WO2011110268A1

WO2011110268A1 - Checkpoint with a camera system

Info

Publication number: WO2011110268A1
Application number: PCT/EP2011/000627
Authority: WO
Inventors: Gerhard Niebling
Original assignee: Muehlbauer Ag
Priority date: 2010-03-12
Filing date: 2011-02-10
Publication date: 2011-09-15
Also published as: DE102010011225B3

Abstract

System for determining the number of people inside an area to be monitored, having: a lighting device for lighting a floor of the area to be monitored with electromagnetic radiation, wherein at least some of the electromagnetic radiation has a wavelength outside the visible light spectrum; a first camera for detecting electromagnetic radiation emitted by the lighting device and for converting said radiation into corresponding image information, wherein the spectral sensitivity of the first camera is outside the visible light spectrum; and an evaluation unit for determining whether one or more people are inside the area to be monitored using the image information acquired by the first camera.

Description

Personal check-in with camera systems

description

background

A device and a method for determining the number of persons in a person passage control is described. In particular, the device and the method are suitable for making a statement as to whether more than one person is within an area to be controlled.

In order to meet security requirements for access to buildings, building areas, border crossings, etc., automatic access control systems are increasingly being used. These access control systems are designed to ensure that only authorized persons are granted access to closed / protected areas. The automatic access control systems may replace the security checks carried out by security staff, for example at border crossings or in front of buildings with special security requirements.

A problem of all access control systems is to ensure that only identified and authorized persons pass through the access control system. In order to prevent an unauthorized person from passing through the access control system simultaneously with an authorized person, it is usually necessary for security personnel to monitor the access control system.

However, this type of monitoring is very labor-intensive and, in particular in the case of several parallel access control systems, is not always satisfactory. Therefore, systems are needed that automatically monitor access control systems. One possibility is to design the access control system as a lock, which can / may only be entered by one person. By monitoring the number of people inside the lock, the lock can be locked if more than one person is in the lock. This ensures that the access control system can only pass if only one person within the access control system If this person is correctly identified and the identified person is also authorized to pass the inspection.

Thus, for example, document DD 289 837 A5 describes a method and arrangement for the automatic recognition of the number of persons. A personal airlock is monitored by an electronic camera mounted on the ceiling. This electronic camera is connected to an image processing computer. In order to distinguish the states "empty", "1 person" and "> 1 person", the intensity, texture, shape and structure properties are extracted from the image information captured by the electronic camera and evaluated In the case of operation of the personal airlock, the features of the standardized reference image are then compared with the characteristics of the currently determined images.

A major disadvantage of this system, however, is that the system must be adapted to the respective lighting conditions. In addition, it must be ensured that the lighting conditions remain constant. If the lighting conditions change, the system parameters must be adjusted accordingly. Thus, this system is only suitable for closed rooms with constant lighting.

Document DE 10 2007 041 333 A1 describes a contactless counting system. In this case, a reference area which has increased contrast properties with respect to the surroundings is illuminated by a light source. A camera creates images of the reference region at successive times, and an evaluation device merges the images along a time axis into a flow image. As a result, images of a counting object appearing on the images are connected to each other along the time axis to form a cluster represented on the flowchart. A tearing off or emergence of a cluster in the direction of the time axis of the flowchart is recognized as an event to be counted.

Document DE 10 2004 011 780 AI describes an access control device. By a light emitting radiation emitting element and a light receiving radiation receiving element, a light field is formed. Since the light radiation is emitted as a flat light field, a complete and complete monitoring of an access area is created. Document CH 695 123 A5 describes a device and method for the automatic detection of an object and / or a person by image signal processing. In order to increase the evaluation speed and to improve the quality of the evaluation, a high-contrast background is proposed, which has, for example, a checkerboard-like pattern.

However, many people perceive locked locks as restrictive. In addition, such security locks are difficult to monitor by security personnel. Therefore, access control systems are often used whose barriers are approximately waist high. Controlled persons find such barriers less restrictive. In addition, security personnel can better understand the access control system and intervene more quickly in the event of disruptions, vandalism or sabotage attempts. In addition, there is often the requirement to install the access control systems in existing rooms with daylighting. However, since the brightness of the daylight illumination has strong fluctuations, such illumination makes image processing more difficult.

problem

Thus, the object is to provide a system that is suitable for reliably determining the number of persons within a controlled area of an access control system. In addition, the system should be insensitive to changes in lighting conditions and be used in rooms that are illuminated with natural light.

Suggested solution

A system is proposed for determining the number of people within a controlled area. The system includes an illumination device for illuminating a bottom of the region to be controlled with electromagnetic radiation, wherein at least a portion of the electromagnetic radiation has a wavelength that is outside of the visible

Light spectrum lies; a first camera for detecting electromagnetic radiation emitted by the illumination device and for converting to corresponding image information, the spectral sensitivity of the first camera being outside the visible light spectrum; and an evaluation unit for determining, on the basis of the image information captured by the first camera, whether one or more several people are within the area to be controlled. In this case, the illumination device is set up to illuminate only one illuminated area, so that all objects which are located above the illuminated area appear as dark objects in the image information captured by the first camera.

In the present system, since at least a portion of the electromagnetic radiation emitted by the illumination device and the spectral sensitivity of the camera are outside the spectrum of visible light, the present system is substantially insensitive to daylight or ambient light. Therefore, the system can also be used in rooms that are illuminated with changing daylight and / or uneven artificial light. In addition, the system parameters do not have to be adapted to the ambient light conditions. Thus, the system is also suitable for mobile applications, for example. Another advantage is that the present system is insensitive to the color of the clothing or the hair or skin color of the persons to be checked. Therefore, the evaluation of the image information is much easier than with systems that work in the light spectrum.

In addition, to solve the problem, a method is proposed for determining the number of persons within an area to be controlled, the method comprising: illuminating the floor of the area to be controlled with electromagnetic radiation whose wavelengths are at least partially shorter or longer than the wavelengths of visible light; Acquiring image information with a first camera, wherein a plurality of pixels are each assigned a brightness value, and wherein the image information corresponds to a first image of the region to be controlled; Determining the objects in the first image of the region to be controlled, the objects corresponding to dark regions located in the first image of the region to be controlled; Determining the size of the objects, wherein the size of an object corresponds to the number of pixels whose brightness value is smaller than the brightness threshold and which are arranged side by side in the first image of the area to be controlled; Comparing the size of an object with a first size threshold; Outputting a first signal corresponding to the information "1 person" if the object is smaller than the first size threshold, and outputting a second signal corresponding to the information "> 1 person" if the object is larger than the first sizes Threshold value is. Design and characteristics

The illuminated area may include the floor and an area above the floor. The area above the ground can be knee-high. Alternatively, it is also possible that the area above the ground has a height between about 5 cm and about 30 cm. In one embodiment, the area above the floor may have a height of between about 10 cm and about 20 cm.

The first camera may have a daylight filter, so that the spectral sensitivity of the first camera comprises a wavelength range of about 800 nm to about 1800 nm. In one embodiment, the spectral sensitivity of the first camera may include a wavelength range from about 800 nm to about 100 nm.

The spectrum of the radiation emitted by the illumination device can be adapted to the spectral sensitivity of the first camera. For example, the illumination device can emit infrared radiation in a wavelength range from about 800 nm to about 1800 nm. However, the wavelength spectrum of the electromagnetic radiation emitted by the illumination device can also be substantially wider than the spectral sensitivity of the first camera.

The first camera may be centrally located over the area to be controlled such that an optical axis of the first camera is substantially perpendicular to the floor of the area to be controlled. The first camera can detect the outline / shadow of a person who is within the area to be controlled.

The first camera can also be arranged laterally to the area to be controlled. In this case, at least one deflection mirror can be attached to a ceiling of the area to be controlled, so that the electromagnetic radiation reflected from the floor of the area to be controlled is deflected in the direction of the first camera.

The at least one deflecting mirror may have a curvature. The curvature of the at least one deflection mirror can be convex. In addition, at least a deflection mirror be arranged and curved so that the first camera can fully capture the area to be controlled.

The bottom of the region to be controlled may contain color pigments, so that the bottom of the region to be controlled in particular reflects the part of the electromagnetic radiation whose wavelength corresponds to the spectral sensitivity of the first camera. In this case, the bottom of the area to be checked in the image information appear as a bright area and the shadow of a person who is located between the bottom of the area to be controlled and the first camera, appear in the image information as a dark area.

In addition, a second and a third camera may be provided, which are arranged so that optical axes of the second and third camera include an angle with the bottom of the area to be controlled, which is not equal to 90 °.

The second and third cameras can capture image information corresponding to a side view.

An access control system may comprise the present system for determining the number of persons located within an area to be controlled, wherein the illumination device may be arranged to illuminate only a lower portion of the area to be controlled.

In the access control system, the area to be controlled may be bounded by lateral boundary walls and first and second barriers; zusätz ^¬ Lich a Personenidentifikati ^¬ onsvorrichtung may be located within the region to be inspected. In this case, the access control system may allow a person access to the restricted area if the system for determining the number of persons has determined on the basis of the image information that a person is within the area to be controlled and the personal identification device has determined that the person is entitled to enter the closed area.

The step of determining the objects of the present method may comprise the following two substeps: comparing the individual brightness values of the plurality of pixels with a brightness threshold value; and summarizing the Pixels whose brightness value is smaller than the brightness threshold and which are juxtaposed in the first image, to objects.

In the present method, a person within the area to be controlled may be prompted, for example by an optical display, to place a piece of luggage at a specific position within the area to be controlled if the object is greater than the first size threshold. This particular position is to be defined so that it is separated from the place of the person. After this request, the number of pixels of the objects can then be compared to a second size threshold if more than one object within the area to be controlled is detected. In this case, the objects whose number of pixels is smaller than the second size threshold can be classified as baggage items.

In the present method, a signal corresponding to the information "> 1 person" may be output when at least two objects have a number of pixels larger than the second size threshold.

In the present method, a position of the objects can be monitored. Where the objects whose position changes, are classified as moving objects, and the objects whose position does not change, are classified as rigid objects. In addition, a signal corresponding to the information "> 1 person" can be output if two moving objects within the area to be controlled are detected.

In the present method, additional image information can be acquired with a second and a third camera. In this case, the additional image information can correspond to a second and a third lateral image of the persons and objects within the area to be monitored.

In the present method, the image information acquired by the second and third cameras may be used to estimate the height of the persons and objects within the area to be controlled.

In the present method, the estimated altitude may be used to detect an attempt by the person to skip a barrier of an access control system. Brief description of the drawings

Other objects, features, advantages and applications will become apparent from the following description of non-limiting embodiments to be understood and the accompanying drawings.

Fig. 1 shows a sectional view of an access control system having a first embodiment of a system for determining the number of persons;

Fig. 2 is a graph showing image information detected in the system shown in Fig. 1;

3 shows a flow chart with a method for evaluating the image information.

Fig. 4 is a side view of an access control system including a second embodiment of a system for determining the number of persons;

Figures 5a, 5b, 5c are graphs of image information detected in the system shown in Figure 4;

Detailed description of embodiments

Fig. 1 shows a sectional view of an access control system 12 having a first embodiment of the present system. The access control system 12 has a substantially U-shaped profile. The access control system 12 comprises lateral boundary walls 14, 16, a bottom 18 and a first and a second barrier 20, 22 (shown in FIG. 2). In this case, the space enclosed by the lateral boundary walls 14, 16 and the first and second barriers 20, 22 forms a region 24 to be controlled. This region 24 to be monitored is monitored by a system for determining the number of persons. This system comprises an illumination device 26, a first camera 28 and an evaluation unit 30.

The lighting device 26 is arranged in the lower area of the access control system in such a way that it adjusts the floor 18 of the area 24 to be monitored. shine. In addition, the lighting device 26 irradiates an area above the floor 18 which is approximately knee-high. The illuminated area may also be lower and, for example, have a height of about 10 cm to about 20 cm. This is shown in Fig. 1 by the dashed lines. The floor 18 and the illuminated area above the floor 18 are referred to as illuminated area 25. The lighting device 26 is designed so that the illuminated area 25 is illuminated as evenly as possible. Alternatively, it is also possible to integrate the lighting device 26 in the bottom 18. For example, the floor 18 may be designed as a surface radiator.

The illumination device 26 generates electromagnetic radiation whose wavelengths are at least partially outside the visible light spectrum. For example, the illumination device 26 may emit ultraviolet (UV) radiation or infrared (IR) radiation.

The bottom 18 of the area 24 to be controlled may contain color pigments to reflect the electromagnetic radiation emitted by the illumination device 26. In this case, the color pigments can be selected such that they primarily reflect the electromagnetic radiation whose wavelength corresponds to the spectral sensitivity of the first camera 28. Color pigments which, for example, reflect IR radiation can be obtained, for example, from BASF under the trade names Sicotan®, Sicopal® or Sicotrans®.

The first camera 28 is arranged centrally above the area 24 to be controlled. In this case, the first camera 28 has a spectral sensitivity which lies outside the range of the visible light spectrum. For example, a digital camera based on silicon sensors can be used whose spectral sensitivity covers a range from 400 nm to 100 nm. In order to limit the spectral sensitivity of the first camera 28 to a wavelength range outside the visible light spectrum and thus to avoid disturbances by daylight, a daylight filter (29) which is impermeable to radiation having a wavelength between 400 nm and 800 nm can be used , But it is also possible to use a digital camera whose spectral sensitivity is also without daylight filter (29) in the IR spectrum. For example, the spectral sensitivity of germanium photodiodes is in the range of 800nm to 1800nm. Since the first camera 28 is arranged centrally above the area 24 of the access control system 12 to be checked, the first camera 28 can acquire image information which corresponds to a plan view of the area 24 to be monitored. Since the floor 18 reflects the radiation emitted by the lighting device 26, the area 24 to be inspected appears as a bright area in the plan view shown in FIG. Since in the illustrated embodiment the lateral boundary walls 14, 16 and the first and second barriers 20, 22 do not reflect the electromagnetic radiation emitted by the illumination device, they appear as dark areas.

If a person 32 or an object 34 is located within the area 24 to be controlled, as shown in FIG. 2, the person 32 or the object 34 covers the electromagnetic radiation reflected from the floor 18. Thus, the camera 28 detects the shadow of the person 32 or the object 34 and they appear as dark areas. The dark areas correspond to the outline of the person 32 or the object 34th

Since the spectral sensitivity of the camera 28 is outside the wavelength range of visible light, the image information is not affected by ambient light or by daylight. Therefore, the present system can also be used in rooms that are illuminated with natural light. In this case, the intensity and the spectrum of the emitted electromagnetic radiation should be chosen such that the image information is not impaired by IR rays or UV rays contained in the sunlight. Likewise, the intensity and the wavelength spectrum can be selected such that the image information is not affected by the IR radiation emitted by persons due to their body temperature. Thus, the intensity of the illumination can be chosen such that all persons / objects which are located above the illuminated area 25 appear in the image information as dark areas.

In an embodiment not shown in detail, the first camera 28 may also be arranged laterally to the area to be controlled 24 and detect the area to be controlled 24 via one or more deflecting mirrors. The first camera 28 is aligned obliquely upward on the one or more deflection mirrors. In this case, the first camera 28 can be integrated in a lateral boundary wall 14, 16. The area to be controlled 24 is observed via one or more deflecting mirrors, which at the ceiling of the room above the zu controlling area 24 are attached. The mirror elements are preferably aligned so that the camera viewing angle of each element is directed vertically downwards. This avoids or reduces overlapping of objects, as occurs with a camera from above, especially at low observation height and a large viewing angle. In order for the composite image of the region 24 to be controlled to have no gaps from the individual mirror elements, it is advantageous to make the mirrors slightly convex.

The image information captured by the first camera 28 can be evaluated in the evaluation unit 30 using the methods illustrated in FIG. The evaluation unit 30 can be a computer equipped with image processing software. Furthermore, it is possible that the evaluation unit 30 is designed as a central evaluation unit, in which the image information of several systems for determining the number of persons is evaluated. This central evaluation unit can then provide information on the number of persons in the respective access control system to a corresponding number of access control systems. It is also possible that the evaluation unit 30 is part of a control unit of the access control system.

In step S100, the image information is acquired, in which case a plurality of pixels are each assigned a brightness value. This image information is then evaluated in the evaluation unit 30.

In step S110, the objects that are within the area to be controlled 24 are determined. An object consists of pixels whose brightness value is smaller than a brightness threshold and which are arranged side by side in a graphical representation of the image information. Thus, the particular objects correspond to the contours of the persons 32 and the objects 34 within the area 24 to be controlled.

In addition, the image information can also be binarized. This means that a first brightness value, for example 255, is assigned to the pixels whose brightness value is greater than the brightness threshold value, and a second brightness value, for example 0, to the pixels whose brightness value is none than the brightness threshold value. is assigned. By binarizing the image information, the following steps can be performed with less computational effort. In step S120, the size of the objects is determined. The number of pixels of an object corresponds to the size of the object.

In step S130, the number of objects is checked. If there is only one object within the area to be controlled 24, the process proceeds to step S140. If a plurality of objects are within the area to be controlled 24, the process continues in step S170.

In step S140, the size of the object is compared to a first size threshold. In this case, the first size threshold value is selected such that it is possible to reliably distinguish between one person and several persons. For example, two touching persons represented as an object may be distinguished by one person because the number of pixels of the two persons exceeds the first size threshold. The size threshold can be determined, for example, based on statistical data.

If the number of pixels of the object is smaller than the first size threshold value, the evaluation unit 30 can forward a signal to the access control system 12, which corresponds to the information "1 person", in step S180 a display unit of the access control system is prompted to identify itself, After successful identification and after verification of the authorization, the person can then pass the access control system 12.

If the number of pixels of the object exceeds the first size threshold, the process proceeds to step S150. In this case, the evaluation unit 30 can output a signal to the access control system 12, which corresponds to the information "> 1 person".

One cause for exceeding the first size threshold may be that the person 32 is carrying a piece of luggage 34. In the graphical representation of the image information, this results in some of the person's corresponding pixels being directly adjacent to some of the pixels corresponding to the item of luggage 34. Therefore, the person 32 and the luggage 34 are detected as an object. To exclude this, in step S150, in response to the "> 1 person" signal, the person is prompted, for example, via the above-mentioned display unit, all persons. to leave baggage at certain positions within the area 24 to be inspected. These positions are chosen such that in the graphical representation of the image information, the person 32 and the luggage pieces 34 are detected as two objects.

If two objects within the area to be controlled are detected in step S160, the process proceeds to step S170. If only one object continues to be detected in step S160, further information can be evaluated. Thus, for example, in an identification document of the person data about the size and weight of the person can be stored. This data can then be used to adjust the first size threshold accordingly. Alternatively, it is also possible that the process is aborted and the person is requested to undergo control by the security personnel.

In step S170, the number of pixels of each object is compared with a second size threshold. In this case, objects whose number of pixels is smaller than the second size threshold, as pieces of luggage

classified. If this comparison shows that the number of pixels of a first object exceeds the second size threshold value and the number of pixels of a second object or of each further object does not exceed the second size threshold value, the evaluation unit 30 can output a signal which is the Information "1 person" corresponds.

If this comparison yields a different result, for example that the number of pixels of two objects exceeds the second size threshold, then further differentiation criteria can be used. These may, for example, be the shape of the objects or positional changes of the objects within the area to be controlled. Alternatively, the process can also be aborted. The person is then asked to undergo an inspection by the security personnel.

In order to increase the reliability of the system, additional criteria can be evaluated in addition to the method described above. For example, the position of a person can be monitored. Position changes correspond to movements of the objects. For example, two objects moving independently of each other are an indication that two persons are within the area to be controlled. In this case, a first object that is in moving into the area to be controlled, to be classified as 1st person. If a second object subsequently moves into the area to be controlled, it can be assumed that two persons are within the area to be controlled, regardless of whether the objects move simultaneously or successively.

Fig. 4 shows a side view of an access control system having a second embodiment of a system for determining the number of persons. The second embodiment may be an extension of the first embodiment or a system independent of the first embodiment.

Fig. 4 shows a person 42 who is in an access control system. The access control system can correspond to the access control system shown in FIGS. 1 and 2. The access control system is passed in the direction of arrow 44. However, it can be seen that the access control system can also be passed in the opposite direction or in both directions.

The lateral boundary wall 14 shown in FIG. 4 is approximately waist-high. Low boundary walls increase the acceptance of automatic access control systems. In addition, the security staff can better survey them and thus react faster to disturbances or sabotage attempts. Furthermore, low containment walls create a less restrictive impression on the controlled persons than fully enclosed spaces.

4 shows three cameras 46, 48, 50. The first camera 46 may correspond to the first camera 28 of the first embodiment. The first camera 46 is centrally located over the area of the access control system to be controlled. The second and third cameras 50 are arranged laterally, for example one above the first barrier 20 and one above the second barrier 22. The three cameras 46, 48, 50 may also be arranged along a straight line. The optical axes of the second and third cameras 48, 50 are inclined so that the optical axes and the bottom of the area to be controlled 24 enclose an angle other than 90 °. This angle is called α. In FIG. 4, for the better illustration of the angle α auxiliary lines are drawn, which are parallel to the floor of the access control system. The angles of inclination of the second and the third camera 48, 50 may be the same or different. The same inclination angles allow a simpler evaluation of the image information. However, it can be in In some cases, due to the shape or arrangement of the access control system, it may be more useful to use different angles of inclination.

The three cameras 46, 48, 50 shown in FIG. 4 can be insensitive to the visible light spectrum analogous to the camera 28 shown in FIG. However, it is also possible that the spectral sensitivity of the second and third camera 48, 50 is substantially in the wavelength range of visible light and thus these cameras are substantially insensitive to the radiated by the illumination device 26 electromagnetic radiation. In the embodiment described here, however, all three cameras 46, 48 and 50 have the same spectral sensitivity, which is substantially outside the wavelength range of visible light. In particular, when using three identical cameras, the evaluation of the image information is easier, since, for example, the same brightness threshold value can be used.

The image information captured by the cameras 46, 48 and 50 is shown in FIGS. 5a, 5b and 5c, wherein FIG. 5a illustrates the image information captured by the first camera 46, FIG. 5b illustrates the image information captured by the second camera 48 and Fig. 5c illustrates the image information captured by the third camera 50. Corresponding to the arrangement and the angle of inclination of the cameras, Fig. 5a corresponds to a plan view and Figs. 5b and 5c corresponds to a side view.

5a shows two objects, wherein object 52 corresponds to a person and object 54 to a piece of luggage parked next to the person. Due to the perspective of the lateral view, the item of luggage is depicted in FIG. 5b at the level of the upper body of the person. In addition, the person and the luggage are represented as an object. Due to the perspective of the lateral view, the item of luggage is depicted in FIG. 5c at the level of the person's feet. Again, person and baggage are represented as one object. By imaging the item of luggage at the level of the person's feet, the length B of the object in FIG. 5c is greater than the length A of the object in FIG. 5b.

The image information of the second and third cameras 48 and 50 can be used to estimate the height of the objects. For this purpose, for example, the lengths A and B of the objects shown in FIGS. 5b and 5c may be used. The detection of the objects is carried out as in the method shown in connection with FIG. 3. To determine the lengths A and B, for example, the positives values of two pixels in the graphical representation are subtracted from each other. Depending on the format of the image information, the position values of the X axis or the Y axis can be used. In this case, one pixel is arranged on the right and the other pixel on the left edge of the object. The amount of the difference then corresponds to the length of the object.

After the lengths A and B have been determined, the two length values are compared. The smaller length value is then multiplied by the tangent of α and a correction factor. The correction factor can be determined, for example, from empirical data.

The thus estimated height can then be used as an additional security feature. For example, the estimated value may be compared with biometric data stored in an identification document. In addition, the estimation of the altitude can be repeated in an infinite loop. Since a sudden change in altitude is a clear indication that a person is trying to skip a barrier of the access control system, the altitude can be monitored to detect a skip. In the event of a sudden change in altitude, an alarm is triggered which alerts the security personnel to an access control system being skipped. Likewise, maximum and minimum allowable quantities can be defined and monitored as thresholds.

From the image information captured by the second and third cameras 48 and 50, additional information about objects concealed in the image information of the first camera 46 can be detected. This is particularly relevant if the acquisition of the image information is affected by the spatial conditions. For example, the observation distance may not be optimal so that the objects obscure each other due to the wide viewing angle of a camera.

It is also possible to carry out the method illustrated in connection with FIG. 3 in succession for the image information captured by the cameras 46, 48 and 50. In this case, the evaluation unit 30 can output a signal to the access control system which corresponds to the information "1 person" if at least the evaluation of two of the three recorded image information leads to the result "1 person". It will also be appreciated that the second embodiment may be used without the first embodiment. Thus, it can be seen that the second and third camera 48, 50, a corresponding lighting device and a corresponding evaluation unit can be used as a separate system. For example, this system can be retroactively integrated into existing access control systems to provide a skip control.

Although the above-explained details of the method and the device are shown in conjunction; It should be noted, however, that they are at least as far as individual claims directed to them, are also independent of each other and are also freely combinable.

Claims

claims

A system for determining the number of persons located within an area to be controlled (24), comprising:

an illumination device (26) for illuminating a bottom (18) of the region (24) to be monitored with electromagnetic radiation, wherein at least part of the electromagnetic radiation has a wavelength which lies outside the visible axis spectrum;

a first camera (28, 46) for detecting electromagnetic radiation emitted by the illumination device (26) in the area to be controlled (24) and converting to corresponding image information, the spectral sensitivity of the first camera (28, 46) being outside the visible Light spectrum lies; and

an evaluation unit (30) for determining, based on the image information acquired by the first camera (28, 46), whether one or more persons are located within the area (24) to be controlled;

wherein the illumination device (26) is adapted to illuminate only one illuminated area (25) so that all objects located above the illuminated area (25) in the image information captured by the first camera (28) are rendered dark objects appear.

2. System according to claim 1, wherein the spectrum of the radiation emitted by the illumination device (26) radiation is adapted to the spectral sensitivity of the first camera (28).

3. System according to one of claims 1 or 2, wherein:

the first camera (28) has a daylight filter (29), such that the spectral sensitivity of the first camera (28) comprises a wavelength range from about 800 nm to about 1800 nm; and or

the illumination device (26) emits infrared radiation and the wavelength of the emitted infrared radiation comprises a wavelength range from about 800 nm to about 1800 nm.

4. System according to one of claims 1 or 2, wherein the first camera (28) is arranged centrally above the area to be controlled (24), so that an optimal The axis of the first camera (28) is substantially perpendicular to the bottom (18) of the area to be controlled (24).

5. System according to one of claims 1 to 3, wherein the first camera (28) is arranged laterally to the area to be controlled (24), and wherein on a ceiling of the area to be controlled (24) one or more deflection mirrors are mounted so that the electromagnetic radiation reflected from the bottom (18) of the area to be controlled (24) is deflected in the direction of the first camera (28).

6. The system of claim 5, wherein the one or more deflection mirrors have a convex curvature.

7. System according to one of claims 1 to 6, wherein the bottom (18) of the area to be controlled (24) contains color pigments, so that the bottom (18) of the area to be controlled (24) in particular reflects the part of the electromagnetic radiation whose Wavelength of the spectral sensitivity of the first camera (28, 46) corresponds.

A system according to any one of claims 1 to 7, further comprising a second and a third camera (48, 50) arranged such that optical axes of the second and third cameras (48, 50) are at an angle with the ground (18) of the area to be controlled (24), which is not equal to 90 °.

An access control system (12) comprising a system for determining the number of persons as defined in the preceding claims, wherein the lighting device (26) is arranged to illuminate only a lower part of the area (24) to be inspected.

10. Access control system (12) according to claim 9, wherein

the area to be controlled (24) is bounded by a first barrier (20), a second barrier (22) and lateral boundary walls (14, 16), within the area to be controlled (24) is a personal identification device, and

the access control system (12) allows a person access to the restricted area only if the system for determining the number of persons has determined on the basis of the image information that a person is within the area to be inspected (24) and the identity of the person has determined that the person is authorized to enter the restricted area.

11. A method for determining the number of persons within an area to be controlled (24) comprising the steps of:

Illuminating the bottom (18) of the area to be controlled (24) with electromagnetic radiation whose wavelengths are at least partially outside the wavelength spectrum of visible light;

Acquiring image information with a first camera (28), wherein a plurality of pixels are each assigned a brightness value, and wherein the image information corresponds to a first image of the region (24) to be controlled;

Determining the objects in the first image of the area to be inspected (24), the objects corresponding to dark areas located in the first image of the area (24) to be inspected;

Determining the size of the objects, wherein the size of an object corresponds to the number of pixels whose brightness value is smaller than the brightness threshold and which are arranged side by side in the first image of the area to be controlled;

Comparing the size of an object to a first size threshold;

Outputting a first signal corresponding to the information "1 person" if the object is smaller than the first size threshold, and outputting a second signal corresponding to the information "> 1 person" if the object is larger than the first sizes Threshold value is.

12. The method of claim 11, wherein:

a person in the area to be inspected (24) is asked to place a piece of luggage at a certain position within the area to be inspected (24) if the object is greater than the first size threshold; and or

the number of pixels of the objects is additionally compared to a second size threshold when more than one object within the region to be controlled (24) is detected, and wherein the objects whose number of pixels is smaller than the second size threshold be classified as luggage; and or a signal is output that corresponds to the information "> 1 person" if at least two objects have a number of pixels which is greater than the second size threshold value.

13. The method of claim 11, wherein a position of the objects is monitored and the objects whose position changes, classified as moving objects, and the objects whose position does not change are classified as rigid objects; and outputting a signal corresponding to the information "> 1 person" when detecting two moving objects within the area to be checked (24).

14. The method according to any one of claims 11 to 13, wherein additionally with a second and a third camera (48, 50) image information is detected, which corresponds to a second and third lateral image of the persons and the objects within the area to be controlled (24) ,

The method of claim 14, wherein the image information captured by the second and third cameras (48, 50) is used to estimate the height of the persons and the objects within the area to be controlled (24);

and / or to detect an attempt by the person to skip a barrier (20, 22) of an access control system (12).