Method and apparatus for processing image data

• US 20060110063A1
• Filed: 01/11/2006
• Published: 05/25/2006
• Est. Priority Date: 04/20/2001
• Status: Active Grant

• Current Assignee: Shell & Slate Software Incorporated
• Original Assignee: Shell & Slate Software Incorporated
• Inventors: Weiss, Ben
• Granted Patent: US 7,680,327 B2
• Time in Patent Office: Days
• US Class Current: 382/262
• CPC Class Codes: G06T 2207/20016 Hierarchical, coarse-to-fine, multiscale or multiresolution image processing; Pyramid transform
• CPC Class Codes: G06T 2207/20032 Median filtering
• CPC Class Codes: G06T 5/20 using local operators
• CPC Class Codes: G06T 5/40 using histogram techniques
• CPC Class Codes: G06T 5/70 Denoising; Smoothing

Abstract

The invention is a method and apparatus for improving image-processing applications. Embodiments of the invention provide methods for preserving computation results and offer intermediary computation steps to allow the processing of images at any location to take advantage of previously processed image areas. The preferred embodiment offers a method for building a median filter that significantly improves the processing speed over basic techniques. By building a histogram hierarchy, image data statistics are added and subtracted using a multiplicity of histograms from the histogram hierarchy, where each histogram describes an image area. Furthermore, a histogram hierarchy is built using multiple layers, each layer defining a level of statistical resolution.

Claims

1. A method for processing image data through a median filter comprising:

obtaining a set of histograms;

loading a first image data set into said set of histograms;

computing a first plurality of median values of said first image data set from said set of histograms;

partially updating said set of histograms with a second image data set; and

,

computing a second plurality of median values of said second image data set from said set of histograms.

2. The method of claim 1, wherein said obtaining said histogram further comprises obtaining a plurality of histogram hierarchies.

3. The method of claim 2, wherein said plurality of histogram hierarchies further comprise each one of said plurality of histogram hierarchies having at least one memory array.

4. The method of claim 3, wherein said at least one memory array further comprises associating said memory array with a histogram'"'"'s numerical resolution.

5. The method of claim 1, wherein said obtaining said set of histograms further comprises aligning at least two memory arrays associated said set of histograms.

6. The method of claim 1, wherein said first image data set further comprises a set of multiple color channels of an image area.

7. The method of claim 1, wherein said loading said first image data set further comprises applying at least one statistical process to said image data set.

8. The method of claim 7, wherein said at least one statistical process further comprises computing a frequency for each value in said first image data set.

9. The method of claim 1, wherein said loading further comprises incrementing at least one memory location in said set of histograms.

10. The method of claim 1, wherein said computing said first plurality of median values further comprises finding a fiftieth percentile value in each histogram in said set of histograms.

11. The method of claim 1, wherein said computing said first plurality of median values further comprises finding a percentile value exceeding a threshold for selecting each one said median values.

12. The method of claim 1, wherein said computing said first plurality of medians further comprises traversing said set of histograms through more than one numerical resolution.

13. The method of claim 1, wherein said partially updating said set of histograms further comprising:

removing a first non-overlapping image data set from said set of histograms, wherein said first non-overlapping image data set is part of said first image data and is not part of said second image data; and

adding a second non-overlapping image data set to said set of histograms, wherein said second non-overlapping image data set is part of said second image data and is not part of said first image data.

14. The method of claim 1, wherein said computing said second plurality of median values further comprises finding a fiftieth percentile value in said set of histograms.

15. The method of claim 1, wherein said computing said second plurality of median values further comprises finding a percentile value exceeding a threshold for selecting said each one of said median values.

16. The method of claim 1, wherein said computing a second plurality of said median values further comprises traversing said set of histograms.