Uniform and non-uniform dynamic range remapping for optimum image display

• US 5,835,618 A
• Filed: 09/27/1996
• Issued: 11/10/1998
• Est. Priority Date: 09/27/1996
• Status: Expired due to Term

• Current Assignee: Siemens Corp. (Siemens AG)
• Original Assignee: Siemens Corporate Research Incorporated (Siemens AG)
• Inventors: Fang, Ming, Qian, Jianzhong, Schramm, Helmuth
• Primary Examiner(s): ROGERS, SCOTT A
• Application Number:
  US08/722,756
• Time in Patent Office: 774 Days
• Field of Search: 382/132, 382/274, 382/275, 382/264, 382/263, 382/262, 382/261, 382/260, 382/266
• US Class Current: 382/132
• CPC Class Codes: G06T 2207/20192 Edge enhancement; Edge preservation
• CPC Class Codes: G06T 2207/30004 Biomedical image processing
• CPC Class Codes: G06T 5/40 using histogram techniques
• CPC Class Codes: G06T 5/92 based on global image properties
• CPC Class Codes: H04N 1/4072 dependent on the contents of the original

Abstract

A contexture dependent dynamic range remapping system utilizes background estimation, mask generation, parameter estimation and dynamic range remapping of an original image to provide an optimum output image. The dynamic range remapping includes an additive algorithm or a multiplicative algorithm or a combination of both algorithms. The output from the algorithms could be filtered by an edge-preserving filter to provide a filtered output image. The system adaptively compresses the dynamic range of the DC and slow-varying signals in the original image significantly while preserving and enhancing fine structures.

Claims

1. A uniform and non-uniform dynamic range remapping system for optimum image display comprising:

background estimation means for receiving an original image and for providing an estimated background image;

mask generation means for receiving an original image;

parameter estimation means for receiving an original image; and

,

dynamic range remapping means connected to said background estimation means, said mask generation means and said parameter estimation means for transmitting an output image;

wherein said dynamic range remapping means comprises;

additive algorithm means; and

means for using a minimum and maximum value of said original image to remap dynamic range of processed image to said original image.

2. A uniform and non-uniform dynamic range remapping system for optimum image display as claimed in claim 1 further comprising:

edge preserving filter means connected to said dynamic range remapping means.

3. A uniform and non-uniform dynamic range remapping system for optimum image display as claimed in claim 1 wherein:

output of said dynamic range remapping means is fed back into said background estimation means, said mask generation means and said parameter estimation means.

4. A uniform and non-uniform dynamic range remapping system for optimum image display as claimed in claim 1 wherein said background estimation means comprises:

filter means for estimating DC and slow-varying signal components from said original image therefore providing said estimated background image.

5. A uniform and non-uniform dynamic range remapping system for optimum image display as claimed in claim 1 wherein said mask generation means comprises:

means for compressing DC and slow-varying signals for providing a binary mask image.

6. A uniform and non-uniform dynamic range remapping system for optimum image display as claimed in claim 1 wherein said mask generation means comprises:

means for subtracting said original image by a portion of said estimated background image for providing a binary mask image.

7. A uniform and non-uniform dynamic range remapping system for optimum image display as claimed in claim 1 wherein said mask generation means comprises:

means for dividing said original image by a portion of said estimated background image for providing a binary mask image.

8. A uniform and non-uniform dynamic range remapping system for optimum image display as claimed in claim 1 wherein said parameter estimation means comprises:

means for computing image parameters such as mean, maximum and minimum values from said original image.

9. A uniform and non-uniform dynamic range remapping system for optimum image display as claimed in claim 1 wherein said dynamic range remapping means further comprises:

fusion means for fusing said remapped images together to form said output image.

10. A uniform and non-uniform dynamic range remapping system for optimum image display as claimed in claim 1 wherein:

said dynamic range remapping means takes out DC and slow-varying signal components from said original image independent of estimated background intensity.

11. A uniform and non-uniform dynamic range remapping system for optimum image display as claimed in claim 1 wherein said additive algorithm means comprises:

positive threshold means for deciding if said estimated background image or said original image should be used for remapping process.

12. A uniform and non-uniform dynamic range remapping system for optimum image display as claimed in claim 1 wherein:

said additive algorithm means is defined by ##EQU3## where;

θ

(x,y) represents said original image;

ν

(x,y) represents a lowpass filtered image;

β

(x,y) represents said output image;

ρ

represents mean intensity value of image;

<

α

<

1represents a user defined control parameter; and

,

Δ

1 and Δ

2 are two small positive thresholds.

13. A uniform and non-uniform dynamic range remapping system for optimum image display as claimed in claim 1 wherein said dynamic range remapping means further comprises:

multiplicative algorithm means.

14. A uniform and non-uniform dynamic range remapping system for optimum image display as claimed in claim 13 further comprising:

edge preserving filter means connected to said dynamic range remapping means.

15. A uniform and non-uniform dynamic range remapping system for optimum image display as claimed in claim 13 wherein:

output of said dynamic range remapping means is fed back into said background estimation means, said mask generation means and said parameter estimation means.

16. A uniform and non-uniform dynamic range remapping system for optimum image display as claimed in claim 13 wherein said background estimation means comprises:

filter means for estimating DC and slow-varying signal components from said original image therefore providing said estimated background image.

17. A uniform and non-uniform dynamic range remapping system for optimum image display as claimed in claim 13 wherein said parameter estimation means comprises:

means for computing image parameters such as mean, maximum and minimum values from said original image.

18. A uniform and non-uniform dynamic range remapping system for optimum image display as claimed in claim 13 wherein said dynamic range remapping means further comprises:

fusion means for fusing said remapped images together to form said output image.

19. A uniform and non-uniform dynamic range remapping system for optimum image display as claimed in claim 13 wherein:

said dynamic range remapping means takes out DC and slow-varying signal components from said original image independent of estimated background intensity.

20. A uniform and non-uniform dynamic range remapping system for optimum image display as claimed in claim 13 wherein:

said multiplicative algorithm means compresses DC and slow-varying signal components and also makes signal contrast in bright regions smaller than signal contrast in dark regions.

21. A uniform and non-uniform dynamic range remapping system for optimum image display as claimed in claim 13 wherein:

said multiplicative algorithm means is defined by ##EQU4## where;

θ

(x,y) represents said original image;

v(x,y) represents a lowpass filtered image;

β

(x,y) represents said output image;

ρ

represents a mean intensity value of image;

<

α

<

1represents a user defined control parameter; and

,

Δ

1 and Δ

2 are two small positive thresholds.

22. A uniform and non-uniform dynamic range remapping system for optimum image display comprising:

a background estimator for receiving an original image and for providing an estimated background image;

a mask generator connected to said background estimator for receiving said estimated background image and an original image;

a parameter estimator for receiving an original image; and

,

a dynamic range remapper connected to said background estimator, said mask generator and said parameter estimator for transmitting an output image;

wherein said dynamic range remapper comprises;

an additive algorithm; and

means for using a minimum and maximum value of said original image to remap dynamic range of processed image to said original image.

23. A uniform and non-uniform dynamic range remapping system for optimum image display as claimed in claim 22 further comprising:

an edge preserving filter connected to an output of said dynamic range remapper.

24. A method of performing uniform and non-uniform dynamic range remapping for optimum image display comprising the steps of:

estimating a background of an original image to provide an estimated background image;

generating a binary mask image from said original image and said estimated background image;

estimating parameters of said original image; and

,

remapping dynamic range therefore providing an output image;

wherein remapping dynamic range comprises the steps of;

utilizing a multiplicative algorithm; and

using a minimum and maximum value of said original image to remap dynamic range of processed image to said original image.

25. A method of performing uniform and non-uniform dynamic range remapping for optimum image display as claimed in claim 24 further comprising the step of:

edge preserving filtering after remapping dynamic range.

26. A uniform and non-uniform dynamic range remapping system for optimum image display comprising:

background estimation means for receiving an original image and for providing an estimated background image;

mask generation means for receiving an original image;

parameter estimation means for receiving an original image; and

dynamic range remapping means connected to said background estimation means, said mask generation means and said parameter estimation means for transmitting an output image;

wherein said dynamic range remapping means comprises;

multiplicative algorithm means; and

means for using a minimum and maximum value of said original image to remap dynamic range of processed image to said original image.

27. A uniform and non-uniform dynamic range remapping system for optimum image display as claimed in claim 26 further comprising:

edge preserving filter means connected to said dynamic range remapping means.

28. A uniform and non-uniform dynamic range remapping system for optimum image display as claimed in claim 26 wherein:

output of said dynamic range remapping means is fed back into said background estimation means, said mask generation means and said parameter estimation means.

29. A uniform and non-uniform dynamic range remapping system for optimum image display as claimed in claim 26 wherein said background estimation means comprises:

filter means for estimating DC and slow-varying signal components from said original image therefore providing said estimated background image.

30. A uniform and non-uniform dynamic range remapping system for optimum image display as claimed in claim 26 wherein said parameter estimation means comprises:

means for computing image parameters such as mean, maximum and minimum values from said original image.

31. A uniform and non-uniform dynamic range remapping system for optimum image display as claimed in claim 26 wherein said dynamic range remapping means further comprises:

fusion means for fusing said remapped images together to form said output image.

32. A uniform and non-uniform dynamic range remapping system for optimum image display as claimed in claim 26 wherein:

said dynamic range remapping means takes out DC and slow-varying signal components from said original image independent of estimated background intensity.

33. A uniform and non-uniform dynamic range remapping system for optimum image display as claimed in claim 26 wherein:

said multiplicative algorithm means compresses DC and slow-varying signal components and also makes signal contrast in bright regions smaller than signal contrast in dark regions.