Method of sampling feature points, image matching method using the same, and image matching apparatus

1. A method of sampling feature points for image matching, the method comprising:

dividing a first image into a plurality of regions corresponding to a number of feature points to be sampled;

extracting a feature point from each of the regions such that a distance between an extracted feature point of one region and an extracted feature point of another region satisfies a predetermined condition; and

estimating a homography based on extracted feature points of the first image and corresponding feature points of a second image to be matched with the first image,

wherein the extracting comprises:

extracting a first feature point from among feature points in a first region;

extracting a second feature point from among feature points in a second region which are spaced apart by a first distance or more from the first feature point;

extracting a third feature point from among feature points in a third region which are spaced apart by a second distance or more from the second feature point; and

extracting a fourth feature point from among feature points in a fourth region which are spaced apart by a third distance or more from the third feature point.

2. The method of claim 1 , wherein, in the extracting the feature point from each of the regions, the feature point of the other region is extracted after the feature point of the one region is extracted.

3. The method of claim 1 , wherein the extracting the third feature point comprises extracting the third feature point from among feature points in the third region which are spaced apart by the first distance or more from the first feature point and spaced apart by the second distance or more from the second feature point.

4. The method of claim 1 , wherein the extracting the fourth feature point comprises extracting the fourth feature point from among feature points in the fourth region, which are spaced apart by the first distance or more from the first feature point, spaced apart by the second distance or more from the second feature point, and spaced apart by the third distance or more from the third feature point.

5. The method of claim 1 , wherein the first to third distances are greater than distances from the first to third feature points to a center point of the first image, respectively.

6. The method of claim 1 , wherein the first and second images are acquired by a same image sensor or different image sensors.

7. The method of claim 6 , wherein if the first and second images are acquired by the same image sensor, the first and second images are images of a same object captured from different points of view or captured at different times, and

wherein if the first and second images are acquired by different image sensors, the different image sensors have different characteristics and are configured to capture different images of the same object.

8. The image matching method of claim 1 further comprises:

matching the first and second images using the estimated homography; and

fusing the first and second images to generate a fusion image.

9. An image matching method comprising:

extracting a plurality of feature points from a first image and a second image to be matched with the first image;

selecting at least one corresponding feature point pair from among the feature points of the first and second images;

dividing the first image into a plurality of regions corresponding to a number of feature points to be sampled;

sampling a feature point from each of the regions of the first image such that a distance between a sampled feature point of one region and a sampled feature point of another region satisfy a predetermined condition; and

estimating a homography based on sampled feature points of the first image and corresponding feature points of the second image

wherein the sampling comprises:

selecting a first feature point from among feature points in a first region;

selecting a second feature point from among feature points in a second region which are spaced apart by a first distance or more from the first feature point;

selecting a third feature point from among feature points in a third region which are spaced apart by a second distance or more from the second feature point; and

selecting a fourth feature point from among feature points in a fourth region which are spaced apart by a third distance or more from the third feature point.

10. An image matching processor comprising:

a region divider configured to divide a first image into a plurality of regions corresponding to a number of feature points to be sampled;

a feature point extractor configured to extract a feature point from each of the regions such that a distance between an extracted feature point of one region and an extracted feature point of another region satisfies a predetermined condition; and

an estimator configured to estimate a homography based on extracted feature points of the first image and corresponding feature points of a second image to be matched with the first image,

wherein the feature point extractor is further configured to extract a first feature point from among feature points in a first region, extract a second feature point from among feature points in a second region which are spaced apart by a first distance or more from the first feature point, extract a third feature point from among feature points in a third region which are spaced apart by a second distance or more from the second feature point, and extract a fourth feature point from among feature points in a fourth region which are spaced apart by a third distance or more from the third feature point.

11. The image matching processor of claim 10 , wherein the feature point extractor is configured to extract the feature point of the other region after extracting the feature point of the one region.

12. The image matching processor of claim 10 , wherein the feature point extractor configured to extract the third feature point from among feature points in the third region which are spaced apart by the first distance or more from the first feature point and spaced apart by the second distance or more from the second feature point.

13. The image matching processor of claim 10 , wherein the feature point extractor configured to extract the fourth feature point from among feature points in the fourth region which are spaced apart by the first distance or more from the first feature point, spaced apart by the second distance or more from the second feature point, and spaced apart by the third distance or more from the third feature point.

14. The image matching processor of claim 10 , wherein the first to third distances are greater than distances from the first to third feature points to a center point of the first image, respectively.

15. The image matching processor of claim 10 , wherein the first and second images are acquired by a same image sensor or different image sensors.

16. The method of claim 15 , wherein if the first and second images are acquired by the same image sensor, the first and second images are images of a same object captured from different points of view or captured at different times, and

wherein if the first and second images are acquired by different image sensors, the different image sensors have different characteristics and are configured to capture different images of the same object.

17. The image matching processor of claim 10 , further comprising:

a feature point extractor configured to extract a plurality of feature points of the first and second images; and

a feature point pair selector configured to select at least one corresponding feature point pair from among the feature points of the first and second images.

18. The image matching processor of claim 10 , further comprising an image fusion processor configured to fuse the first and second images to generate a fusion image by using the estimated homography.