EP2210417A1

EP2210417A1 - Configuration module for a video surveillance system, surveillance system comprising the configuration module, method for configuring a video surveillance system and computer program

Info

Publication number: EP2210417A1
Application number: EP08804189A
Authority: EP
Inventors: Stephan Heigl
Original assignee: Robert Bosch GmbH
Current assignee: Robert Bosch GmbH
Priority date: 2007-11-12
Filing date: 2008-09-15
Publication date: 2010-07-28
Also published as: CN101855906A; US9549155B2; US20100194859A1; WO2009062770A1; DE102007053812A1; CN101855906B

Abstract

Video surveillance systems typically comprise a plurality of video cameras that are distributed in a surveillance region at different locations. The image data recorded by the surveillance cameras is collected in a surveillance center and automated or evaluated by surveillance personnel. It is known with the automated surveillance that certain image regions of a surveillance camera are selected and continuously monitored by means of digital image processing. A configuration module 6 for a video surveillance system 1 is proposed, comprising a plurality of surveillance cameras 2 a-d, which are and/or can be arranged in a surveillance region 3 at different locations, comprising a model storage 7, which is designed to provide a model 10 of the surveillance region 3, wherein camera data of the surveillance cameras 2 a-d is input and/or referenced in the model 10, further comprising an input interface 11 for the input of at least one surveillance target 5 into the model 10 and comprising a computing device 12, which determines at least one camera-specific surveillance target section 13 a, b, for one of the surveillance cameras 2 a, b based on the surveillance target 5 that was input.

Description

description

title

Video surveillance monitor configuration module, monitoring system with the configuration module method for configuring a video surveillance system and computer program

State of the art

The invention relates to a configuration module for a video surveillance system for configuring a plurality of surveillance cameras, which in one

Surveillance area arranged spatially distributed and / or can be arranged, and a corresponding video surveillance system, a method and a computer program.

Video surveillance systems typically include a plurality of video cameras distributed in a surveillance area. The of the

Surveillance cameras recorded image data are combined in a monitoring center and there automated or evaluated by monitoring personnel. In automated monitoring, it is known that certain image areas of a surveillance camera are selected and constantly monitored by means of digital image processing.

For example, WO2005 / 050971 discloses a video surveillance system having a plurality of surveillance cameras for a building or the like, wherein a surveillance camera is configurable such that exceeding a virtual trip cord placed in the image area of the surveillance camera results in alarm triggering. During configuration, the virtual trip cord is entered into a single video image of the surveillance camera to be configured.

Disclosure of the invention According to the invention, a configuration module with the features of claim 1, a monitoring system with the features of claim 9, a method for configuring a monitoring system with the features of claim 13 and a computer program with the features of claim 14 are disclosed. Advantageous or preferred embodiments of the invention are set forth in the subclaims, the following description and by the figures.

According to the invention, a configuration module which is suitable and / or designed for configuring a video surveillance system with a plurality of surveillance cameras is proposed. The majority of the surveillance cameras are spatially distributed in a surveillance area and / or can be arranged therein. The surveillance area is designed as desired and can be embodied, in particular, as one or more interior spaces in a building, one or more streets, intersections, factory grounds, etc.

The configuration module comprises a model memory, which may be designed as a volatile or as a non-volatile memory and which provides a model of the monitoring area. The model is e.g. formed as an image, a map, a floor plan or the like of the monitoring area and / or includes the geometric dimensions and / or dimensions thereof. In mathematical terms, the model shows the surveillance area arranged in a world coordinate system in a model coordinate representation.

In the model, camera data of the surveillance cameras are registered and / or referenced. In particular, the surveillance cameras are registered in the model in model coordinates in terms of position and orientation.

In one possible embodiment, the model and the camera data are stored permanently in the model memory, with other alternatives, the data mentioned above only for

Running time of the configuration module contracted.

The configuration module has an input interface which is connected and / or connectable, for example, to a human-machine interface MMS, the input interface being used to input at least one monitoring target in the Model is suitable and / or trained. The monitoring goal can z. B. are registered in the model so that this is detected on the detection area, so the area of the video cameras in the surveillance area of at least two of the surveillance cameras or that this extends further than the detection range of the surveillance cameras. In particular, that can

Surveillance target also partially overlapping and partially non-overlapping with the coverage of one of the surveillance cameras are entered. Thus, the monitoring target can be entered into the model regardless of the placement of the surveillance cameras. The monitoring target is preferably entered in model coordinates.

A computing device, as part of the configuration module, is configured to determine at least one camera-specific monitoring target section for at least one of the surveillance cameras on the basis of the input monitoring target. In a preferred embodiment, starting from the registered

Monitoring target, which extends for example over the detection range of at least two surveillance cameras, formed monitoring target sections, each monitoring target section is assigned to exactly one surveillance camera. Preferably, each monitoring target section corresponds to the part of the monitoring destination which is in the detection area of the associated

Security camera is arranged. If the coverage areas of the surveillance cameras overlap, the surveillance target areas of a surveillance destination may also overlap.

A consideration of the invention is that the previously implemented individual configuration of surveillance cameras in a video surveillance system is cumbersome and error prone. In particular, if monitoring targets are to be activated which extend over detection areas of at least two surveillance cameras, the at least two surveillance cameras had to be individually configured so that the transition of the surveillance target between the detection areas actually meets the desired requirements. So far, it has been very expensive to activate a virtual trip cord in a surveillance system that extends beyond the detection range of multiple surveillance cameras, as in a transition from one detection area to the next usually a step or other inconstancy creeps. Also, a change in the position of the monitoring target is - A -

so far cumbersome because each surveillance camera must be reconfigured separately.

In contrast, the invention proposes that the monitoring target is registered in a model and that the conversion of the monitoring target into camera-specific

Monitoring target sections are automated. In other words, a user may, for example, enter a virtual triplet in the model simply by defining two endpoints, with the configuration of one, two, three or more surveillance cameras automatically resulting from the output of the camera-specific surveillance target sections for the surveillance cameras.

The advantages of the invention are thus in a simplified operation and a lower error rate in the operation.

In a preferred embodiment of the invention, the surveillance cameras are arranged stationary, in modified embodiments, the

Surveillance cameras, however, also be designed as movable, in particular PTZ cameras (PanTill zoom cameras), wherein the camera-specific monitoring target sections are created as a function of the current camera position.

In a further preferred embodiment of the invention it is provided that the monitoring target is arranged stationary in the model and / or in the monitoring area. In particular, in this preferred embodiment, the monitoring target is not designed as a moving surveillance object.

In a possible realization of the invention, the monitoring target and / or the monitoring target section is designed as a line or a line, for example with interpolation points and / or interpolation sections, as a surface, in particular as a polygon and / or as a spatial section, in particular a volume. Generally speaking, the monitoring target can be represented as a geometric body, which preferably extends horizontally in the monitoring area, but in modified embodiments shows vertical or arbitrarily oriented sections. In a first possible embodiment, the monitoring target is designed as a virtual trip-rope, in a second possible embodiment the monitoring target is embodied as a surface section which, for example, identifies a blocked area in the monitoring area. It is also possible that the monitoring target is designed as a volume, for example as a cylinder, in whose

Center a protective object is arranged.

In one possible embodiment, the model is designed as a 2-D or as a 3-D model, the 2-D model being realized for example as a plan view of the monitoring area. In a possible 3-D model, on the one hand, the floor plan and stationary objects, such as walls, are modeled. In an extension also quasi-stationary objects, such as cabinets, work machines or the like may be formed in the model.

In particular but not exclusively in the case of the embodiment with a 3-D model, it is preferable for the configuration module to have an occlusion check device which is designed to check a possible occlusion of the monitoring target or the monitoring target section in front of one of the surveillance cameras. One possibility of the implementation is to provide the model or objects in the model with a Z hierarchy and to check whether the

Monitoring target portion is arranged completely or at least partially visible in front of the respective surveillance camera.

In a practical realization of the invention, the camera data entered and / or referenced in the model comprise a camera model, a

Camera position and / or camera orientation. The camera model describes the optical properties, in particular the imaging properties of the surveillance camera and, together with the data on the camera position and / or orientation, allows the projection of the pixels from an image coordinate system of the surveillance camera into the model coordinate system or in the opposite direction.

In a possible concrete implementation of the invention, the calculation device is designed in terms of programming and / or circuitry to determine the monitoring target sections by the monitoring objectives present in model coordinates being in the image coordinate system of the respectively corresponding ones Surveillance camera can be projected and optionally cropped. In particular, the projected surveillance targets are cropped at the edges and / or in areas which are obscured by stationary and / or quasi-stationary objects and / or located outside the image capture area of the respective surveillance camera.

Another object of the invention relates to a video surveillance system having the features of claim 8, wherein the video surveillance system is connected to a plurality of surveillance cameras and / or connectable, wherein the surveillance cameras are spatially distributed and / or arranged in a surveillance area and wherein the video surveillance system, the configuration module as described above or according to one of the preceding claims.

In a practice-oriented development of the invention, the video surveillance system has at least one monitoring device, which is designed programmatically and / or circuitry for monitoring at least one of the monitoring target sections by evaluating the video data streams of the surveillance cameras by means of digital image processing algorithms. The monitoring device is designed in particular for the detection and / or tracking of monitoring objects, object movements or the like in the monitoring target sections.

In one possible architecture, the video surveillance system comprises a plurality of the monitoring devices, each monitoring device being associated with at least or precisely one surveillance camera. This aspect of the invention thus relates to a decentralized distribution of the monitoring devices in the video surveillance system, so that the image data streams of the surveillance cameras can be evaluated at least in part in a parallel architecture.

In a particularly preferred constructional realization of the invention, at least one surveillance camera is designed as a so-called intelligent camera which, for example, has a processor unit in the same housing, wherein the surveillance device is integrated locally in the surveillance camera, in particular in the processor unit. This aspect of the invention allows to easily configure a network of intelligent cameras using the configuration module.

A further subject matter relates to a method for the configuration of a surveillance system or of the surveillance system according to one of the preceding claims. The surveillance target, for example in the form of a geometric figure, is entered in the model of the surveillance area in a first step and in one second step, the monitoring target is converted into the camera-specific monitoring target sections of the individual surveillance cameras.

A last subject of the invention relates to a computer program having the features of claim 13.

Brief description of the drawings

Further features, advantages and effects of the invention will become apparent from the following description of a preferred embodiment. Showing:

Figure 1 is a schematic representation of a video surveillance system with a configuration module as an embodiment of the invention;

FIG. 2 is also a schematic representation of an illustration of the projection of model coordinates into pixel coordinates.

Embodiment (s) of the invention

1 shows a video surveillance system 1, which is wirelessly or wired to a plurality of surveillance cameras 2 a, b, c, d, connected by signal technology. The surveillance cameras 2 a-d are spatially distributed in a real surveillance area, wherein the surveillance area in FIG. 1 is designed as a space 3. In modified embodiments, it is also possible that the monitoring area is designed to be more spacious, twisted, multi-level, etc.

In the surveillance cameras 2 a and 2 b are schematically the detection areas 4 a and 4 b, shown, which are arranged partially overlapping. Also overlapping with the detection areas 4 a, 4 b, a monitoring target 5 is shown, which is designed as a polygon and extends horizontally in the space 3. Such a monitoring target 5 is arranged in a surveillance area, for example, when the surveillance target 5 may not be labeled or crossed by surveillance objects, such as visitors, without an alarm message being output by the monitoring system.

As can be seen from FIG. 1, the monitoring target 5 is completely in the detection area 4 b of the surveillance camera 2 b, but only partially in the area

Detection area 4 a of the surveillance camera 2 a arranged. Without further measures, the monitoring system 1 would have to be configured such that in a first step the monitoring target 5 is entered with regard to the surveillance camera 2b and in a second step with respect to the surveillance camera 2a. It is a particular difficulty that the monitoring objective 5 in the

Detection area 4a only partially protrudes and therefore must be cut off when entering. It is particularly difficult here to enter the transition of the monitoring target 5 in the border or overlap region of the detection areas 4 a and 4 b with sufficient accuracy.

In order to facilitate the configuration of the surveillance cameras 2 a to d, the video surveillance system 1 has a configuration module 6, which allows a simplified configuration of the surveillance cameras 2 a to d.

The configuration module 6 comprises a model memory 7, which is a model 10 of the

Monitoring area, in this case the room 3, with marked surveillance cameras 2 a to d provides. The model 10 is designed as a 2-D model, for example, and can be graphically provided on a display unit 8 to an observer 9. By means of a human-machine interface 11, the observer 9 can easily draw the monitoring target 5 into the model 10, preferably independently of the position of the surveillance cameras 2 a to 2 d.

The configuration module 6 has a computing device 12, which on the one hand to the model memory 7 with the model 10 and the data of the entered monitoring target 5 and based on this data generated camera-specific monitoring target sections 13 a and 13 b.

As is apparent from the representation of the room 3, the monitoring target section 13 b comprises the complete monitoring target 5, since this is completely in the

Detection area 4 b of the surveillance camera 2 b is arranged. On the other hand, the monitoring target section 13 a is formed only as a subset of the monitoring target 5, specifically as the subset recorded by the detection area 4 a of the surveillance camera 2 a.

The conversion of the monitoring target 5 into the monitoring target sections 13 a, b is achieved by means of a projection of the monitoring target 5 from the model coordinates of the model 10 shown on the display device 8 into the image coordinates of the surveillance cameras 2 a to 2 d.

FIG. 2 illustrates the projection of the monitoring target 5 from the model coordinates, which are indicated by a model coordinate system 14, into an image coordinate system 15a or 15b of the surveillance cameras 2 a and 2 b, respectively. The model coordinate system is in a fixed, unique relationship with a world coordinate system in which the space 3 is located.

When projecting the monitoring target 5 onto the image coordinate system 15b of the surveillance camera 2b, the complete surveillance target 5 is imaged onto an image plane in the image coordinate system, the position of the image plane being defined by the image capturing unit of the surveillance camera 2b. The

Mapping specification is generated by the camera data, ie in particular the position and orientation of the surveillance cameras 2a and 2b and a camera model which describes the optical imaging properties.

In a further step, the imaged monitoring target 5 is trimmed at the edge, wherein the intersection boundaries are defined by the detection range of the surveillance cameras 2 a, b, ie ultimately by the (effective) size of the image capturing unit or another aperture of the surveillance cameras 2 a, b. Optionally, a Verdeckungsprüfvorrichtung 16 may be provided in the configuration module 6, which is a possible occlusion of the Monitoring target 5 checked by other objects in the model 10 and disabled in a cover other areas, especially the hidden areas of the monitoring target 5. The remainder of the imaged monitoring target 5 is the monitoring target portion 13 a, b, which can be observed by the surveillance cameras 2a and b, respectively. The

Monitoring target sections 13 a, b correspond to real areas in the space 3 in world coordinates.

The position and extent of the monitoring target sections 13 a, b in image coordinates is output to monitoring devices 17a-d, which are decentralized in the

Surveillance cameras 2 a-d are integrated. For example, the surveillance cameras 2 a-d are designed as so-called "intelligent cameras" which have integrated and / or individually assigned processor units in their own housing.

On the monitoring devices 17 a-d, the monitoring target sections 13 a, b are monitored by means of digital image processing for events detectable by image processing algorithms during monitoring operation. Such events may include entering, crossing, touching, etc. of the monitoring target sections 13a, b, color, contrast, content changes, and / or object detection, or recognition in the

Monitoring target sections 13 a, b be.

Claims

claims

1. configuration module (6) for a video surveillance system (1) with a plurality of surveillance cameras (2 a-d), which are spatially distributed in a surveillance area (3) and / or can be arranged,

with a model memory (7), which is designed to provide a model (10) of the monitoring area (3), in which model (10) camera data of the surveillance cameras (2 a-d) are registered and / or referenced,

with an input interface (11) for inputting at least one monitoring target (5) into the model (10) and

with a calculation device (12) which, on the basis of the entered monitoring target (5), has at least one camera-specific monitoring target section

(13 a, b) for at least one of the surveillance cameras (2 a, b) determined.

2. Configuration module (6) according to claim 1, characterized in that the monitoring target extends over the detection range of at least two of the surveillance cameras and the calculation device (12) based on the input monitoring target (5) camera-specific monitoring target sections (13 a, b) for the at least two surveillance cameras (2 a, b) determined.

3. Configuration module (6) according to claim 1 or 2, characterized in that the monitoring target (5) is arranged stationarily in the monitoring area.

4. Configuration module (6) according to any one of the preceding claims, characterized in that the monitoring target (5) and / or the monitoring target section (13 a, b) as a line, polyline, surface and / or a space portion is formed.

5. Configuration module (6) according to one of the preceding claims, characterized in that the model (10) is designed as a 2D or a 3D model.

6. Configuration module (6) according to any one of the preceding claims, characterized by a Verdeckungsprüfvorrichtung (16), which is designed to check a possible occlusion of the monitoring target (5) in front of one of the surveillance cameras (2 a-d).

7. Configuration module (6) according to one of the preceding claims, characterized in that the camera data is a camera model, a camera position and / or

alignment.

8. Configuration module (6) according to one of the preceding claims, characterized in that the calculation device (12) is formed programmatically and / or circuitry, the monitoring target sections (13 a, b) by a

Projection of the monitoring target (5) in the image coordinate system (15 a, b) of the corresponding surveillance camera (2 a-d) to produce.

9. Video surveillance system (1) with a plurality of surveillance cameras (2 a-d), which are spatially distributed in a surveillance area (3) and / or arranged, characterized by a configuration module (6) according to one of the preceding claims.

10. Video surveillance system (1) according to claim 9, characterized by at least one monitoring device (17 a-d), which is designed for monitoring at least one of the monitoring target sections (13 a, b) by means of digital image processing algorithms.

11. A video surveillance system (1) according to claim 9 or 10, characterized by a plurality of the monitoring devices (17 a-d), each one

Monitoring device (17 a-d) at least or exactly a surveillance camera (2 a-d) is assigned,

12. Video surveillance system (1) according to one of the preceding claims, characterized in that at least one surveillance camera (2 ad) as intelligent camera is formed and / or the monitoring device (17 ad) is integrated locally in the surveillance camera (2 ad).

13. A method for configuring the monitoring system (1) according to one of the preceding claims,

wherein the monitoring target (5) is entered in the model (10) of the monitoring area (3),

wherein the monitoring target (5) in at least one camera-specific

Monitoring target section (13 a, b) is converted.

14. Computer program with program code means for performing all the steps of the method according to claim 13, when the program is performed on a computer or the video surveillance system according to one of claims 1 to 12.