Image defect map creation using batches of digital images
View Patent ↗A method of automatically determining a need to service a digital image acquisition system including a digital camera with a lens assembly includes analyzing pixels within one or more acquired digital images according to probability determinations that such pixels correspond to blemish artifacts. It is automatically determined whether a threshold distribution of blemish artifacts is present within one or more of the digital images. A need for service is indicated when at least the threshold distribution is determined to be present.
1. A method of correcting defects in digital images, comprising:
using a processor; acquiring a set of images with a device including a lens and image sensor;
computing a gray-scale version of a current image of the set of images;
computing a local mean of the current image computing a difference image based on differences between the gray-scale version and local mean version of the current image;
computing a current binary map that results after thresholding the difference image;
growing a gray map by adding to the current binary map with subsequent binary maps of subsequent images of the set;
generating a final binary map; and correcting a defect within a digital image based on the final binary map.
2. The method of claim 1 , comprising facilitating implementation of a binarization with an adaptive threshold.
3. The method of claim 1 , further comprising thresholding to obtain a binary map.
4. The method of claim 3 , wherein the threshold comprises an adaptive threshold that depends on number of processed images.
5. The method of claim 4 , wherein the threshold is based on a percentage of the number of images.
6. The method of claim 3 , further comprising calculating a label image.
7. The method of claim 6 , further comprising eliminating defect regions that have an area greater than a defect size threshold.
8. The method of claim 7 , wherein the defect size threshold comprises 800 pixels.
9. The method of claim 1 , wherein the computing a local mean comprises computing low-pass filtering.
10. One or more processor-readable non-transitory media having embedded therein code for programming a processor to perform a method of correcting defects in digital images, wherein the method comprising:
programming the processor;
acquiring a set of images with a device including a lens and image sensor; computing a gray-scale version of a current image of the set of images;
computing a local mean of the current image computing a difference image based on differences between the gray-scale version and local mean version of the current image;
computing a current binary map that results after thresholding the difference image; and
growing a gray map by adding to the current binary map with subsequent binary maps of subsequent images of the set.
11. The one or more processor-readable media of claim 10 , the method comprising facilitating implementation of a binarization with an adaptive threshold.
12. The one or more processor-readable media of claim 10 , wherein the method further comprises thresholding to obtain a binary map.
13. The one or more processor-readable media of claim 12 , wherein the threshold comprises an adaptive threshold that depends on number of processed images.
14. The one or more processor-readable media of claim 13 , wherein the threshold is based on a percentage of the number of images.
15. The one or more processor-readable media of claim 12 , wherein the method further comprises calculating a label image.
16. The one or more processor-readable media of claim 15 , wherein the method further comprises eliminating defect regions that have an area greater than a defect size threshold.
17. The one or more processor-readable media of claim 16 , wherein the defect size threshold comprises 800 pixels.
18. The one or more processor-readable media of claim 10 , wherein the computing a local mean comprises computing low-pass filtering.
19. An image acquisition and processing device that acquires and corrects defects in digital images, comprising a lens, an image sensor, a processor and a memory having code embedded therein for programming the processor to perform within the device the following method:
acquiring a set of images with a device including a lens and image sensor;
computing a gray-scale version of a current image of the set of images;
computing a local mean of the current image computing a difference image based on differences between the gray-scale version and local mean version of the current image;
computing a current binary map that results after thresholding the difference image; and
growing a gray map by adding to the current binary map with subsequent binary maps of subsequent images of the set.
20. The device of claim 19 , the method comprising facilitating implementation of a binarization with an adaptive threshold.
21. The device of claim 19 , wherein the method further comprises thresholding to obtain a binary map.
22. The device of claim 21 , wherein the threshold comprises an adaptive threshold that depends on number of processed images.
23. The device of claim 22 , wherein the threshold is based on a percentage of the number of images.
24. The device of claim 21 , wherein the method further comprises calculating a label image.
25. The device of claim 24 , wherein the method further comprises eliminating defect regions that have an area greater than a defect size threshold.
26. The device of claim 25 , wherein the defect size threshold comprises 800 pixels.
27. The device of claim 19 , wherein the computing a local mean comprises computing low-pass filtering.