Flash artifact eye defect correction in blurred images using anisotropic blurring
First Claim
Patent Images
1. A method for digital image flash artefact eye defect correction comprising:
- acquiring a digital image;
identifying one or more candidate eye defect regions in the image comprising one or more flash artefact regions including pixels in one or more pupil regions that are red, yellow, golden or white, or combinations thereof;
determining whether the image is blurred;
correcting the one or more candidate eye defect regions, including replacing said red, yellow, golden or white pixels with corrected pixels to form one or more corrected regions, wherein the corrected regions comprise probabilistically uncertain defect pixels utilizing between 10% and 70% information derived from pixels of the acquired digital image, as well as probabilistically certain pixels not utilizing information from pixels of the acquired digital image, and uncorrected glint pixels having a luminance above a certain threshold to be classified as certain glint pixels; and
responsive to the image being blurred, anisotropically blurring the one or more corrected regions of the image.
5 Assignments
0 Petitions
Accused Products
Abstract
A method and apparatus for providing defect correction in blurred images is disclosed. For one embodiment of the invention, a digital image is acquired. One or more candidate defect regions in the image are identified. The candidate defect regions are corrected and responsive to the image being blurred, the corrected regions of the image are blurred. For one embodiment of the invention, a digital image is acquired and one or more candidate eye defect regions in the image are identified. The candidate eye defect regions are corrected and if the image is blurred, the corrected regions of the image are blurred accordingly.
176 Citations
56 Claims
-
1. A method for digital image flash artefact eye defect correction comprising:
-
acquiring a digital image; identifying one or more candidate eye defect regions in the image comprising one or more flash artefact regions including pixels in one or more pupil regions that are red, yellow, golden or white, or combinations thereof; determining whether the image is blurred; correcting the one or more candidate eye defect regions, including replacing said red, yellow, golden or white pixels with corrected pixels to form one or more corrected regions, wherein the corrected regions comprise probabilistically uncertain defect pixels utilizing between 10% and 70% information derived from pixels of the acquired digital image, as well as probabilistically certain pixels not utilizing information from pixels of the acquired digital image, and uncorrected glint pixels having a luminance above a certain threshold to be classified as certain glint pixels; and responsive to the image being blurred, anisotropically blurring the one or more corrected regions of the image. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17)
-
-
18. A digital image processing device including a processor arranged to correct digital image flash artefact eye defects, comprising:
-
a flash artefact eye defect correction component configured both to identify one or more candidate eye defect regions in the image comprising one or more flash artefact regions including pixels in one or more pupil regions that are red, yellow, golden or white, or combinations thereof; and
to correct the one or more candidate eye defect regions, including replacing said red, yellow, golden or white pixels with corrected pixels to form one or more corrected regions, wherein the one or more corrected regions comprise probabilistically uncertain defect pixels utilizing between 10% and 70% information derived from pixels of the acquired digital image, as well as probabilistically certain pixels not utilizing information from pixels of the acquired digital image, and uncorrected glint pixels having a luminance above a certain threshold to be classified as certain glint pixels; andan anisotropic blurring component configured to determine whether the image is blurred; and
, responsive to the image being blurred, anisotropically blurring the one or more corrected regions of the image. - View Dependent Claims (19, 20, 21, 22)
-
-
23. One or more non-transitory processor readable hardware components having code embedded therein for programming one or more processors to perform a method for digital image flash artefact eye defect correction, wherein the method comprises:
-
identifying one or more candidate eye defect regions in said image comprising one or more flash artefact regions including pixels in one or more pupil regions that are red, yellow, golden or white, or combinations thereof; determining whether said image is blurred; correcting said one or more candidate eye defect regions, including replacing said red, yellow, golden or white pixels with corrected pixels to form one or more corrected regions, wherein the corrected regions comprise probabilistically uncertain defect pixels utilizing between 10% and 70% information derived from pixels of the acquired digital image, as well as probabilistically certain pixels not utilizing information from pixels of the acquired digital image, and uncorrected glint pixels having a luminance above a certain threshold to be classified as certain glint pixels; and responsive to said image being blurred, anisotropically blurring said one or more corrected regions of said image. - View Dependent Claims (24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39)
-
-
40. A digital image acquisition device, comprising:
-
a camera including a flash for acquiring a digital image; a processor; a memory having code embedded therein for programming the processor to perform a method for digital image flash artefact eye defect correction, wherein the method comprises; identifying one or more candidate eye defect regions in said digital image comprising one or more flash artefact regions each including pixels in a pupil region that are red, yellow, golden or white, or combinations thereof; determining whether said digital image is blurred; correcting said one or more candidate eye defect regions, including replacing said red, yellow, golden or white pixels with corrected pixels to form one or more corrected regions, wherein the corrected regions comprise probabilistically uncertain defect pixels utilizing between 10% and 70% information derived from pixels of the acquired digital image, as well as probabilistically certain pixels not utilizing information from pixels of the acquired digital image, and uncorrected glint pixels having a luminance above a certain threshold to be classified as certain glint pixels; and responsive to said image being blurred, anisotropically blurring said one or more corrected regions of said digital image. - View Dependent Claims (41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56)
-
Specification