Method and system for improving surveillance of PTZ cameras

1. A method, comprising:

obtaining a scene from a video camera and defining at least one point of the scene;

creating a first scene terrain model of the scene, said first scene terrain model comprises a human object size in pixels in said at least one point of the scene;

creating a second scene model of the scene, said second model defines a false positive determination that said at least one point comprises a human object;

wherein said second scene model is created when the scene does not contain a human object;

obtaining an image of the scene, said image is captured by a video camera;

determining whether the human object is detected at the at least one point of said captured image by applying said first scene terrain model and said second scene model on the at least one point;

wherein the second scene model of the scene comprises a Panoramic HOG map;

obtaining a position of a PTZ camera, the position including values of pan, tilt and zoom;

detecting a PTZ frame by the PTZ camera at the obtained PTZ position;

obtaining a successful detection of the human object in a specific location in the frame of the PTZ camera;

determining the polar coordinates of the human object;

determining an altitude of the human object; and

obtaining a panoramic map of the detected PTZ frame and identifying a point of the panoramic map closest to the detected human object, according to the determined polar coordinates.

2. The method according to claim 1 , further comprises determining a matching point in the panoramic map closest to the specific pixel.

3. The method according to claim 1 , wherein determining the altitude of the object after obtaining a bounding box of human object, and selecting two pixels of the bounding box.

4. The method according to claim 3 , wherein the two pixels are a top pixel having top coordinates (x, y1) and a bottom pixel having bottom coordinates (x, y2), defining x as the horizontal middle of the bounding box.

5. The method according to claim 4 , further comprises converting the top pixel and the bottom pixel into polar coordinates, thereby obtaining a polar top pixel (Π1,θ1) and a polar bottom pixel (Π2, θ2).

6. The method according to claim 4 , further comprises converting the polar coordinates of top pixel and the bottom pixel into 3D world coordinates.

7. The method according to claim 4 , further comprises obtaining 2 rays starting from a 3D origin located at a first ray pixel P1=z1v1, and a second ray pixel P2=z2v2.

8. The method according to claim 7 , wherein determining the altitude of a standing human, wherein a line connecting the first ray pixel and the second ray pixel is defined as vertical and parallel to Y-axis.

9. The method according to claim 7 , further comprises determining the object's altitude in a specific point after determining a Y-coordinate of the second ray pixel.

10. The method according to claim 1 , wherein creating the second scene model comprises determining HOG matching scores for all pixels of the image of the scene.

11. The method according to claim 10 , further comprises obtaining an object's location on the frame in pixels.

12. The method according to claim 11 , further comprises converting the human object's location in pixels into polar coordinates on a panoramic map.

13. The method according to claim 10 , further comprises obtaining Panoramic HOG point associated to the polar coordinates of the pixels in which the human object is located.

14. The method of claim 1 , wherein the human object size is determined according to scene geometry and altitude maps.

15. A method, comprising:

obtaining a scene from a video camera and defining at least one point of the scene;

creating a first scene terrain model of the scene, said first scene terrain model comprises a human object size in pixels in said at least one point of the scene;

creating a second scene model of the scene, said second model defines a false positive determination that said at least one point comprises a human object; wherein said

second scene model is created when the scene does not contain a human object;

obtaining an image of the scene, said image is captured by a video camera;

determining whether the human object is detected at the at least one point of said captured image by applying said first scene terrain model and said second scene model on the at least one point;

obtaining a position of a PTZ camera, the position including values of pan, tilt and zoom;

detecting a PTZ frame by the PTZ camera at the obtained PTZ position;

obtaining a successful detection of the human object in a specific location in the frame of the PTZ camera;

determining the polar coordinates of the human object;

determining an altitude of the human object;

obtaining a panoramic map of the detected PTZ frame and identifying a point of the panoramic map closest to the detected human object, according to the determined polar coordinates.

16. A method, comprising:

obtaining a scene from a video camera and defining at least one point of the scene;

creating a first scene terrain model of the scene, said first scene terrain model comprises a human object size in pixels in said at least one point of the scene;

creating a second scene model of the scene, said second model defines a false positive determination that said at least one point comprises a human object; wherein said

second scene model is created when the scene does not contain a human object;

obtaining an image of the scene, said image is captured by a video camera;

determining whether the human object is detected at the at least one point of said captured image by applying said first scene terrain model and said second scene model on the at least one point;

wherein creating the second scene model comprises determining HOG matching scores for all pixels of the image of the scene;

obtaining a position of a PTZ camera, the position including values of pan, tilt and zoom;

detecting a PTZ frame by the PTZ camera at the obtained PTZ position;

obtaining a successful detection of the human object in a specific location in the frame of the PTZ camera;

determining the polar coordinates of the human object;

determining an altitude of the human object; and

obtaining a panoramic map of the detected PTZ frame and identifying a point of the panoramic map closest to the detected human object, according to the determined polar coordinates.