Method and apparatus for detecting posture of surveillance camera

1. A method of detecting a posture of a camera, the method comprising:

acquiring an image captured by the camera;

acquiring ground information about the ground in the image based on the image; and

detecting the posture of the camera based on the ground information,

wherein the acquiring the ground information comprises:

detecting a first object and a second object from the image;

detecting an intersection between a first extension line and a second extension line, the first extension line connecting first corresponding feature points between the first and second objects, and the second extension line connecting second corresponding feature points between the first and second objects; and

determining one of linear edges detected from the image that is closest to the intersection to be a horizon in the image.

2. The method of claim 1 , wherein the posture of the camera comprises a tilting angle of the camera, and

wherein the detecting the posture of the camera comprises detecting the tilting angle of the camera based on the horizon.

3. The method of claim 1 , wherein the detecting the first and second objects is performed by detecting the first object from a first frame of the image and detecting the second object from a second frame of the image, wherein the first and second frames are captured by the camera at different times.

4. The method of claim 1 , wherein the detecting the first and second objects is performed by tracing the first object in the first frame and detecting the first object traced in the second frame as the second object.

5. The method of claim 1 , wherein the first object is detected using a motion detecting method.

6. The method of claim 1 , wherein the detecting the first and second objects is performed by a feature recognition method based on same feature values.

7. The method of claim 1 , wherein the detecting the first and second objects, the detecting the first and second feature points, and the detecting the intersection are repeated a plurality of times, and

wherein the determining the horizon is performed using a plurality of intersections obtained by the repetition.

8. The method of claim 1 , wherein the camera is operated in a top-view mode to capture images of an area from an upper side of the area or in a side-view mode to capture images of the area from a lateral side of the area, and

wherein the detecting the posture of the camera is performed by detecting a posture of the camera corresponding to the top-view mode or side-view mode of the camera.

9. The method of claim 8 , wherein the acquiring the ground information comprises:

detecting motion vectors of moving objects in the image; and

acquiring the ground information based on a direction of the detected motion vectors.

10. The method of claim 9 ,

wherein the acquiring the ground information further comprises determining whether the motion vectors have directivity, and

wherein the acquiring the ground information further comprises determining that the posture of the camera is in the top-view mode if the motion vectors are determined as having no directivity, and determining that the posture of the camera is in the side-view mode

if the motion vectors are determined as having directivity.

11. An apparatus for detecting a posture of a camera, the apparatus comprising at least one processor to implement:

an image acquisition unit configured to acquire an image captured by the camera coupled to a vertically-rotatable tilting unit;

a ground information acquisition unit configured to acquire information about the ground in the image; and

a posture detection unit configured to detect the posture of the camera based on the ground information,

wherein the ground information acquisition unit is further configured to:

detecting a first object and a second object from the image,

detecting an intersection between a first extension line and a second extension line, the first extension line connecting first corresponding feature points between the first and second objects, and the second extension line connecting second corresponding feature points between the first and second objects, and

determining one of linear edges detected from the image that is closest to the intersection to be a horizon in the image.

12. The apparatus of claim 11 , wherein the camera is operated in a top-view mode to capture images of an area from an upper side of the area or in a side-view mode to capture images of an area from a lateral side of the area,

wherein the ground information acquisition unit comprises a movement line detection part configured to detect motion vectors of moving objects in the image, and

wherein the posture detection unit detects a posture of the camera corresponding to one of the top-view mode and the side-view mode based on whether the motion vectors have directivity.

13. The apparatus of claim 12 , wherein the movement line detection part detects a plurality of motion vectors, and

wherein the ground information acquisition unit further comprises a directivity detection part configured to determine whether the motion vectors have directivity,

wherein if the motion vectors are determined as having no directivity, the posture detection unit determines the posture of the camera as a posture corresponding to the top-view mode, and

wherein if the motion vectors are determined as having directivity, the posture detection unit determine the posture of the camera as a posture corresponding to the side-view mode.