Granted Patent B2

US 9,007,472 · App. 13/611,725 · Granted Apr 14, 2015

Abstract

Claims (51)

1. A reference image setting unit comprising:

a sample frame extract unit which receives a plurality of image frames taken for a certain period of time from image data generated by photographing an object; and

a reference frame extract unit which compares the plurality of image frames with one another based on distances between a center point and a plurality of characterizing points in each image frame, extracts the least shaken image frame from among the plurality of image frames according to a result of the comparison, and sets the extracted least shaken image frame as a reference image frame.

2. The reference image setting unit of claim 1 , wherein the plurality of image frames are generated by continuously photographing the object.

3. The reference image setting unit of claim 1 , wherein the reference frame extract unit comprises:

a center point detector which detects the center point of each image frame of the plurality of image frames;

a characterizing point detector which detects the plurality of characterizing points in each image frame; and

a reference frame selector which extracts the least shaken image frame from among the plurality of image frames by analyzing the distances between the center point and the plurality of characterizing points in each image frame.

4. The reference image setting unit of claim 3 , further comprising:

a frame average calculator which calculates a frame average of each image frame by using the center point and the plurality of charactering points in each image frame; and

a frame comparison value calculator which calculates a frame comparison value for each image frame by using the frame average of each image frame,

wherein the reference frame selector selects an image frame having the smallest value from among a plurality of frame comparison values for the plurality of image frames, which is the least shaken image frame, as the reference image frame.

5. The reference image setting unit of claim 4 , wherein the frame average calculator calculates the frame average of each image frame by averaging distances between the center point and the plurality of characterizing points in each image frame.

6. The image stabilizing apparatus of claim 5 , wherein the frame comparison value calculator calculates the plurality of frame comparison values by summing absolute values, which are obtained by subtracting the frame averages of other image frames from the frame average of each image frame.

7. An image stabilizing apparatus comprising:

the reference image setting unit of claim 1 ; and

an image adjusting unit which compares a currently input image frame of an image of the photographed object with the reference image frame, and adjusts the currently input image frame according to a shaken degree of the currently input image frame in view of the reference image frame.

8. The image stabilizing apparatus of claim 7 , further comprising:

a characterizing point checking region setting unit which sets a characterizing point checking region comprising a plurality of characterizing points in the image,

wherein, in the characterizing point checking region, the image adjusting unit calculates an optical flow, and moves the currently input image frame using the optical flow.

9. The image stabilizing apparatus of claim 8 , wherein the characterizing point checking region setting unit comprises:

a center point detector which detects a representative centroid point of the plurality of image frames; and

a checking region setting/adjusting unit which sets the characterizing point checking region so that the region comprises the plurality of characterizing points which are greater than or equal to a predetermined level.

10. The image stabilizing apparatus of claim 9 , wherein the optical flow comprises a plurality of directions and magnitudes for the plurality of characterizing points, and

wherein the image adjusting unit calculates a representative direction and magnitude from the plurality of directions and magnitudes, and moves the currently input image frame in a direction opposite to the representative direction by the representative magnitude.

11. An image stabilizing method comprising:

performing the method of claim 1 ; and

comparing a currently input image frame of an image of the photographed object with the reference image frame, and adjusting the currently input image frame according to a shaken degree of the currently input image frame in view of the reference image frame.

12. The method of claim 11 , further comprising:

setting a characterizing point checking region comprising a plurality of characterizing points in the image,

wherein the adjusting the currently input image frame comprises calculating an optical flow in the characterizing point checking region, and moving the currently input image frame using the optical flow.

13. The method of claim 12 , wherein the setting the characterizing point checking region comprises:

detecting a representative centroid point of the plurality of image frames; and

setting the characterizing point checking region so that the region comprises the plurality of characterizing points which are greater than or equal to a predetermined level.

14. The method of claim 13 , wherein the optical flow comprises a plurality of directions and magnitudes for the plurality of characterizing points,

wherein the calculating the optical flow comprises calculating a representative direction and magnitude from the plurality of directions and magnitudes, and

wherein the moving the currently the moves the currently input image frame in a direction opposite to the representative direction by the representative magnitude.

15. A reference image setting method comprising:

generating, by using a camera, a plurality of image frames by photographing an object for a certain period of time; and

comparing the plurality of image frames with one another based on distances between a center point and a plurality of characterizing points in each image frame, extracting a least shaken image frame from among the plurality of image frames using a result of the comparing, and setting the extracted least shaken image frame as a reference image frame.

16. The method of claim 15 , wherein the plurality of image frames are generated by continuously photographing the object.

17. The method of claim 15 , wherein the comparing, extracting and setting comprises:

detecting the center point of each image frame of the plurality of image frames;

detecting the plurality of characterizing points in each image frame; and

extracting the least shaken image frame from among the plurality of image frames by analyzing the distances between the center point and the plurality of characterizing points in each image frame.

18. The method of claim 17 , wherein the extracting the least shaken image frame comprises:

calculating a frame average of each image frame by using the center point and the plurality of charactering points in each image frame;

calculating a frame comparison value for each image frame by using the frame average of each image frame; and

selecting an image frame having the smallest value from among a plurality of frame comparison values for the plurality of image frames, which is the least shaken image frame, as the reference image frame.

19. The method of claim 18 , wherein the calculating the frame average comprises calculating the frame average of each image frame by averaging distances between the center point and the plurality of characterizing points in each image frame.

20. The method of claim 19 , wherein the calculating the frame comparison value for each image frame comprises calculating the plurality of frame comparison values by summing absolute values, which are obtained by subtracting the frame averages of other image frames from the frame average of each image frame.

Assignments (12)