Method for monitoring a monitored region recorded by a camera

Abstract

The invention relates to a method for monitoring a monitored region (10) recorded by a camera, wherein a content analysis is performed for a sub-region (20) of the monitored region (10); wherein the sub-region (20) is determined in dependence on one or more parameters; and wherein the determination of the sub-region (20) is performed anew when at least one of the parameters changes during the monitoring.

Claims

1. A method for monitoring a monitored region recorded by a camera, the method comprising: determining, with an electronic processor, a sub-region of the monitored region based on one or more parameters; in response to a change in at least one of the one or more parameters, modifying the sub-region of the monitored region; and performing, with the electronic processor, a content analysis on images recorded by the camera, wherein the content analysis is performed for the sub-region of the monitored region and is not performed for an inactive region of the monitored region.

2. The method according to claim 1, wherein the determination of the sub-region is automatically performed when at least one of the parameters changes during monitoring.

3. The method according to claim 1, wherein the sub-region is based on a planar surface in the monitored region.

4. The method according to claim 3, wherein the sub-region is determined on the basis of the planar surface in the monitored region and height information.

5. The method according to claim 1, wherein the one or plurality of parameters comprises an output value of a sensor.

6. The method according to claim 1, wherein the one or plurality of parameters comprises at least one image processing parameter in the content analysis.

7. The method according to claim 1, wherein the sub-region comprises a continuous region or at least two regions separated from one another.

8. The method according to claim 1, wherein the content analysis for a sub-region of the monitored region, the determination of the sub-region, or both are performed in the camera or in a higher-level monitoring system when at least one of the parameters changes during the monitoring.

9. A system for monitoring a monitored region, the system comprising: a camera installed in the monitored region and configured to record images of the monitored region, the camera including an electronic processor configured to determine a sub-region of the monitored region based on one or more parameters, in response to a change in at least one of the one or more parameters, modify the sub-region of the monitored region, and perform a content analysis on the images recorded by the camera, wherein the content analysis is performed for the sub-region of the monitored region and is not performed for an inactive region of the monitored region.

10. A non-transitory computer-readable medium including instructions that, when executed by an electronic processor, cause the electronic processor to perform a set of functions, the set of functions comprising: determine a sub-region of the monitored region based on one or more parameters; in response to a change in at least one of the one or more parameters, modify the sub-region of the monitored region; and perform a content analysis on images recorded by a camera, wherein the content analysis is performed for the sub-region of the monitored region and is not performed for an inactive region of the monitored region.

11. The method according to claim 1, wherein performing the content analysis includes at least one selected from a group consisting of performing object detection, performing object tracking, performing object classification, and checking a plausibility of an object.

12. The system according to claim 9, wherein the inactive region is masked out.

13. The system according to claim 9, wherein the monitored region includes the sub-region, the inactive region, and an irrelevant region.

14. The system according to claim 13, wherein the irrelevant region is a region of the monitored region in which no alarm or metadata are generated for.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) FIG. 1 shows a monitored scene and a region selected thereon for the content analysis for a method which is not according to the invention.

(2) FIG. 2 shows a monitored scene comprising different sub-regions selected thereon for the content analysis for a method according to the invention in a preferred embodiment.

DETAILED DESCRIPTION

(3) A monitored scene comprising a monitored region 10, by way of example an entry area to a super market, is depicted in FIG. 1. The complete monitored region 10 is shown in FIG. 1a.

(4) In FIG. 1b, a sub-region 20 of the monitored region 10 is selected, which is used for the content analysis during the image processing in a camera. The hatched region is excluded from the content analysis in this case because no people are allowed to gather there or, respectively, said region is not relevant, as a comparison with FIG. 1a shows.

(5) The entry area is only of interest particularly, e.g., when counting people who enter and/or leave the super market. The disadvantage here is, however, that the entire entry area is permanently used for the content analysis irrespective of whether or not people are actually currently present in the entry area.

(6) The same monitored region 10 as in FIG. 1 is shown in FIG. 2. The selected sub-regions 20, which are used for the content analysis, are different from one another, as explained below.

(7) In FIG. 2a, the sub-region 20 extends to the region in front of the entrance or, respectively, the front region of the entrance, i.e. up to the gates, which are placed in the passage area of the entrance. Because there are no people present in the passage area when the gates are closed, a content analysis is also not necessary in this region.

(8) In FIG. 2b, the sub-region 20 additionally extends to the right passageway. In this case, two sensors 30 are depicted, which are configured on the gates in the passageway. The sensors can, e.g., output different values as parameters depending on whether the corresponding gate is open or closed. A gate is usually opened if a person walks up to it and wants to pass through said gate. A corresponding value of the sensors 30 is then emitted which is recorded by the camera and is used to determine respectively define the sub-region 20.

(9) In the present case, the sub-region is expanded to include the right passageway because people are or at least may be present there. The sub-region is thus only expanded to the larger region if this is necessary on the basis of the monitoring, i.e. if people may be present there.

(10) It can furthermore be seen that the sub-region 20 is based on a planar surface, namely the floor, of the monitored region 10. People usually move only upon said surface. The surface is preferably marked manually in the camera image, on the basis of which the sub-region to be analyzed is automatically determined by the camera or a higher-level monitoring system superior to the camera.

(11) In FIG. 2c, the sub-region 20 is, opposed to FIG. 2a, expanded to the left passage area. In this case, two sensors 30, which are configured on the gates in the passageway, are likewise depicted. The sub-region is again only expanded by a region, in which people are located. In the present case, this relates to people who are passing through the left passageway.

(12) The application of a method according to the invention is not exhausted merely by the monitoring of people. Other moving objects, such as, e.g., vehicles, can likewise be correspondingly monitored.

(13) Suitable sensors for outputting values as parameters, which are used to determine the sub-region, can also be situated on other objects or articles which are suitable for defining possible monitored regions. This applies here, for example, to doors, in particular sliding doors or revolving doors, as they are, e.g., to be found at the entrance to department stores. Sensors at light barriers can also emit corresponding values.

(14) A further option for parameters are, e.g., image processing parameters during the content analysis. In this case, inactive regions can, e.g., be identified as background and furthermore be excluded from the content analysis. It is, e.g., likewise conceivable that a detected direction of movement of an object expands the sub-region in the corresponding direction.