Apparatus and method for automatically recognizing object by using low-speed camera in dual photographing mode

1. An apparatus for automatically recognizing an object by using a low-speed camera in a dual photographing mode, the apparatus comprising:

an image acquirer configured to acquire successive images in a single exposure mode and high-speed images in a multiple exposure mode by using two low-speed cameras capable of performing successive photographing and high-speed photographing operations; and

an image information analyzer configured to generate object movement information by analyzing the acquired successive images to determine a movement state of the object, and by analyzing successive images and high-speed images acquired by controlling photographing operations of the two low-speed cameras based on the determination,

wherein the image information analyzer further comprises:

an object movement information generator configured to determine the movement state of the object to be a state of waiting for movement in response to a determination that the object has yet to start movement, and generate object movement information in response to a determination that the object has started movement.

2. The apparatus of claim 1 , wherein the image acquirer comprises:

a first camera configured to include a first timer and a first shutter, and controlled to perform successive photographing and high-speed photographing operations; and

a second camera configured to include a second timer and a second shutter, and controlled to perform high-speed photographing operations.

3. The apparatus of claim 2 , wherein the first timer and the second timer are synchronized with each other, and the first shutter and the second shutter perform successive photographing and high-speed photographing operations respectively, according to the movement state of the object based on the first timer and the second timer which are synchronized with each other.

4. The apparatus of claim 3 , wherein:

in a case where the movement state of the object is a state of waiting for movement, the image acquirer controls the first shutter to operate in a single exposure mode to enable the first camera to perform successive photographing operations, and to stop photographing operations of the second camera; and

in a case where the movement state of the object is a state of being ready for movement, the image acquirer controls the first shutter to operate in a multiple exposure mode to enable the first camera to perform high-speed photographing operations, and also controls the second shutter to operate in a multiple exposure mode to enable the second camera to perform high-speed photographing operations.

5. The apparatus of claim 3 , wherein the image acquirer acquires high-speed images in a multiple exposure mode by repeatedly operating the first shutter and the second shutter N times at regular time intervals while one image frame is acquired.

6. The apparatus of claim 1 , wherein the image information analyzer further comprises:

a successive image analyzer configured to analyze the acquired successive images to search for foreground regions in the successive images, and to determine that the movement state of the object to be a state of waiting for movement or a state of being ready for movement depending on whether there is a foreground region and an object region having a size similar to an expected size of an object among sizes of the searched foreground regions in the successive images;

a photographing controller configured to control photographing operations of the two cameras included in the image acquirer according to the determined movement state of the object; and

a high-speed image analyzer configured to analyze the acquired successive images to search for foreground regions in the high-speed images, and to determine whether the object has started movement by detecting an object region from the searched foreground regions, and by comparing a central point of an object region in a high-speed image of a current frame with a central point of an object region in a high-speed image of a previous frame.

7. The apparatus of claim 6 , wherein the successive image analyzer searches for the foreground regions of the successive images by generating a background by locally calculating an average value or a mean value of the brightness of the successive images, by removing regions having a lower brightness value than the brightness of the generated background, and by analyzing connectivity of regions remaining in the successive images from which the background is removed.

8. The apparatus of claim 6 , wherein the high-speed image analyzer searches for the foreground regions of the high-speed images by generating a background by locally calculating an average value or a mean value of the brightness of the high-speed images, by removing regions having a lower brightness value than the brightness of the generated background, and by analyzing connectivity of regions remaining in the high-speed images from which the background is removed.

9. The apparatus of claim 6 , wherein the successive image analyzer determines whether there is a foreground region having a size similar to the size of the object, wherein in response to a determination that there is no foreground region having a size similar to the size of the object, the successive image analyzer determines the movement state of the object to be a state of waiting for movement, and in response to a determination that there is a foreground region having a size similar to the size of the object, the successive image analyzer compares an expected shape of the object with a shape of the foreground region to determine again whether there is a foreground region having a size similar to the size of an object, wherein in response to a determination that there is no foreground region having a size similar to the size of the object, the successive image analyzer determines the movement state of the object to be a state of waiting for movement, and in response to a determination that there is a foreground region having a size similar to the size of the object, the successive image analyzer determines the movement state of the object to be a state of being ready for movement.

10. The apparatus of claim 6 , wherein the object movement information generator further comprises:

an object speed information generator configured to generate object speed information, which includes the movement speed and direction of the object, by reconstructing three-dimensional positions of central points in an object region to generate object movement information, and by calculating displacement of the reconstructed 3D positions of central points in the object region;

an object rotation information generator configured to generate object rotation information, which includes a force and axis of rotation of the object, by reconstructing a 3D pattern on a surface of the object by using images of regions surrounding the central points of the object region, and by calculating rotation of the reconstructed 3D pattern of the regions surrounding the central points of the object region; and

an object movement information generator configured to generate object movement information that includes the object speed information and the object rotation information.

11. A method of automatically recognizing an object by using a low-speed camera in a dual photographing mode, the method comprising:

acquiring successive images in a single exposure mode and high-speed images in a multiple exposure mode by using two low-speed cameras capable of performing successive photographing and high-speed photographing operations; and

generating object movement information by analyzing the acquired successive images to determine a movement state of the object, and by analyzing successive images and high-speed images acquired by controlling photographing operations of the two low-speed cameras based on the determination,

wherein the generating of the object movement information further comprises:

determining the movement state of the object to be a state of waiting for movement in response to a determination that the object has yet to start movement, and generating object movement information in response to a determination that the object has started movement.

12. The method of claim 11 , wherein the acquiring of the images comprises:

acquiring successive and high-speed images by using a first camera including a first timer and a first shutter, and controlled to perform successive photographing and high-speed photographing operations, and a second camera including a second timer and a second shutter, and controlled to perform high-speed photographing operations.

13. The method of claim 12 , wherein the first timer and the second timer are synchronized with each other, and the first shutter and the second shutter perform successive photographing and high-speed photographing operations respectively, according to the movement state of the object based on the first timer and the second timer which are synchronized with each other.

14. The method of claim 13 , wherein the acquiring of the images comprises:

in a case where the movement state of the object is a state of waiting for movement, controlling the first shutter to operate in a single exposure mode, so as to enable the first camera to perform successive photographing operations and to stop photographing operations of the second camera; and

in a case where the movement state of the object is a state of being ready for movement, controlling the first shutter to operate in a multiple exposure mode to enable the first camera to perform high-speed photographing operations, and also controlling the second shutter to operate in a multiple exposure mode to enable the second camera to perform high-speed photographing operations.

15. The method of claim 13 , wherein the acquiring of the images comprises acquiring high-speed images in a multiple exposure mode by repeatedly operating the first shutter and the second shutter N times at regular time intervals while one image frame is acquired.

16. The method of claim 11 , wherein the generating of the object movement information further comprises:

analyzing the acquired successive images to search for foreground regions in the successive images, and determining that the movement state of the object to be a state of waiting for movement or a state of being ready for movement depending on whether there is a foreground region and an object region having a size similar to an expected size of an object among sizes of the searched foreground regions in the successive images;

controlling photographing operations of the two cameras included in the image acquirer according to the determined movement state of the object; and

analyzing the acquired successive images to search for foreground regions in the high-speed images, and determining whether the object has started movement by detecting an object region from the searched foreground regions, and by comparing a central point of an object region in a high-speed image of a current frame with a central point of an object region in a high-speed image of a previous frame.

17. The method of claim 16 , wherein the determining that the movement state of the object to be a state of waiting for movement or a state of being ready for movement comprises searching for the foreground regions of the successive images by generating a background by locally calculating an average value or a mean value of the brightness of the successive images, by removing regions having a lower brightness value than the brightness of the generated background, and by analyzing connectivity of regions remaining in the successive images from which the background is removed.

18. The method of claim 16 , wherein the determining whether the object has started movement comprises searching for the foreground regions of the high-speed images by generating a background by locally calculating an average value or a mean value of the brightness of the high-speed images, by removing regions having a lower brightness value than the brightness of the generated background, and by analyzing connectivity of regions remaining in the high-speed images from which the background is removed.

19. The method of claim 16 , wherein the determining whether the object has started movement comprises determining whether there is a foreground region having a size similar to the size of the object, wherein in response to a determination that there is no foreground region having a size similar to the size of the object, the movement state of the object is determined to be a state of waiting for movement, and in response to a determination that there is a foreground region having a size similar to the size of the object, an expected shape of the object is compared with a shape of the foreground region to determine again whether there is a foreground region having a size similar to the size of an object, wherein in response to a determination that there is no foreground region having a size similar to the size of the object, the movement state of the object is determined to be a state of waiting for movement, and in response to a determination that there is a foreground region having a size similar to the size of the object, the movement state of the object is determined to be a state of being ready for movement.

20. The method of claim 16 , wherein the generating of the object movement information further comprises:

generating object speed information, which includes the movement speed and direction of the object, by reconstructing three-dimensional positions of central points in an object region to generate object movement information, and by calculating displacement of the reconstructed 3D positions of central points in the object region;

generating object rotation information, which includes a force and axis of rotation of the object, by reconstructing a 3D pattern on a surface of the object by using images of regions surrounding the central points of the object region, and by calculating rotation of the reconstructed 3D pattern of the regions surrounding the central points of the object region; and

generating object movement information that includes the object speed information and the object rotation information.