Virtual sensor systems and methods

1. A method for detecting activation of a virtual sensor in a scene, wherein the virtual sensor comprises data representing a three-dimensional geometric form, data representing a position in the scene and data representing one or several trigger conditions, the method comprising:

capturing a first data representation of the scene using at least one physical sensor; and

determining whether one or several of the virtual sensor trigger conditions is/are fulfilled based on an analysis of the captured first data representation in a first volume area corresponding to the geometric form and position of the virtual sensor,

wherein the determining whether a virtual sensor trigger condition is fulfilled comprises

analyzing point cloud data representing surfaces of objects detected by the at least one physical sensor in the scene, the point cloud data comprising data representing non-empty points of a second volume area corresponding to the captured first data representation and respective associated position in the second volume area and

determining whether an object of the scene enters the first volume area corresponding to the geometric form and position of the virtual sensor by determining whether a number of non-empty points whose position fall in the first volume area which correspond to the geometric form and position of the virtual sensor is beyond a threshold specified by a trigger condition.

2. The method according to claim 1 , wherein the first data representation of the scene includes an image data representation of the scene.

3. The method according to claim 1 , further comprising:

capturing a second data representation of the scene;

configuring the geometric form and position of the virtual sensor based on an image generated from the captured second data representation.

4. The method according to claim 1 , wherein analyzing point cloud data representing points in the first volume area which corresponds to the geometric form and position of the virtual sensor comprises:

detecting the points of the point cloud which are positioned in the first volume area corresponding to the geometric form and position of the virtual sensor; and

comparing the number of detected points to a predetermined threshold.

5. The method according to claim 1 , wherein analyzing point cloud data representing points in the first volume area which correspond to the geometric form and position of the virtual sensor comprises:

detecting the points of the point cloud which are positioned in the area corresponding to the geometric form and position of the virtual sensor;

determining an intensity value representing intensity amounts for some of the detected points; and

comparing the determined intensity value to a predetermined threshold.

6. The method according to claim 1 , further comprising processing the captured first data representation of the scene for removing data that represent a background of the scene.

7. The method according to claim 1 , further comprising defining and configuring one or several virtual sensors in the scene using the first data representation of the scene.

8. A system for detecting activation of a virtual sensor in a scene, the system comprising:

a scene capture sub-system configured to capture a scene using at least one physical sensor and generate a first data representation of the scene;

a virtual sensor sub-system in operative communication with the scene capture sub-system, the virtual sensor sub-system comprising an interface for receiving data representing the first data representation of the scene, a processor, and a memory operatively coupled to the processor, wherein the virtual sensor sub-system is configured for detecting activation of the virtual sensor in the scene, wherein the virtual sensor comprises data representing a three-dimensional geometric form, data representing a position in the scene and data representing one or several trigger conditions, and wherein the detecting activation of the virtual sensor in the scene comprises:

determining whether one or several of the virtual sensor trigger condition is/are fulfilled based on an analysis of the captured first data representation in a first volume area corresponding to the geometric form and position of the virtual sensor,

wherein the determining whether a virtual sensor trigger condition is fulfilled comprises

analyzing point cloud data representing surfaces of objects detected by said at least one sensor in the scene, the point cloud data comprising data representing non-empty points of a second volume area corresponding to the captured first data representation and respective associated position in the second volume area and

determining whether an object of the scene enters the first volume area corresponding to the geometric form and position of the virtual sensor by determining whether a number of non-empty points whose position fall in the first volume area which correspond to the geometric form and position of the virtual sensor is beyond a threshold specified by a trigger condition.

9. A system according to claim 8 , wherein the virtual sensor sub-system includes functionality to define and configure one or several virtual sensors in the scene using the first data representation of the scene.

10. A non-transitory computer-readable storage medium storing computer-executable instructions for detecting activation of a virtual sensor in a scene, the computer executable instructions comprising instructions for:

receiving data representing a three-dimensional geometric form of the virtual sensor, data representing a position of the virtual sensor in the scene, and data representing one or several trigger conditions of the virtual sensor;

receiving data representing a first data representation of the scene captured by at least one physical sensor; and

determining whether one or several of the virtual sensor trigger condition is/are fulfilled based on the analysis of the received data,

wherein the determining whether a virtual sensor trigger condition is fulfilled comprises

analyzing point cloud data representing surfaces of objects detected by said at least one sensor in the scene, the point cloud data comprising data representing non-empty points of a second volume area corresponding to the captured first data representation and respective associated position in the second volume area and

determining whether an object of the scene enters the first volume area corresponding to the geometric form and position of the virtual sensor by determining whether a number of non-empty points whose position fall in the first volume area which correspond to the geometric form and position of the virtual sensor is beyond a threshold specified by a trigger condition.

11. A non-transitory computer program product comprising computer program code tangibly embodied in a computer readable medium, said computer program code comprising instructions to, when provided to a computer system and executed, cause said computer to perform a method for detecting activation of a virtual sensor in a scene,

wherein the virtual sensor comprises data representing a three-dimensional geometric form, data representing a position in the scene and data representing one or several trigger conditions,

the method comprising

determining whether one or several of the virtual sensor trigger conditions is/are fulfilled based on an analysis of data representing a captured first data representation of the scene in a first volume area corresponding to the geometric form and position of the virtual sensor,

wherein the determining whether a virtual sensor trigger condition is fulfilled comprises

analyzing point cloud data representing surfaces of objects detected by said at least one sensor in the scene, the point cloud data comprising data representing non-empty points of a second volume area corresponding to the captured first data representation and respective associated position in the second volume area and

determining whether an object of the scene enters the first volume area corresponding to the geometric form and position of the virtual sensor by determining whether a number of non-empty points whose position fall in the first volume area which correspond to the geometric form and position of the virtual sensor is beyond a threshold specified by a trigger condition.

12. A non-transitory computer-readable storage medium storing a computer program that, when executed, causes a system comprising a processor operatively coupled with a memory, to perform a method for detecting activation of a virtual sensor in a scene,

wherein the virtual sensor comprises data representing a three-dimensional geometric form, data representing a position in the scene and data representing one or several trigger conditions,

the method comprising

determining whether one or several of the virtual sensor trigger conditions is/are fulfilled based on an analysis of data representing a captured first data representation of the scene in first volume area corresponding to the geometric form and position of the virtual sensor,

wherein the determining whether a virtual sensor trigger condition is fulfilled comprises

analyzing point cloud data representing surfaces of objects detected by said at least one sensor in the scene, the point cloud data comprising data representing non-empty points of a second volume area corresponding to the captured first data representation and respective associated position in the second volume area and

determining whether an object of the scene enters the first volume area corresponding to the geometric form and position of the virtual sensor by determining whether a number of non-empty points whose position fall in the first volume area which correspond to the geometric form and position of the virtual sensor is beyond a threshold specified by a trigger condition.

13. The method according to claim 1 , wherein the virtual sensor and trigger condition are chosen based on a way an object of the scene is expected to enter the first volume area which corresponds to the geometric form and position of the virtual sensor.

14. The method according to claim 1 , wherein the size of the virtual sensor is chosen based on an object of the scene which is expected to enter the first volume area which corresponds to the geometric form and position of the virtual sensor.