Adjusting pixels by desired gains and factors

• US 7,426,314 B2
• Filed: 07/30/2004
• Issued: 09/16/2008
• Est. Priority Date: 07/30/2004
• Status: Expired due to Fees

• Current Assignee: Hewlett-Packard Development Company, L.P. (HP Inc.)
• Original Assignee: Hewlett-Packard Development Company, L.P. (HP Inc.)
• Inventors: Kimbell, Benjamin D., Miller, Casey L.
• Primary Examiner(s): Kassa; Yosef
• Assistant Examiner(s): Patel; Jayesh A
• Application Number:
  US10/909,214
• Publication Number:
  US 20060023965A1
• Time in Patent Office: 1,509 Days
• Field of Search: 382/54, 382/263, 382/294, 382/254, 358/284
• US Class Current: 382/260
• CPC Class Codes: H04N 1/58 Edge or detail enhancement; Noise or error suppression, e.g. colour misregistration correction H04N1/62 takes precedence

Abstract

A low frequency value for a pixel value is adjusted by a desired gain. A first high frequency value for the pixel value is calculated. A second high frequency value is generated for the first high frequency value. The second high frequency value is the first high frequency value adjusted by a factor. The factor is resolved as a function of the first high frequency value and the desired gain. A modified value for the pixel value is outputted, where the modified value is the adjusted low frequency value augmented by the second high frequency value.

Claims

1. A method comprising:

adjusting a low frequency value for a pixel value by a desired gain;

calculating a first high frequency value for the pixel value;

generating a second high frequency value as the first high frequency value adjusted by a factor, wherein the factor is a function of the first high frequency value and the desired gain, where the function determines the factor as a percentage value that increases or decreases based on a comparison of an absolute value of the first high frequency value to a first and second threshold to indicate whether noise is present; and

displaying a modified pixel value as the adjusted low frequency value augmented by the second high frequency value.

2. The method of claim 1 further comprising, processing a low pass filter on an image or portion of the image associated with the pixel value to acquire the low frequency value.

3. The method of claim 1 further comprising, iterating the method for each pixel of an image or portion of the image.

4. The method of claim 1 further comprising, receiving the desired gain as a configurable option for processing an image or portion of the image.

5. The method of claim 1, wherein adjusting the low frequency value further includes multiplying the low frequency value by the desired gain.

6. The method of claim 5, wherein calculating the first high frequency value further includes subtracting the low frequency value from the pixel value.

7. The method of claim 6, wherein generating the second high frequency value further includes setting the second high frequency value as the first high frequency multiplied by the desired gain, which is then further multiplied by the factor.

8. The method of claim 1, wherein generating further includes determining the factor as a function of the first high frequency value, the desired gain, and the adjusted low frequency value.

9. A method comprising:

receiving a desired gain for processing an image or portion of the image;

iterating each pixel of the image or the portion of the image by;

adjusting a low frequency value for each pixel;

determining a high frequency value for each pixel;

modifying the high frequency value by a factor where the factor is resolved as a function of a first high frequency value, acquired by subtracting the low frequency value from an original value of the pixel being iterated, and the desired gain, and includes;

comparing the first high frequency value multiplied by the desired gain to a first threshold if the comparison is at or below the first threshold then the factor is set to an inverse of the desired gain;

comparing the first high frequency value multiplied by the desired gain to a second threshold, if the comparison is at or above the second threshold then the factor is set to the desired gain; and

interpolating to acquire a value for the factor between the inverse of the desired gain and the desired gain if the first high frequency value is in between the first and second thresholds; and

saving a modified pixel for each pixel that is represented by the adjusted low frequency value added to the modified high frequency value; and

displaying a modified version of the image or portion of the image as a collection of the saved modified pixels.

10. The method of claim 9, wherein in receiving further includes receiving the desired gain for at least one of a luminance channel and a color channel.

11. The method of claim 9, wherein the processing of the method is embedded within a device.

12. A device comprising:

a memory comprising stored instructions that when executed provide;

a low pass function configured to determine an adjusted low frequency value for a pixel value in response to a desired gain;

a high pass function that determines a high frequency value for the pixel value and then adjusts the high frequency value by a factor, the factor being a percentage value that increases or decreases based on a comparison of an absolute value of the high frequency value to a first and second threshold to indicate whether noise is present; and

a display for displaying an image using the pixel values from the memory.

13. The device of claim 12 further comprising, an pixel modifier function that outputs a modified value for the pixel value represented by the sum of the adjusted low frequency value and the adjusted high frequency value.

14. The device of claim 12 further comprising, an administrative interface that provides an image or portion of the image associated with the pixel value and the desired gain to the system.

15. The device of claim 12, wherein the low pass function is a low pass filter that averages values for surrounding pixels within a configurable area of the pixel value to acquire the low frequency value for the pixel value.

16. The device of claim 12, wherein the high pass function assigns the factor a value represented by an inverse of the desired gain when the high pass frequency value multiplied by the desired gain compares to the first threshold, and wherein the high pass function assigns the factor a value as the desired gain when the high pass frequency value multiplied by the desired gain compares to the second threshold.

17. A system, comprising:

a low pass filter for acquiring an adjusted low frequency value for a pixel value in response to a desired gain;

means for generating a high frequency value for the pixel value and adjusting the high frequency value by a factor, the factor being a percentage value that increases or decreases based on a comparison of an absolute value of the high frequency value to a first and second threshold to indicate whether noise is present in an area of the pixel value; and

an output function that displays a modified value for the pixel value as the adjusted low frequency value combined with the adjusted high frequency value.

18. The system of claim 17, wherein the means for generating further includes means for comparing the desired gain and a first high frequency value to one or more thresholds in order to assign a value to the factor.

19. The system of claim 18, wherein the high frequency value is adjusted by multiplying the first high frequency value with the factor.

20. The system of claim 17, wherein the modified value represents a non linear adjustment to the pixel value.

21. A computer readable medium encoded with computer readable instructions for modifying an image or portion of the image, the instructions when executed cause a computer to perform a method comprising:

receiving a desired gain for an image or portion of the image;

calculating a low pass value for the image or portion of the image in response to the desired gain;

calculating a high pass value for the image or portion of the image;

modifying the high pass value by factors associated with each pixel of the image to minimize noise, where a factor is determined as a value represented by an inverse of the desired gain when the high pass value multiplied by the desired gain is near a first threshold, and where the factor increases toward a value of one when the high pass frequency value multiplied by the desired gain approaches a second threshold; and

displaying a modified version of the image or portion of the image representing the low pass value of the image or portion of the image added to the modified high pass value of the image or portion of the image.

22. The computer readable medium of claim 21 where the instructions are stored and executable on a server as a service that is accessible over a network.

23. The computer readable medium of claim 21 where the instructions are stored and executable on a device.

24. The computer readable medium of claim 21, wherein modifying further includes resolving the factors as a function of a first high frequency value calculated for each pixel and the desired gain.