Device and method for removing distortion

1. A device for removing distortion, the device comprising:

a photographing unit configured to photograph a plurality of frames in succession by using a rolling shutter;

a storage unit configured to store photographed images of the plurality of frames;

a local motion vector generator configured to generate a local motion vector between a reference image and an input image among the photographed images;

a parameter estimator configured to estimate a parameter set indicative of an amount of distortion in the input image in view of the reference image, based on the local motion vector; and

an image reconstruction unit configured to generate a reconstructed image by using the estimated parameter set and information about the input image,

wherein the parameter estimator is further configured to estimate the parameter set by selecting at least one scan line in the reference image and at least one corresponding line in the input image, and

wherein the parameter set is a set of variables representing coordinates of a start point and an end point of the corresponding line which are located at a first curved boundary line and a second curve boundary line of the input image, respectively.

2. The device of claim 1 , wherein the at least one scan line comprises a line traversing the reference image horizontally so that the start point and the end point have the same y coordinate.

3. The device of claim 1 , wherein the parameter estimator estimates the parameter set by selecting at least one point present on the scan line in the reference image and at least one corresponding point in the corresponding line in the input image.

4. The device of claim 3 , wherein the parameter estimator selects the corresponding point based on the local motion vector.

5. The device of claim 3 , wherein the parameter estimator estimates the parameter set by selecting the point and the corresponding point such that the two points internally divide the scan line in the reference image and the corresponding line in the input image in the same ratio, respectively.

6. The device of claim 3 , wherein the parameter estimator estimates the parameter set by setting all points in the corresponding line by functions of a y-coordinate of the scan line in the reference image.

7. The device of claim 6 , wherein the parameter estimator calculates coordinates of the corresponding point in the input image by using a linear equation in the form of Ax =b, where A is a matrix represented by a function of a y-coordinate of the scan line, x is the parameter set, and b is coordinates of the corresponding point in the input image.

8. The device of claim 1 , wherein the image reconstruction unit searches for at least one point in the input image corresponding to at least one arbitrary point in the reconstructed image by using the estimated parameter set, and generates the reconstructed image by using a pixel value of the at least one point in the input image.

9. A method of removing distortion from an input image in a device for removing the distortion, the method comprising:

storing a plurality of frames photographed in succession by a photographing unit by using a rolling shutter, in a storage unit;

generating a local motion vector between a reference image and an input image among the photographed images in a local motion vector generator;

estimating a parameter set to determine an amount of distortion in the input image in view of the reference image, based on the local motion vector, in the wobbling parameter estimator; and

generating a reconstructed image by using the estimated parameter set and information about the input image in the image reconstruction unit,

wherein the estimating the parameter set comprises selecting at least one scan line in the reference image and at least one corresponding line in the input image, and

wherein the parameter set is a set of variables representing coordinates of a start point and an end point of the corresponding line which are located at a first curve boundary line and a second curved boundary line of the input image, respectively.

10. The method of claim 9 , wherein the at least one scan line comprises a line traversing the reference image horizontally so that the start point and the end point have the same y coordinate.

11. The method of claim 9 , wherein in the estimating the parameter set, the parameter set is estimated by selecting at least one point present on the scan line in the reference image and at least one corresponding point in the corresponding line in the input image.

12. The method of claim 11 , wherein in the estimating the parameter set, the corresponding point is selected based on the local motion vector.

13. The method of claim 11 , wherein in the estimating the parameter set, the point and the corresponding point are selected such that the two points internally divide the scan line in the reference image and the corresponding line in the input image in the same ratio, respectively.

14. The method of claim 11 , wherein in the estimating the parameter set, all points in the corresponding line are set by functions of a y-coordinate of the scan line in the reference image.

15. The method of claim 14 , wherein in the estimating the parameter set, coordinates of the corresponding point in the input image are calculated by using a linear equation in the form of Ax =b, where A is a matrix represented by a function of a y-coordinate of the scan line, x is the parameter set, and b is coordinates of the corresponding point in the input image.

16. The method of claim 9 , wherein in the generating the reconstructed image, at least one point in the input image corresponding to at least one arbitrary point in the reconstructed image is searched for by using the estimated parameter set, and the reconstructed image is generated by using a pixel value of the at least one point in the input image.