Image processing device and method of correcting images
An image processing device to correct for wobble of an image constructed according to exemplary embodiments of the invention includes an input interface to communicate with an image capture sensor, and a motion vector detector to process first and second images received from the image sensor through the input interface, to detect, in the first image, feature points having feature values higher than a threshold value, and to compare the first image to the second image by using at least parts of the feature points to determine a global motion vector of the first image. The threshold value is adjustable depending on a gain of the image sensor.
1. An image processing device to correct for wobble of an image, the image processing device comprising:
an input interface to communicate with an image sensor; and
a motion vector detector to process first and second images captured by the image sensor using a single gain value,
wherein the motion vector detector is configured to:
receive the first and second images from the image sensor through the input interface;
obtain the single gain value of the image sensor;
detect, in the first image, feature points having feature values higher than a threshold value; and
compare the first image to the second image by using at least parts of the feature points to determine a global motion vector of the first image, the threshold value being adjustable depending on the single gain value of the image sensor.
2. The image processing device of claim 1 , wherein the motion vector detector is configured to increase the threshold value as the gain value of the image sensor increases.
3. The image processing device of claim 2 , wherein:
the threshold value is a first value when the gain value of the image sensor is lower than a first level;
the threshold value is a second value higher than the first value when the gain value of the image sensor is higher than a second level that is higher than the first level; and
the threshold value increases between the first value and the second value as the gain value of the image sensor increases when the gain value of the image sensor is between the first level and the second level.
4. The image processing device of claim 1 , wherein the feature values are contrast levels.
5. The image processing device of claim 1 , wherein:
the first image comprises a plurality of sub blocks; and
the motion vector detector is configured to detect feature point candidates in the plurality of sub blocks, respectively, and to select one or more of the feature point candidates having feature values higher than the threshold value as the feature points.
6. The image processing device of claim 1 , further comprising:
an image stabilizer to correct the first image in response to a signal from the global motion vector.
7. The image processing device of claim 1 , wherein the motion vector detector is configured to generate signals representative of local motion vectors of feature point groups of the first image by comparing them to the second image, and to generate signals representative of the global motion vector based upon the local motion vectors, and wherein the feature point groups are defined by grouping the at least parts of the feature points.
8. The image processing device of claim 7 , wherein the motion vector detector is configured to adjust the number of feature points included in each of the feature point groups depending on the gain value of the image sensor.
9. The image processing device of claim 8 , wherein the number of feature points included in each of the feature point groups increases as the gain value of the image sensor increases.
10. The image processing device of claim 8 , wherein:
the number of feature points is a first value when the gain value of the image sensor is lower than a first level;
the number of feature points is a second value higher than the first value when the gain value of the image sensor is higher than a second level that is higher than the first level; and
the number of feature points increases between the first value and the second value as the gain value of the image sensor increases when the gain value of the image sensor is between the first level and the second level.
11. A method for correcting wobble in an image, the method comprising steps of:
receiving first and second images captured by an image sensor using a single gain value;
receiving the single gain value of the image sensor;
detecting, in the first image, feature points having feature values higher than a threshold value; and
comparing the first image to the second image by using at least parts of the feature points to generate signals to correct for the wobble of the first image,
wherein the threshold value is adjustable depending on the single gain value of the image sensor.
12. The method of claim 11 , wherein the threshold value increases as the gain value of the image sensor increases.
13. The method of claim 11 , wherein the step of comparing the first image to the second image comprises:
defining feature point groups by grouping the at least parts of the feature points;
generating signals representative of local motion vectors of the feature point groups of the first image by comparing to the second image; and
generating signals representative of a global motion vector of the first image based upon the local motion vectors.
14. The method of claim 13 , wherein the number of feature points included in each of the feature point groups is adjusted depending on the gain value of the image sensor.
15. The method of claim 14 , wherein the number of feature points included in each of the feature point groups increases as the gain value of the image sensor increases.
16. An image processing device to correct for wobble in an image, the image processing device comprising:
an input interface to communicate with an image sensor; and
a motion vector detector to process first and second images captured by the image sensor using a single gain value,
wherein the motion vector detector is configured to:
receive the first and second images from the image sensor through the input interface;
obtain the single gain value of the image sensor; and
generate signals representative of a global motion vector of the first image based upon local motion vectors of feature point groups of the first image by comparing them to the second image, the feature point groups being defined by grouping at least parts of feature points included in the first image,
wherein the motion vector detector is configured to adjust the number of feature points included in each of the feature point groups depending on the single gain value of the image sensor.
17. The image processing device of claim 16 , wherein the number of feature points included in each of the feature point groups increases as the gain value of the image sensor increases.
18. The image processing device of claim 16 , wherein:
the feature points included in the first image have feature values higher than a threshold value,
the threshold value increases as the gain value of the image sensor increases.
19. The image processing device of claim 18 , wherein:
the first image includes a plurality of sub blocks; and
the motion vector detector is configured to detect feature point candidates in the plurality of sub blocks, respectively, and to select one or more of the feature point candidates having feature values higher than the threshold value as the feature points included in the first image.
20. The image processing device of claim 18 , wherein the threshold value increases as the gain value of the image sensor increases.