locating intersections of said plurality of smooth curves;

means for generating a correction table including a two-dimensional correction coordinate corresponding to each refined two-dimensional coordinate in the image frame; and

means for correcting said distortion of an image of the object under test in accordance with said correction coordinates in the correction table, and for providing a corrected image frame of the object under test; and

said means for correcting the data values including means for correcting the data values of the corrected image frame.

25. A method for reducing the effects of extraneous signals, including offsets, veiling glare, and/or scatter in an imaging system, the imaging system including means for generating a beam of x rays, the beam having a cross section projected on the detecting means to define a directly irradiated area of the detecting means and at least one non-irradiated area of the detecting means adjacent to the directly irradiated area, and means for generating an image frame representative of the detected x rays of the x-ray beam, the image frame having columns and rows of data values corresponding to at least one of detected x rays and extraneous signals, said reducing method including the steps of:

defining a first area of the image frame corresponding to the directly irradiated area of the detecting means;

defining at least one second area of the image frame corresponding to the at least one non-irradiated area of the detecting means; and

correcting the data values in the first area of the image frame in accordance with the data values from the at least one second area of the image frame.

26. The method according to claim 25, including the steps of:

locating each row in a region of interest of the image frame containing at least one data value above a predetermined threshold value and locating each row in a region of interest of the image frame containing all data values below the certain threshold value;

summing the data values in each column for the rows containing at least one data value above the threshold value to provide an image sum and summing the data values in each column for the rows containing all data values below the threshold value to provide a extraneous signal sum; and

said step of correcting includes adjusting the data values in locations above the threshold value in the region of interest as a function of the image sum and the extraneous signal sum.

27. The method according to claim 25, wherein the rows with data values above a predetermined threshold value are located in a single contiguous image area in a region of interest of the image frame and the rows with values below the threshold value are located in first and second contiguous extraneous signal areas in a region of interest of the image frame directly adjacent to the image area and respectively on opposite sides thereof;

wherein the step of defining a first area of the image frame includes means for determining the first area as the image area of the image frame;

wherein the step of defining at least one second area includes;

determining the second area as the first extraneous signal area of the image frame; and

determining a third area as the second extraneous signal area of the image frame.

28. The method according to claim 27, wherein the steps of removing the effects of extraneous signals from the image frame is preceded by a method of removing distortion of the image wherein said method includes the steps of:

generating a first image frame of a distortion measuring object with a plurality of reference point attenuators at predetermined spaced locations, said first image frame having representations of said reference point attenuators, said first image frame including spatial distortions;

generating a second image frame of a background measuring object;

subtracting the two generated image frames to produce a subtracted image frame including spatial distortion; and

removing said spatial distortion, including;

automatically detecting approximate two-dimensional coordinates of said reference point attenuator representations in said subtracted image frame;

fitting a plurality of smooth curves in two approximately orthogonal directions having curvatures corresponding to the reference point attenuator representations;

determining refined two-dimensional coordinates in said image frame by locating intersections of said plurality of smooth curves;

generating a correction table including a two-dimensional correction coordinate corresponding to each refined two-dimensional coordinate in the image frame; and

correcting said distortion of an image of the object under test in accordance with said correction coordinates in the correction table, and providing a corrected image frame of the object under test.

29. The method according to claim 27, wherein sequential column numbers and sequential row numbers are respectively assigned to the columns and rows of data values, wherein said step of defining a first area further includes:

identifying a plurality of rows in each column of said image frame with a data value above the threshold value;

identifying the largest and the smallest row number from said plurality of rows in each column and defining the first area as a function of the largest and smallest row number.

30. The method according to claim 29, wherein:

said step of determining the second area includes defining the second area as a function of the smallest row number in each column decremented by a first predetermined number of rows; and

said step of determining a third area includes defining the third area as a function of the largest row number in each column incremented by a second predetermined number of rows.

31. The method according to claim 25, wherein the imaging system is for imaging an object under test, the imaging system further including a distortion measuring object having a plurality of reference point attenuators at predetermined spaced locations in said distortion measuring object and a background measuring object, said method of reducing extraneous signals further includes the steps of:

generating a first image frame of said distortion measuring object with said plurality of reference point attenuators in said locations, said first image frame having representations of said reference point attenuators, said first image frame including spatial distortions;

generating a second image frame of said background measuring object without said plurality of removable reference point attenuators;

subtracting the two generated image frames to produce a subtracted image frame including spatial distortion; and

removing said spatial distortion including;

automatically detecting approximate two-dimensional coordinates of said reference point attenuator representations in said subtracted image frame;

fitting a plurality of smooth curves in two approximately orthogonal directions having curvatures corresponding to the reference point attenuator representations;

determining refined two-dimensional coordinates in said image frame by locating intersections of said plurality of smooth curves;

generating a correction table including a two-dimensional correction coordinate corresponding to each refined two-dimensional coordinate in the image frame; and

correcting said distortion of an image of the object under test in accordance with said correction coordinates in the correction table, and for providing a corrected image frame of the object under test; and

correcting the data values including means for correcting the data values of the corrected image frame.