Method for image processing using local statistics convolution

1. A method to filter a digital image comprising a plurality of pixels, the method comprising:

segmenting the plurality of pixels into a first plurality of regions;

computing a first plurality of regional histograms by calculating a regional histogram corresponding to each region in the first plurality of regions, wherein the regional histogram is a three-dimensional histogram;

computing a first plurality of regional transfer functions by calculating for each region in the first plurality of regions a regional transfer function comprising a convolution of the corresponding regional histogram with a first convolution kernel;

deriving an image transfer function by interpolation of two or more of the first plurality of regional transfer functions;

filtering the digital image by applying the image transfer function to each of the plurality of pixels to generate a filtered digital image; and

causing the filtered digital image to be displayed.

2. The method of claim 1 , wherein the plurality of pixels comprises all pixels of the digital image.

3. The method of claim 2 , wherein each pixel of the plurality of pixels is contained within a single region of the first plurality of regions.

4. The method of claim 1 , wherein the three-dimensional histogram includes a mapping of pixel counts again a luminance and two chrominances.

5. The method of claim 1 , further comprising performing a bilinear interpolation on results retrieved from the three-dimensional histogram.

6. The method of claim 1 , wherein the filtering includes performing a filtering operation specific to one color.

7. The method of claim 1 , further comprising determining an entropy of a given object in the digital image, wherein the entropy is determined independent of similar luminosity objects in the digital image having a different color than the given object.

8. A program storage device readable by a machine, tangibly embodying a program of instructions executable by the machine to perform operations to transform a digital image comprising a plurality of pixels, said operations comprising:

segmenting the plurality of pixels into a plurality of regions;

computing a plurality of regional histograms by calculating a regional histogram corresponding to each region in the plurality of regions;

computing a first plurality of regional transfer functions by calculating for each region in the plurality of regions a regional transfer function comprising a convolution of the corresponding regional histogram with a first convolution kernel;

deriving an image transfer function by interpolation of two or more of the first plurality of regional transfer functions; and

filtering the digital image by applying the image transfer function to each of the plurality of pixels.

9. The program storage device of claim 8 , wherein deriving the image transfer function by the interpolation of the two or more of the first plurality of regional transfer functions comprises:

selecting a pixel having a luminosity value from a first region of the plurality of regions;

computing a plurality of single value results by inputting the luminosity value of the selected pixel into each of a second plurality of regional transfer functions, said second plurality of regional transfer functions including regional transfer functions corresponding to regions other than the first region; and

interpolating a resulting value for the selected pixel based on two of the plurality of single value results.

10. The program storage device of claim 8 , wherein the first convolution kernel is a Gaussian function.

11. The program storage device of claim 8 , where the regional transfer function further comprises a second convolution comprising the convolution of the corresponding regional histogram with a second convolution kernel.

12. The program storage device of claim 11 , where the first convolution kernel is a Gaussian function, and the second convolution kernel is a product of a Gaussian function and a step function.

13. The program storage device of claim 11 , where the second convolution is weighted by pixel luminosity.

14. The program storage device of claim 8 , where the plurality of regional histograms are a function of one or more of luminosity, pixel difference, color, depth along a z-axis, and saturation.

15. A method to filter a digital image, the method comprising:

segmenting the digital image into a plurality of tiles;

computing tile histograms corresponding to each of the plurality of tiles, wherein each tile histogram includes a three-dimensional histogram;

deriving a plurality of tile transfer functions from the tile histograms;

interpolating a tile transfer function from the plurality of tile transfer functions; and

filtering the digital image with the interpolated tile transfer function.

16. The method of claim 15 , where the tiles are non-overlapping, and each pixel of the digital image is contained within a single tile.

17. The method of claim 15 , wherein the three-dimensional histogram includes a mapping of pixel counts again a luminance and two chrominances.

18. The method of claim 15 , wherein the filtering includes performing a filtering operation specific to one color.

19. The method of claim 15 , further comprising determining an entropy of a given object within the digital image without being affected by a presence of one or more similar luminosity objects within the digital image having a different color than the given object.

20. The method of claim 15 , wherein the interpolating comprises:

selecting a pixel having a luminosity value from a first tile of the plurality of tiles;

computing a plurality of single value results by inputting the luminosity value of the selected pixel into each of a second plurality of regional transfer functions, said second plurality of regional transfer functions including regional transfer functions corresponding to tiles not comprising the selected pixel; and

interpolating a resulting value for the selected pixel as a function of the plurality of single value results.