Color and geometry distortion correction system and method
View Patent ↗Systems and methods for correcting color and geometry in an image are disclosed. In one example, image data are received in one format and are then converted to another format to perform color correction. In another example, geometric corrections are first applied to each color component of an image data to correct for different distortions including lateral chromatic aberrations. Next, color distortion corrections are performed to correct for each color component of the image data independently. In one particular exemplary embodiment, color distortion corrections are applied using surface function representation of a grid data transformation relating uncorrected values to corrected values.
1. A method for correcting color distortions in an image, said method comprising:
(a) obtaining input image data in a first format color space;
(b) converting the input image data from the first format color space into a second format color space;
(c) separately correcting color distortions in the converted image data for each color component, using a color correction transformation, in the second format color space and thereby generating color distortion corrected data, wherein the corrected color component depends on all color components and the color correction transformation is represented by surface functionals,
such that a displayed image based on the color distortion corrected data is visually free of color distortions.
2. The method of claim 1 , wherein the second format color space is gamma de-compensated.
3. The method of claim 1 , wherein the first format color space is gamma compensated.
4. The method of claim 1 , wherein the color corrected data in the second format color space is converted to a third format color space for display.
5. The method of claim 1 , wherein the color distortions are caused by at least one of an image acquisition device, a broadcasting system, and a display system.
6. The method of claim 5 , wherein the color distortions include those caused by at least one of compression, coding, transmission, decoding, and decompression.
7. The method of claim 1 , wherein correcting color distortions depends on specific regions of the image.
8. The method of claim 7 , wherein the specific regions are specified by at least one pixel.
9. The method of claim 1 , further correcting luminance non-uniformities for each color component.
10. The method of claim 1 , further correcting chrominance non-uniformities for each color component.
11. The method of claim 1 , wherein the color corrected values are adjusted by applying an offset function and a gain function.
12. The method of claim 11 wherein said offset function and gain function are represented as warp maps.
13. The method of claim 1 , wherein the color corrected values are further adjusted by applying a soft-clipping function.
14. The method of claim 1 , further applying filtering to obtain pixel color values.
15. The method of claim 1 , wherein color correction is performed using a floating point representation in the second format color space.
16. The method of claim 1 , further using a floating point representation format according to required precision and hardware resources.
17. The method of claim 1 , wherein format conversion is achieved using look-up-tables.
18. A method for correcting color distortions in an image having color data points, said method comprising:
(a) obtaining grid transformation data relating distorted color data points to corrected color data points, said grid transformation data spanning color space and image coordinate space,
(b) dividing the displayed image into a plurality of patches wherein size of each patch varies depending on the region of the displayed image,
(c) representing the grid transformation data corresponding to each patch using surface functionals,
(d) separately applying the surface functionals for said each patch in the image for each color component, wherein the corrected color component depends on all color components, to correct for color distortions in the image data.
19. The method of claim 18 , further dividing color space and image coordinate space into patches according to distortion complexity level in each patch, and representing the grid transformation data in each patch for each color component with a surface functional.
20. The method of claim 18 , further comparing the difference between color values rendered by a surface functional and color values rendered by the grid transformation data against a pre-set tolerance level, and further dividing the patches if the difference is greater than the pre-set tolerance level.
21. The method of claim 18 , wherein the color distortions are caused by at least one of an image acquisition device, a broadcasting system, and a display system.
22. The method of claim 21 , wherein the color distortions include those caused by at least of compression, coding, transmission, decoding, and decompression.
23. A method for correcting an image having color components, in an image display system having geometric and color distortions, said method comprising:
(a) obtaining image data for each of said color components;
(b) correcting for geometric distortions in the image display system for each color component independently and thereby generating geometrically corrected image data,
(c) separately correcting for color distortions in the geometric corrected image data for each color component using a color correction transformation, to generate geometric distortion corrected and color distortion corrected image data, wherein the corrected color component depends on all color components and the color correction transformation is represented by surface functionals,
such that a displayed image based on the geometric distortion corrected and color distortion corrected image data is visually geometric distortion free and color distortion free.
24. An electronic system for correcting color distortions in an image, said system compromising:
(a) an interface adapted to obtain image data in a first format color space;
(b) conversion means adapted to convert the input image data from the first format color space into a second format color space;
(c) color correction means adapted to correct color distortions in the image data separately for each color component using a color correction transformation in the second format color space and thereby generating color distortion corrected data, wherein the corrected color component depends on all color components and the color correction transformation is represented by surface functionals,
such that a displayed image based on the color distortion corrected data is visually free of color distortions.
25. The electronic system of claim 24 , further adapted to make correcting color distortions depend on specific regions of an image.
26. The electronic system of claim 25 , wherein the specific regions are specified by at least one pixel.
27. The electronic system of claim 24 , further adapted to correct luminance non-uniformities for each color component.
28. The electronic system of claim 24 , further adapted to correct chrominance non-uniformities for each color component.
29. The electronic system of claim 24 , further adapted to obtain color corrected values by applying an offset function and a gain function.
30. The electronic system of claim 29 , further adapted to represent the offset function and the gain function as warp maps.
31. The electronic system of claim 24 , further adapted to adjust color corrected values by applying a soft-clipping function.
32. The electronic system of claim 24 , further adapted to apply filtering to obtain pixel color values.
33. The electronic system of claim 24 , further adapted to perform format conversion using look-up-tables.