Multi-projector intensity blending system

1. A method of projecting an image utilizing a plurality of projectors, wherein at least two of the projectors project overlapping portions of the image in a global display space, the method comprising:

identifying overlapping pixels in the global display space;

generating an attenuation map for the projectors, wherein the attenuation values for the overlapping pixels of the attenuation map are at least partially a function of the number of projectors contributing to the overlap;

establishing pixel intensity values for the projectors by applying one or more intensity transfer functions to the attenuation maps generated for the projectors, wherein the intensity transfer functions are configured to at least partially account for the non-linear response of the output intensity of the projectors, as a function of an input intensity control signal applied to the projectors, and the intensity transfer functions are established by methodology comprising:

generating a blended image using the attenuation map;

adjusting pixel intensity for overlapping pixels in the blended image to create an enhanced blended image;

repeating the pixel intensity adjustment for overlapping pixels in the blended image at one or more additional image intensity values; and

using the initial and repeated intensity adjustments to establish the intensity transfer function.

2. A method as claimed in claim 1 wherein the intensity transfer function is established by fitting a mathematical function to the initial and repeated intensity adjustments.

3. A method as claimed in claim 1 wherein the intensity transfer function is established by correlating the initial and repeated intensity adjustments with target output intensities and storing the correlated values in a lookup table.

4. A method as claimed in claim 1 wherein:

pixels in the overlapping portions of the image are characterized by two or more distance-to-edge values d 1 , d 2 , d i , representing the respective distances of a selected pixel from the closest edge of each projected portion of the image in the global display space; and

the relative magnitudes of the distance-to-edge values for the selected pixel are used in establishing the pixel intensity values for the selected pixel.

5. A method as claimed in claim 4 wherein the distance-to-edge values are measured as a direct pixel distance in the frame buffer of each projector or as distances in a planar or spherical global coordinate system.

6. A method as claimed in claim 4 wherein:

the distance-to-edge values are used to define a degree to which each projector dominates pixel intensity for the pixels in the overlapping portions of the image; and

pixel intensity is established such that more dominant projectors are subject to less attenuation for pixels where the distance-to-edge values define the projector as a more dominant projector.

7. A method as claimed in claim 4 wherein the relative magnitudes of the distance-to-edge values for the selected pixel are used in establishing the pixel intensity values for the selected pixel independent of the intensity transfer function.

8. A method as claimed in claim 4 wherein the relative magnitudes of the distance-to-edge values for the selected pixel are used in establishing the pixel intensity values for selected pixels in the attenuation map.

9. A method as claimed in claim 1 wherein:

pixels in the overlapping portions of the image are characterized by two or more distance-to-edge values d 1 , d 2 , d i , representing the respective distances of a selected pixel from the closest edges of projected portions of the image in the global display space; and

an average or weighted average of the distance-to-edge values for the selected pixel is used in establishing the pixel intensity values for the selected pixel.

10. A method as claimed in claim 1 wherein the method further comprises an artifact reduction operation, the artifact reduction operation comprising an attenuation map perturbation routine adding noise, attenuation factor smoothing, digital filtering, or JND metrics to the attenuation values in the map.

11. A method as claimed in claim 1 wherein the method further comprises an artifact reduction operation where the attenuation map is compared to a model of the human visual system in order to determine what characteristics of the attenuation map are likely to lead to artifacts that can be readily detected by the human visual system and, once identified, the attenuation map is modified to account for such artifacts.

12. An image projection system comprising a plurality of projectors and an intensity blending controller programmed to execute the method of projecting an image recited in claim 1 .

13. A method of calibrating the intensity response function of a projector or operating the image projector according to a calibrated intensity response function, the method comprising:

generating a calibration image comprising an intensity-adjusted portion and a dithered portion, wherein pixels of the intensity-adjusted portion are driven at a fraction of maximum input intensity and the ratio of on/off pixels in the dithered portion is selected to correspond to the fractional input intensity of the intensity-adjusted portion;

adjusting pixel intensity in either the intensity-adjusted portion or the dithered portion to match the apparent intensity of the remaining image portion;

repeating the pixel intensity adjustment for one or more additional fractional input intensities;

using the initial and repeated intensity adjustments to establish an intensity response function for the image projector, wherein the intensity response function is configured to at least partially account for the non-linear response of the output intensity of the projector as a function of an input intensity control signal applied to the projector; and

utilizing the intensity response function as an input to an image-blending intensity transfer function to establish pixel intensity values for the projector.

14. A method of calibrating the intensity response function of a projector or operating the image projector according to a calibrated intensity response function, the method comprising:

generating an attenuation map for the projectors, wherein the attenuation values for the overlapping pixels of the attenuation map are at least partially a function of the number of projectors contributing to the overlap;

generating a blended image using the attenuation map;

adjusting pixel intensity for overlapping pixels in the blended image to create an enhanced blended image;

repeating the pixel intensity adjustment for overlapping pixels in the blended image at one or more additional image intensity values;

using the initial and repeated intensity adjustments to establish an intensity transfer function for the image projector, wherein the intensity response function is configured to at least partially account for the non-linear response of the output intensity of the projector as a function of an input intensity control signal applied to the projector; and

utilizing the intensity response function as an input to an image-blending intensity transfer function to establish pixel intensity values for the projector.

15. A method of projecting an image utilizing a plurality of projectors, wherein at least two of the projectors project overlapping portions of the image in a global display space, the method comprising:

identifying overlapping pixels in the global display space;

generating an attenuation map for the projectors, wherein the attenuation values for the overlapping pixels of the attenuation map are at least partially a function of the number of projectors contributing to the overlap; and

establishing pixel intensity values for the projectors by applying one or more intensity transfer functions to the attenuation maps generated for the projectors, wherein the intensity transfer functions are configured to at least partially account for the non-linear response of the output intensity of the projectors, as a function of an input intensity control signal applied to the projectors, and the intensity transfer functions are established by methodology comprising:

generating a calibration image comprising an intensity-adjusted portion and a dithered portion, wherein pixels of the intensity-adjusted portion are driven at a fraction of maximum input intensity and the ratio of on/off pixels in the dithered portion is selected to correspond to the fractional input intensity of the intensity-adjusted portion;

adjusting pixel intensity in either the intensity-adjusted portion of the calibration image or the dithered portion of the calibration image to match the apparent intensity of the remaining portion of the calibration image;

repeating the pixel intensity adjustment for one or more additional fractional input intensities; and

using the initial and repeated intensity adjustments to establish the intensity transfer function.

16. A method as claimed in claim 15 wherein:

pixels in the overlapping portions of the image are characterized by two or more distance-to-edge values d 1 , d 2 , d i , representing the respective distances of a selected pixel from the closest edge of each projected portion of the image in the global display space; and

the relative magnitudes of the distance-to-edge values for the selected pixel are used in establishing the pixel intensity values for the selected pixel.

17. A method as claimed in claim 16 wherein the distance-to-edge values are measured as a direct pixel distance in the frame buffer of each projector or as distances in a planar or spherical global coordinate system.

18. A method as claimed in claim 16 wherein:

the distance-to-edge values are used to define a degree to which each projector dominates pixel intensity for the pixels in the overlapping portions of the image; and

pixel intensity is established such that more dominant projectors are subject to less attenuation for pixels where the distance-to-edge values define the projector as a more dominant projector.

19. A method as claimed in claim 16 wherein the relative magnitudes of the distance-to-edge values for the selected pixel are used in establishing the pixel intensity values for the selected pixel independent of the intensity transfer function.

20. A method as claimed in claim 16 wherein the relative magnitudes of the distance-to-edge values for the selected pixel are used in establishing the pixel intensity values for selected pixels in the attenuation map.

21. A method as claimed in claim 15 wherein:

pixels in the overlapping portions of the image are characterized by two or more distance-to-edge values d 1 , d 2 , d i , representing the respective distances of a selected pixel from the closest edges of projected portions of the image in the global display space; and

an average or weighted average of the distance-to-edge values for the selected pixel is used in establishing the pixel intensity values for the selected pixel.

22. A method as claimed in claim 15 wherein the method further comprises an artifact reduction operation, the artifact reduction operation comprising an attenuation map perturbation routine adding noise, attenuation factor smoothing, digital filtering, or JND metrics to the attenuation values in the map.

23. A method as claimed in claim 15 wherein the method further comprises an artifact reduction operation where the attenuation map is compared to a model of the human visual system in order to determine what characteristics of the attenuation map are likely to lead to artifacts that can be readily detected by the human visual system and, once identified, the attenuation map is modified to account for such artifacts.

24. An image projection system comprising a plurality of projectors and an intensity blending controller programmed to execute the method of projecting an image recited in claim 15 .

25. A method as claimed in claim 15 wherein the calibration image is generated from a single projector.

26. A method as claimed in claim 15 wherein the initial and repeated intensity adjustments are used to establish the intensity transfer function by fitting the intensity transfer function to values representing the adjustments.

27. A method as claimed in claim 15 wherein the initial and repeated intensity adjustments are used to establish the intensity transfer function by correlating the initial and repeated intensity adjustments with target output intensities and storing the correlated values in a lookup table.