Device, method and system for automatic stabilization of a series of retinal images

1. A method of stabilizing a series of retinal angiography images to provide a coherent serial display, said method comprising:

locally matching each pair of overlapping images, said locally matching includes at least providing a best offset vector for the images, wherein the best offset vector is provided based upon a matching of features from overlapping portions of the images;

globally matching at least a triplet of matching pairs of images, wherein pairs of globally matching images include locally matching pairs whose best offset vector sum is substantially zero;

choosing a key image;

aligning remaining images from the series of images to the key image, wherein said aligning comprises computing distortions between the key image and each of the other images; and

computing corrective transformations to compensate for the distortions using one of: a free quadratic model, an affine model and a translation model, and rendering the other images according to the computed corrective transformations.

2. The method according to claim 1 , wherein said images comprise an internal fixation device and further comprising removing said internal fixation device from the series of images prior to said step of locally matching.

3. The method according to claim 2 , further comprising restoring the internal fixation device to said series of images following said step of aligning.

4. The method according to claim 1 , wherein said choosing a key image comprises choosing the image having a highest amount of correlative data with the other images.

5. The method according to claim 1 , wherein said computing comprises matching features in said key image and in the other images using correlation.

6. An apparatus for stabilizing a series of retinal angiography images to provide a coherent serial display, comprising:

means for locally matching each pair of overlapping images, said means for locally matching including at least means for providing a best offset vector for the images, wherein the best offset vector is provided based upon a matching of features from overlapping portions of the images;

means for globally matching at least a triplet of matching pairs of images, wherein pairs of globally matching images include locally matching pairs whose best offset vector sum is substantially zero;

means for choosing a key image; and

means for aligning remaining images from the series of images to the key image, wherein the means for aligning comprises means for computing distortions between the key image and each of the other images;

means for computing corrective transformations to compensate for the distortions using one of: a free quadratic model, an affine model and a translation model;

means for rendering the other images according to the computed corrective transformations.

7. The apparatus according to claim 6 , wherein said images comprise an internal fixation device and further comprising means for removing said internal fixation device from the series of images.

8. The apparatus according to claim 7 , further comprising means for restoring the internal fixation device to said series of images.

9. The apparatus according to claim 6 , wherein said means for choosing a key image comprise means for choosing the image having a highest amount of correlative data with the other images.

10. The apparatus according to claim 6 , wherein said means for computing comprise means for matching features in said key image and in the other images using correlation.

11. A non-transitory program storage device readable by machine, tangibly embodying a program of instructions executable by the machine to perform method steps of enabling a stabilization of a series of retinal angiography images to provide a coherent serial display, said method comprising:

locally matching a pair of overlapping images, said locally matching includes at least providing a best offset vector for the images, wherein the best offset vector is provided based upon a matching of features from overlapping portions of the images;

globally matching at least a triplet of matching pairs of images, wherein pairs of globally matching images include locally matching pairs whose best offset vector sum is substantially zero;

choosing a key image;

aligning remaining images from the series of images to the key image, wherein said aligning comprises computing distortions between the key image and each of the other images; and

computing corrective transformations to compensate for the distortions using one of: a free quadratic model, an affine model and a translation model, and rendering the other images according to the computed corrective transformations.

12. The program storage device according to claim 11 , wherein said images comprise an internal fixation device and further comprising removing said internal fixation device from the series of images prior to said locally matching.

13. The program storage device according to claim 12 , further comprising restoring the internal fixation device to said series of images following said aligning.

14. The program storage device according to claim 11 , wherein said choosing a key image comprises choosing the image having a highest amount of correlative data with the other images.

15. The program storage device according to claim 11 , wherein said computing comprises matching features in said key image and in the other images using correlation.