Pixel location calibration image capture and processing
What is disclosed are systems and methods for optical correction for correcting for non-uniformity in active matrix light emitting diode device (AMOLED) and other emissive displays, using iterative processing of images of calibration patterns including features of coarse and fine granularity to successively generate a high-resolution estimate of the panel pixel locations.
1. An optical correction method for correcting display of images on a display panel having pixels, each pixel having a light-emitting device, the method comprising:
arranging a camera in front of the display panel;
displaying one or more calibration patterns on the display panel while capturing one or more calibration images of said calibration patterns with said camera, said one or more calibration patterns comprising a spaced pattern of coarse features and a spaced pattern of fine features, a spacing of the coarse features larger than a spacing of the fine features;
generating a coarse estimate of panel pixel locations within the calibration images from the images of the coarse features in the calibration images;
locating images of the fine features within the calibration images with use of the coarse estimate;
generating a high-resolution estimate of panel pixel locations within the calibration images from the located images of the fine features in the calibration images, the high-resolution estimate having greater accuracy than the coarse estimate; and
generating correction data for correcting images displayed in the display panel with use of the high-resolution estimate.
2. The optical correction method of claim 1 , wherein the one or more calibration patterns comprises a single calibration pattern, and wherein the once or more calibration images comprises a single image.
3. The optical correction method of claim 1 , wherein the coarse features are spaced apart from a periphery of the one or more calibration patterns.
4. The optical correction method of claim 1 , wherein the fine features are distributed throughout the one or more calibration patterns.
5. The optical correction method of claim 1 , wherein each fine feature includes pixels of a foreground color, and each coarse feature includes pixels of a foreground color surrounded by an area of a background color, said area absent other coarse features or fine features.
6. The optical correction method of claim 1 , wherein the coarse estimate comprises a first 2D polynomial function, the high-resolution estimate comprises a second 2D polynomial function, and the second 2D polynomial function has an order greater than an order of the first 2D polynomial function.
7. The optical correction method of claim 1 , wherein said generating the coarse estimate includes:
locating images of the coarse features in the one or more calibration images;
identifying the coarse features of the one or more calibration patterns corresponding to said images of the coarse features; and
generating a coarse mapping between panel pixel locations and calibration image pixel locations from locations of the images of the coarse features in the one or more calibration images and known locations of the coarse features in the one or more calibration patterns,
wherein said locating images of the fine features includes:
estimating expected locations of images of the fine features within the one or more calibration images with use of the coarse estimate and known locations of the fine features in the one or more calibration patterns, and
wherein said generating the high-resolution estimate includes:
identifying the fine features of the one or more calibration patterns corresponding to said images of the fine features; and
generating a high-resolution mapping between panel pixel locations and calibration image pixel locations from locations of the images of the fine features in the one or more calibration images and known locations of the fine features in the one or more calibration patterns.
8. The optical correction method of claim 7 , wherein identifying the fine features of the one or more calibration patterns corresponding to said images of the fine features includes:
for each expected location of an image of a fine feature, determining the closest image of a fine feature in the one or more calibrations images which falls within a distance threshold.
9. The optical correction method of claim 1 , wherein the one or more calibration patterns comprises a single calibration pattern and the once or more calibration images comprises a single image, wherein the coarse features are spaced apart from a periphery of the one or more calibration patterns and include a single pixel of a foreground color surrounded by a square area of a background color, said square area absent other coarse features or fine features, wherein the fine features are distributed throughout the single calibration pattern and each fine feature includes a single pixel of a foreground color, and wherein the coarse estimate comprises a first 2D polynomial function, the high-resolution estimate comprises a second 2D polynomial function, and the second 2D polynomial function has an order greater than an order of the first 2D polynomial function.
10. The optical correction method of claim 9 , wherein said generating the coarse estimate includes:
locating images of the coarse features in the single calibration image;
identifying the coarse features of the single calibration pattern corresponding to said images of the coarse features; and
generating a coarse mapping between panel pixel locations and calibration image pixel locations from locations of the images of the coarse features in the single calibration image and known locations of the coarse features in the single calibration pattern,
wherein said locating images of the fine features includes:
estimating expected locations of images of the fine features within the single calibration image with use of the coarse estimate and known locations of the fine features in the single calibration pattern, and
wherein said generating the high-resolution estimate includes:
for each expected location of an image of a fine feature, determining the closest image of a fine feature in the single calibration image which falls within a distance threshold to identify the fine features of the single calibration pattern corresponding to said images of the fine features; and
generating a high-resolution mapping between panel pixel locations and calibration image pixel locations from locations of the images of the fine features in the single calibration image and known locations of the fine features in the single calibration pattern.
11. An optical correction system for correcting display of images on a display panel having pixels, each pixel having a light-emitting device, the system comprising:
a camera arranged in front of the display panel;
an optical processing circuit coupled to said camera adapted to
display one or more calibration patterns on the display panel while capturing one or more calibration images of said calibration patterns with said camera, said one or more calibration patterns comprising a spaced pattern of coarse features and a spaced pattern of fine features, a spacing of the coarse features larger than a spacing of the fine features;
generate a coarse estimate of panel pixel locations within the calibration images from the images of the coarse features in the calibration images;
locate images of the fine features within the calibration images with use of the coarse estimate;
generate a high-resolution estimate of panel pixel locations within the calibration images from the located images of the fine features in the calibration images, the high-resolution estimate having greater accuracy than the coarse estimate; and
generate correction data for correcting images displayed in the display panel with use of the high-resolution estimate.
12. The optical correction system of claim 11 , wherein the one or more calibration patterns comprises a single calibration pattern, and wherein the once or more calibration images comprises a single image.
13. The optical correction system of claim 11 , wherein the coarse features are spaced apart from a periphery of the one or more calibration patterns.
14. The optical correction system of claim 11 , wherein the fine features are distributed throughout the one or more calibration patterns.
15. The optical correction system of claim 11 , wherein each fine feature includes pixels of a foreground color, and each coarse feature includes pixels of a foreground color surrounded by an area of a background color, said area absent other coarse features or fine features.
16. The optical correction system of claim 11 , wherein the coarse estimate comprises a first 2D polynomial function, the high-resolution estimate comprises a second 2D polynomial function, and the second 2D polynomial function has an order greater than an order of the first 2D polynomial function.
17. The optical correction system of claim 1 , wherein the optical processing circuit is adapted to generate the coarse estimate by:
locating images of the coarse features in the one or more calibration images;
identifying the coarse features of the one or more calibration patterns corresponding to said images of the coarse features; and
generating a coarse mapping between panel pixel locations and calibration image pixel locations from locations of the images of the coarse features in the one or more calibration images and known locations of the coarse features in the one or more calibration patterns,
wherein the optical processing circuit is adapted to locate images of the fine features by:
estimating expected locations of images of the fine features within the one or more calibration images with use of the coarse estimate and known locations of the fine features in the one or more calibration patterns, and
wherein the optical processing circuit is adapted to generate the high-resolution estimate by:
identifying the fine features of the one or more calibration patterns corresponding to said images of the fine features; and
generating a high-resolution mapping between panel pixel locations and calibration image pixel locations from locations of the images of the fine features in the one or more calibration images and known locations of the fine features in the one or more calibration patterns.
18. The optical correction system of claim 17 , wherein the optical processing circuit is adapted to identify the fine features of the one or more calibration patterns corresponding to said images of the fine features by:
for each expected location of an image of a fine feature, determining the closest image of a fine feature in the one or more calibrations images which falls within a distance threshold.
19. The optical correction system of claim 11 , wherein the one or more calibration patterns comprises a single calibration pattern and the once or more calibration images comprises a single image, wherein the coarse features are spaced apart from a periphery of the one or more calibration patterns and include a single pixel of a foreground color surrounded by a square area of a background color, said square area absent other coarse features or fine features, wherein the fine features are distributed throughout the single calibration pattern and each fine feature includes a single pixel of a foreground color, and wherein the coarse estimate comprises a first 2D polynomial function, the high-resolution estimate comprises a second 2D polynomial function, and the second 2D polynomial function has an order greater than an order of the first 2D polynomial function.
20. The optical correction system of claim 19 , wherein the optical processing circuit is adapted to generate the coarse estimate by:
locating images of the coarse features in the single calibration image;
identifying the coarse features of the single calibration pattern corresponding to said images of the coarse features; and
generating a coarse mapping between panel pixel locations and calibration image pixel locations from locations of the images of the coarse features in the single calibration image and known locations of the coarse features in the single calibration pattern,
wherein the optical processing circuit is adapted to locate images of the fine features by:
estimating expected locations of images of the fine features within the single calibration image with use of the coarse estimate and known locations of the fine features in the single calibration pattern, and
wherein the optical processing circuit is adapted to generate the high-resolution estimate by:
for each expected location of an image of a fine feature, determining the closest image of a fine feature in the single calibration image which falls within a distance threshold to identify the fine features of the single calibration pattern corresponding to said images of the fine features; and
generating a high-resolution mapping between panel pixel locations and calibration image pixel locations from locations of the images of the fine features in the single calibration image and known locations of the fine features in the single calibration pattern.