Interpolated image filtering method and apparatus

• US 20040008901A1
• Filed: 07/11/2002
• Published: 01/15/2004
• Est. Priority Date: 07/11/2002
• Status: Active Grant

• Current Assignee: GE Medical Systems Global Technology Company LLC (GE Healthcare Technologies, Inc.)
• Original Assignee: GE Medical Systems Global Technology Company LLC (GE Healthcare Technologies, Inc.)
• Inventors: Avinash, Gopal B.
• Granted Patent: US 7,599,579 B2
• Time in Patent Office: Days
• US Class Current: 382/260
• CPC Class Codes: G06T 2207/10088 Magnetic resonance imaging [MRI]
• CPC Class Codes: G06T 2207/20012 Locally adaptive
• CPC Class Codes: G06T 2207/20192 Edge enhancement; Edge preservation
• CPC Class Codes: G06T 3/4007 based on interpolation, e.g. bilinear interpolation image demosaicing G06T3/4015; edge-driven or edge-based scaling G06T3/403
• CPC Class Codes: G06T 5/002 Denoising; Smoothing noise processing or correction adapted to be used in an image pickup device containing and electronic image sensor H04N23/81, H04N25/60 - H04N25/67
• CPC Class Codes: G06T 5/003 Deblurring; Sharpening vibration or motion blur correction for cameras comprising an electronic image sensor H04N23/682
• CPC Class Codes: G06T 5/20 using local operators
• CPC Class Codes: G06T 5/70 Denoising; Smoothing
• CPC Class Codes: G06T 5/73 Deblurring; Sharpening
• CPC Class Codes: G06T 7/13 Edge detection

Abstract

A technique for processing a digital image in which an operator selection of a field of view triggers the automatic selection of a shrink factor, so that image processing and quality is uniform regardless of the operator selection. Processing of the image includes shrinking the image by the shrink factor, identifying structural elements within the image, differentially processing the structure and non-structure elements, expanding the image by the shrink factor, and texture blending the processed image with the original image. Optionally, high frequency noise may also be blended with the processed image to improve perceived image quality.

Claims

1. A method for processing a digital image, the method comprising the steps of:

setting a shrink factor based upon an operator selected field of view;

shrinking the image by the shrink factor;

processing the image to selectively enhance one or more selected regions;

expanding the image by the shrink factor; and

blending one or more portions of the expanded image with one or more corresponding portions of the initial image.

2. The method of claim 1, wherein setting the shrink factor is performed automatically.

3. The method of claim 1, further comprising the step of blending a distribution of high frequency noise with the image.

4. The method of claim 3, wherein the distribution of high frequency noise is derived from a uniform distribution of random numbers.

5. The method of claim 3, wherein the blending of the distribution of high frequency noise is performed differentially with regard to the one or more selected regions.

6. The method of claim 1, wherein the step of shrinking is accomplished by a sub-sampling technique.

7. The method of claim 1, wherein the step of shrinking is accomplished by use of a boxcar filter.

8. The method of claim 1, wherein the one or more selected regions represent one or more structure regions.

9. The method of claim 1, wherein the one or more selected regions are selected using one or more gradient thresholds.

10. The method of claim 1, further comprising pre-processing the image prior to shrinking such that the one or more edges of the image are padded to facilitate shrinking.

11. The method of claim 1, wherein the processing of selected regions comprises smoothing pixels based upon a dominant orientation and an orientation orthogonal to the dominant orientation.

12. The method of claim 1, wherein the processing of selected regions comprises selectively applying directional sharpening of pixels in the selected region.

13. The method of claim 1, wherein expanding the image is accomplished using an interpolation technique.

14. The method of claim 1, wherein the blending of one or more selected regions with the one or more corresponding regions is differential such that different selected regions may be blended in different proportions.

15. A method for processing a digital image, the method comprising the steps of:

automatically specifying a shrink factor based upon a field of view selected by an operator;

shrinking the image by the shrink factor;

processing the image; and

expanding the image by the shrink factor.

16. The method of claim 15, wherein the shrink factor is set to 1 for images sized less than or equal to 256×

256, 2 for images sized greater than 256×

256 but less than or equal to 512×

512, and 4 for all images sized greater than 512×

512.

17. The method of claim 15, further comprising blending the image with a distribution of high frequency noise.

18. The method of claim 17, wherein the distribution of high frequency noise is uniformly distributed.

19. The method of claim 17, wherein blending is performed differentially such that one or more selected regions of the image are blended with noise at a different proportion than the remainder of the image.

20. The method of claim 17, wherein the amount of high frequency noise blended with a pixel is less than five percent of the intensity of the pixel.

21. The method of claim 15, wherein shrinking is accomplished by use of a boxcar filter.

22. The method of claim 15, wherein processing comprises enhancing one or more selected regions of the image relative to the non-selected regions.

23. The method of claim 22, wherein the one or more selected regions represent structural elements of the image.

24. The method of claim 15, wherein expanding is accomplished using an interpolation technique.

25. The method of claim 15, further comprising padding the image prior to shrinking to facilitate the shrinking process.

26. A method of improving the perceived image quality of a digital image, the method comprising:

accessing a distribution of random noise;

weighting a number drawn from the distribution by a factor, wherein the factor is determined based upon a characteristic of a pixel of the digital image; and

combining the weighted number and the pixel.

27. The method of claim 26, wherein the distribution is uniformly distributed.

28. The method of claim 26, wherein the distribution is filtered to favor high frequencies.

29. The method of claim 26, wherein the characteristic is intensity.

30. The method of claim 26, wherein the characteristic is a classification based upon a processing step.

31. The method of claim 30, wherein the classification is one of structure and non-structure.

32. A system for processing a digital image, the system comprising:

an input device for controlling image acquisition and processing;

an output device for producing a reconstructed image based upon processed image data; and

a signal processing circuit configured to provide processed image data wherein the signal processing circuit receives a selected field of view from an operator via the input device, assigns a shrink factor in response to the selected field of view, shrinks the image by the shrink factor, processes the image, and expands the processed image by the shrink factor.

33. The system of claim 32, wherein the signal processing circuit pads the image prior to shrinking the image to facilitate the shrinking process.

34. The system of claim 32, wherein the signal processing circuit processes the image by identifying one or more structural elements within the image and differentially enhancing the one or more structural elements.

35. The system of claim 32, wherein the signal processing circuit blends the image with a distribution of noise.

36. The system of claim 35, wherein the distribution is uniform.

37. The system of claim 36, wherein the distribution is filtered to substantially comprise high-frequency noise.

38. The system of claim 35, wherein the signal processing circuit blends the image and the distribution of noise differentially such that one or more regions of the image receive a greater proportion of noise.

39. A system for processing a digital image, the system comprising:

a signal processing circuit configured to access a distribution of noise;

to acquire a number from the distribution, to weight the number by a factor determined by a characteristic of a pixel of the digital image, and to combine the weighted number and the pixel.

40. The system of claim 39, wherein the distribution is uniformly distributed.

41. The system of claim 39, wherein the distribution is filtered to favor high frequencies.

42. The system of claim 39, wherein the characteristic is intensity.

43. The system of claim 39, wherein the characteristic is a classification based upon a processing step.

44. The system of claim 39, wherein the classification is one of structure and non-structure.

45. A system for processing digital images, the system comprising:

an input circuit capable of receiving signals from one or more input devices;

an output circuit capable of receiving signals from a processing circuit; and

a processing circuit capable of receiving signals from the input circuit and sending signals to the output circuit, wherein the processing circuit comprises a means to shrink a digital image in response to an input provided by an operator via the input circuit, wherein the processing circuit processes the shrunken digital image, expands the processed digital image by the same factor by which the image was shrunk, and send the expanded digital image to the output circuit.

46. The system of claim 45, wherein the processing circuit processes the shrunken digital image by identifying one or more structural elements within the image and enhancing the one or more structural elements.

47. The system of claim 45, wherein the processing circuit further possesses a means for improving perceived image quality by blending noise with the digital image.

48. The system of claim 45, wherein the processing circuit, prior to sending the expanded digital image to the output circuit, differentially blends texture from the original digital image with the expanded digital image.