Digital image processing apparatus and method of estimating global motion of image
View Patent ↗Provided are a digital image processing apparatus which corrects a position of each of a plurality of template motion vectors of an image frame of a captured image by considering relative positions of neighboring motion vectors respectively similar to each of the template motion vectors, calculates a score of each of the template motion vectors by considering a number of the neighboring motion vectors, and selects a global motion vector representing the image frame based on the score, and a method of estimating a global motion of an image to stabilize the captured image using the global motion.
1. A digital image processing apparatus comprising:
a template motion vector obtaining unit configured to obtain template motion vectors respectively corresponding to a plurality of templates of a plurality of different areas in an image frame of a captured image;
a position correcting unit configured to correct a position of each of the template motion vectors based on a weighted sum of each of the template motion vectors and at least one neighboring motion vector neighboring each of the template motion vectors;
a score calculating unit configured to calculate a score of each of the template motion vectors based on a confidence of each of the template motion vectors and a number of the at least one neighboring motion vector corresponding to each of the template motion vectors; and
a global motion estimating unit configured to select a global motion vector representing the image frame from among the template motion vectors, based on the score calculated by the score calculating unit.
2. The digital image processing apparatus of claim 1 , wherein the template motion vector obtaining unit is further configured to match the image frame to the plurality of templates to obtain the template motion vectors.
3. The digital image processing apparatus of claim 1 , wherein the weighted sum is calculated by using a confidence of each of the template motion vectors and a confidence of each of the at least one neighboring motion vector corresponding to each of the template motion vectors.
4. The digital image processing apparatus of claim 1 , wherein each of the at least one neighboring motion vector is a motion vector whose Euclidean distance from a template motion vector corresponding to the at least one neighboring motion vector is less than a threshold value.
5. The digital image processing apparatus of claim 1 , further comprising a stabilizing unit configured to stabilize the captured image comprising the image frame based on the global motion vector.
6. The digital image processing apparatus of claim 5 , wherein the stabilizing unit is configured to stabilize the captured image by moving the image frame according to the global motion vector.
7. The digital image processing apparatus of claim 5 , wherein the stabilizing unit is configured to stabilize the captured image by correcting data of the image frame according to the global motion vector.
8. A method of estimating a global motion of an image, the method comprising:
obtaining template motion vectors respectively corresponding to a plurality of templates of a plurality of different areas in an image frame of a captured image;
correcting a position of each of the template motion vectors based on a weighted sum of each of the template motion vectors and at least one neighboring motion vector neighboring with each of the template motion vectors;
calculating a score of each of the template motion vectors based on a confidence of each of the template motion vectors and a number of the at least one neighboring motion vector corresponding to each of the template motion vectors; and
selecting a global motion vector representing the image frame from among the template motion vectors, based on the score calculated by the score calculating unit.
9. The method of claim 8 , further comprising matching the image frame to the plurality of templates to obtain the template motion vectors.
10. The method of claim 8 , wherein the weighted sum is calculated by using a confidence of each of the template motion vectors and a confidence of each of the at least one neighboring motion vector corresponding to each of the template motion vectors.
11. The method of claim 8 , wherein each of the at least one neighboring motion vector is a motion vector whose Euclidean distance from a template motion vector corresponding to the neighboring motion vector is less than a threshold value.
12. The method of claim 8 , further comprising stabilizing the captured image comprising the image frame based on the global motion vector.
13. The method of claim 12 , wherein the stabilizing the captured image comprises stabilizing the captured image by moving the image frame according to the global motion vector.
14. The method of claim 12 , wherein the stabilizing the captured image comprises stabilizing the captured image by correcting data of the image frame according to the global motion vector.
15. A non-transitory computer-readable recording medium having recorded thereon a program for executing the method of claim 8 .
16. A non-transitory computer-readable recording medium having recorded thereon a program for executing the method of claim 10 .
17. A non-transitory computer-readable recording medium having recorded thereon a program for executing the method of claim 11 .
18. A non-transitory computer-readable recording medium having recorded thereon a program for executing the method of claim 12 .