WO2018077766A1

WO2018077766A1 - Entrance system with image sensors

Info

Publication number: WO2018077766A1
Application number: PCT/EP2017/076934
Authority: WO
Inventors: Roger Dreyer
Original assignee: Assa Abloy Entrance Systems Ab
Priority date: 2016-10-24
Filing date: 2017-10-23
Publication date: 2018-05-03

Abstract

The present invention concerns an entrance system for controlling an entrance between a first and a second area. The entrance system comprises an automated door for opening or closing the entrance. Image sensors are configured for imaging an entrance area comprising a first zone located in the first area and a second zone located in the second area. A control system processes the images from the sensors and determines object data. The object data, related to a location or a variation of a location of an object in the entrance area, are compared with pre-defined criteria defining conditions for opening or closing the entrance. A resulting door control command signal is triggered for controlling the automated door.

Description

ENTRANCE SYSTEM WITH IMAGE SENSORS

Field of the invention

The present invention relates to an entrance system for controlling an entrance. More specifically it relates to an entrance system comprising an automated door for opening or closing the entrance.

Description of prior art

Various configurations of entrance systems using automated doors for controlling an entrance exist. The entrance systems control the movement of people and/or objects from one area to another area.

The entrance can for example be an entrance of a building that is controlled by using a revolving door. In other examples swing doors, sliding doors or overhead doors can be used to open and close an entrance between two areas.

Typically, multiple sensors are used to detect an occurrence of a person or an object in an entrance area adjacent to or surrounding the entrance. When one of the sensors detects an occurrence of a person or an object in the entrance area, a door control command signal is sent to a motor drive unit for activating the movement of the door so as to open the door. Generally, after a time-out period and when no further occurrence of a person or object is detected, the entrance is closed again.

Various technologies of sensors are used to detect the occurrence of a person or object in an entrance area. For example photovoltaic sensors, radar detectors, ultrasound detectors, load cells, thermal sensors, infrared sensors (IR) and image sensors have been used.

Generally, entrance systems use a combination of sensors of various technologies and these various sensors are installed for performing various functions . Some sensors, e.g. IR-sensors, are for example used as safety sensors to monitor potential risks of a person being hit by a moving door during the opening and closing operation. Other sensors, e.g. a load cell or mechanical switch can be used to activate the opening of a door when a person stands in front of the entrance.

Image sensors used for the surveillance of an entrance or for detecting the presence of an object at an entrance are typically using CCD technology (charge-coupled device) or CMOS technology.

One of the problems encountered with automated doors is that they do not operate in an efficient way. When opening and closing the door multiple times when it was not always needed to do so, the energy consumption for heating or cooling the building is strongly increasing. In some cases it might be more efficient to leave the door open for a longer period rather than sub-seguentially opening and closing it multiple times. In other examples the door can accidently be opened when it was not necessary, for example if a presence detector was detecting an animal in the entrance zone .

In patent document EP1681424, a control system for a sliding door is disclosed wherein the position and moving direction of an object is determined and a target opening degree for the sliding door is calculated. As shown on Fig. 4B of this document, a seguential algorithm is implemented wherein a first object is detected and a door setting for the sliding door is determined and in a subseguent step, when a second object is detected, a new door setting for the sliding door is determined for the second object and the setting for the second object is compared with the door setting of the first object. Depending on this comparison, either the setting of the second object overrides the first setting if the sliding door needs to be opened more or the second setting is ignored if the door is sufficiently open to have the second object pass the sliding door.

Hence, there exist a need in the art for a more intelligent entrance system for controlling the opening and closing of an entrance in a more efficient way. Summary of the invention

The present invention is defined in the appended independent claims. Preferred embodiments are defined in the dependent claims .

In particular, the present invention concerns an entrance system for controlling an entrance between a first area and a second area. This entrance system comprises an automated door configured for opening and closing the entrance and a control system configured for controlling the automated door for opening or closing the entrance.

The entrance system according to the invention comprises a series of image sensors S(i), with i=l to NS with NS≥2, configured for continuously or repetitively capturing images of an entrance area. The entrance area comprises a first entrance zone located within said first area and a second entrance zone located within the second area, and the series of image sensors S(i) are coupled with the control system for transmitting the images to the control system.

The entrance system according to the invention is characterized in that the control system comprises a processor programmed for repetitively performing steps of: a) acquiring, from each image sensor S(i) of the series of image sensors, an actual image _;(t), with t being a variable expressing a time of capturing the image, b) identifying one or more objects (B(j)) occurring on the actual images j(t) obtained from the sensors S(i), c) determining actual object data OD(t) associated with the one or more objects identified, wherein the actual object data OD(t) comprise at least position data defining a location and/or a variation of a location in the entrance area of the one or more objects identified, d) comparing the actual object data OD(t) associated with the one or more objects with pre-defined criteria, as follows :

• if said actual object data OD(t) are associated with a single object then comparing the actual object data associated with the single object with pre-defined criteria that are based on a presence of a single object in said entrance area and that are defining conditions for opening or closing said entrance, and

• if said actual object data OD(t) are associated with multiple objects (B(j)), then comparing in combination the actual object data associated with the multiple objects with pre-defined criteria that are based on a concurrent presence of multiple objects in said entrance area and that are defining conditions for opening and closing said entrance, and wherein the comparing of the object data associated to a single object or associated to multiple objects results in a selection of a door control command signal,

e) triggering of said door control command signal for

controlling the automated door for opening or closing the entrance. Advantageously, by configuring the image sensors such that the entrance area imaged by the sensors comprises a first entrance zone located in the first area and a second entrance zone located in the second area, the control of the automated door can be performed in a more proactive way taking into account persons arriving at the entrance area and persons leaving the entrance area from each side of the entrance. Indeed, by using criteria that are based on a concurrent presence of multiple objects in the entrance area, the conditions for opening or closing the entrance when multiple objects are present in the entrance area, are optimally defined taking simultaneously into account the position data of each of the objects identified.

Advantageously, by using a control system generating a door control command signal for opening or closing the automated door wherein the generation of the door control command signal is based on processing the images obtained from a plurality of sensors sensing the entire entrance area, the opening and closing can be performed in a more efficient way.

Indeed the door control command signal triggered is not depending on a single event of for example an object detected on one side of the entrance but the door control command signal triggered depends on an evaluation of all objects identified on both side of the entrance thereby taking into account the location and/or variation of location of each of the objects identified in the entrance area. For example, when a person is detected in the first area in proximity of the entrance, the door entrance system according to the invention can postpone the opening or slow down the opening of the door when a second person is observed in the second area also walking to the entrance but still being at a larger distance from the entrance. Advantageously, by using image sensors, a single sensor can cover a larger detection area and hence the number of sensors needed for the entrance system can be reduced, which reduces cost and complexity of the system. In addition, using image sensors and processing the images according to the invention, both safety related and functional related events are detected and associated door control command signals are triggered for controlling the door. Hence the number of additional specific safety related sensors can be reduced or eliminated. Advantageously by using image sensors, processing the images and triggering door control command signals according to the invention, the freguency of opening and closing the door can be reduced resulting in a reduction of power consumption related to heating or cooling an area in a building.

Typically, the data defining the location of an object B(j) comprise position coordinates x(j), y(j) and z(y) measured along the coordinate axes of a coordinates system having axes X,Y and Z, and wherein the X and Y axes are, for example, forming a plane parallel with a floor level of the entrance area.

Preferably, the actual object data OD(t) comprise an object type information for each object identified. This object type information is determined by comparing a shape of the object identified with pre-defined shapes corresponding to various types of objects.

In preferred embodiments, the actual object data OD(t) further comprise an object lateral size for each of the one or more objects identified, and wherein the object lateral size is defined as a maximum width of the object measured in a plane parallel with the X,Y plane.

Advantageously, by taking into account the speed of the object and/or the size of the object when defining the door control command signal for opening or closing the door, the door is only opened and closed just when the object is about to pass the entrance. In this way, the total time that the door is open, is reduced.

In embodiments according to the invention, the predefined criteria that are based on a presence of a single object in the entrance area or that are based on a concurrent presence of multiple objects in the entrance area comprise a plurality of pre-defined object occupancy scenarios. Each object occupancy scenario is defining at least a location and/or a variation of the location of one or more exemplary objects in the entrance area, and for each of the object occupancy scenarios, an associated door control command signal is pre-defined.

Short description of the drawings

These and further aspects of the invention will be explained in greater detail by way of example and with reference to the accompanying drawings in which:

Fig.l schematically illustrates a side view and a top view of an entrance area controlled by an entrance system according to the invention;

Fig.2 shows a schematic block diagram of the entrance

system according to the invention;

Fig.3 shows an entrance system according to the invention comprising a 3-wing revolving door;

Fig.4 shows an entrance system according to the invention comprising a 4-wing revolving door;

Fig.5 shows an entrance system according to the invention comprising a 2-wing revolving door;

Fig.6A shows an entrance system according to the invention comprising a single sliding door; Fig.6B shows an entrance system according to the invention comprising a double sliding door;

Fig.7A shows an entrance system according to the invention comprising a double swing door positioned such that the entrance is in an entrance closed status;

Fig.7B shows an entrance system according to the invention comprising a double sliding door wherein the doors are positioned such that the entrance is in a status in between open and closed;

Fig.7C shows an entrance system according to the invention comprising a double sliding door wherein the doors are positioned such that the entrance is in an entrance open status;

The figures are not drawn to scale. Generally, identical components are denoted by the same reference numerals in the figures .

Detailed description of preferred embodiments

An entrance system for controlling an entrance between a first area 3 and a second area 4 according to the invention is schematically illustrated in Fig. 1. Such an entrance system comprises an automated door 2 configured for opening and closing the entrance and a control system 25 configured for controlling the automated door 2 for opening or closing the entrance. Typically, the first area 3 and second area 4 are separated by a wall 17 and an opening in the wall is forming the entrance for passing between the first and second area, the entrance is for example the entrance of a building.

The entrance system further comprises a series of image sensors S(i), with i=l to NS with NS≥2, configured for capturing continuously or repetitively images of an entrance area wherein the entrance area comprises, as illustrated on Fig. 1, a first entrance zone 14 located within the first area 3 and a second entrance zone 15 located within the second area 4. In other words, with this configuration of the series of image sensors S(i), images are taken from both sides of the entrance. For example, the entrance system shown in Fig. 1 comprises two image sensors S(l) and S(2) wherein the first imaging sensor S(l) is imaging the first entrance zone 14 and the second imaging sensor S(2) is imaging the second entrance zone 15.

The image sensors according to the invention have to be construed as digital cameras for taking a time seguence or a stream of 2-D image frames. Typically, the image sensor is using CCD technology (charge-coupled device) or CMOS technology. The captured images by the image sensor are converted into electronic data files for further processing with a processor.

The series of image sensors S(i) are coupled with the control system 25 for transmitting the captured images _;(t) to the control system, with t being a variable expressing a time of capturing the image. As discussed above, the images are taken continuously or repetitively and hence the control system will acguire continuously or repetitively actual images, i.e. receiving continuously updated images. As illustrated in Fig. 2, the control system 25 receives as input, in this example, the actual images M₁(t) and ₂(t) from the sensors S(l) and S(2), respectively.

The control system 25 comprises a processor 28 programmed for repetitively acguiring from each image sensor S(i) of the series of image sensors, the actual image _;(t), with t being, as mentioned above, a variable expressing a time of capturing the image .

The processor 28 is programmed to identifying one or more objects (B(j)) occurring on the actual images ;(t) received from the sensors S(i) and to determining actual object data OD(t) associated with the one or more objects identified. These actual object data OD(t) comprise at least position data defining a location (P(j)) and/or a variation of a location (AP(j)/At) in the entrance area of the one or more objects (B(j)) identified. In other words, the actual object data OD(t) obtained at a time t can comprise for example position data of one object if only a single object is identified at time t and at another time t' the actual object data OD(t') can for example comprise position data of more than one object, depending on the number of objects identified at time t'.

A variation of a location of an object is determined by comparing or tracking the location of an object obtained from two or more subseguent images captured at different times. The variation of a location can for example be expressed as a speed of the object or a speed component measured along a given axis, for example the speed can be determined along an axis Y perpendicular to the door as shown on Fig. 1. In embodiments, the object data OD(t) comprise information indicating if the object identified is moving towards the entrance or if the object is moving away from the entrance .

More generally, the speed of each object B(i) can be expressed as a speed vector v[ in a three dimensional coordinate system having axes Χ,Υ,Ζ. The length of the vector v[ indicates the actual or the latest measured speed of the object. In the lower panel of Fig.l, the actual speed vector v[ is shown for each of the exemplary objects B(i) in the entrance area.

The processor 28 is further programmed for repetitively comparing the object data OD(t) with pre-defined criteria defining conditions for opening or closing the entrance. As a result of this comparing, a door control command signal 26 for opening or closing the entrance is selected. And in a final step this door control command signal 26 selected is triggered for controlling the automated door 2 for opening or closing the entrance.

If the actual object data OD(t) are associated with a single object then the pre-defined criteria are based on a presence of a single object in the entrance area. On the other hand, if the actual object data OD(t) are associated with multiple objects (B(j)), then the pre-defined criteria are based on a concurrent presence of multiple objects in the entrance area.

If the object data are associated to multiple objects, the comparing with the pre-defined criteria is performed in combination. Comparing in combination has to be construed as taking the object data associated to the multiple objects as a whole, i.e. as a combined set of data, for making the comparison with the pre-defined criteria that are based on a concurrent presence of multiple objects. It is this comparison in combination that results in a selection of a door control command signal. This is in contrast with for example document EP1681424, where a seguential algorithm is implemented wherein a first object detected is compared to criteria and results in a first door control command signal defining a position for a sliding door and sub-seguentially a second object detected is compared with criteria resulting in a second door control command signal, and wherein the second door control command signal is either overriding the first door control command signal or wherein a selection is made between the first and the second door control command signal .

Pre-defined criteria based on a concurrent presence of multiple objects have to be construed as criteria that are specifically defined for the specific situation where there is a presence of multiple objects in the entrance area. These criteria for a concurrent presence of multiple objects comprise criteria related to position data of multiple objects. As a result, the door control command signal resulting from the comparison of the pre-defined criteria with the object data OD(t) will depend on the specific situation. For example, for a situation where there is a single object detected, the pre-defined criteria defining conditions for opening or closing the entrance are for example related to values for the position and speed of the single object detected. If there is however a situation where two objects are detected, the pre-defined criteria for a two object situation are not necessary the same criteria as for the single object situation. Indeed, it is not because the position and speed criteria are met to open the entrance for a single object situation that for a two object situation with one of the objects having the same position and speed as in the one object situation, that the criteria are necessarily met to open the entrance. Therefore, in this example with two objects, the criteria are defined for a concurrent presence of two-objects in the entrance area and are related in this example on the position and speed of both objects in the entrance area. Further examples of criteria involving multiple objects in the entrance area will be given below when discussing detailed embodiments of the invention.

By performing this repetitive process of acguiring actual images, identifying objects, determining actual object data, comparing the object data with predefined criteria and selecting and triggering a door control command signal, the automated door is continuously controlled based on an continuous evaluation of the presence and location of objects in the first 14 and second entrance zone 15. This repetitive process can for example be performed every millisecond .

In Fig. 1, an exemplary coordinate system is shown comprising coordinate axes X and Y forming a plane parallel with a floor level of the entrance area. An axis Z, is perpendicular to axes X and Y. The locations of objects in the entrance area can then be expressed as two-dimensional coordinates x,y or as three-dimensional coordinates x,y,z.

In some embodiments according to the invention, the control system 25 is configured for receiving a door status signal 27 from the automated door 2 or from a door status sensor. This status signal is indicating an entrance status such as for example an entrance closed status, an entrance open status or a transitional status between open and closed. The pre-defined criteria are then defined for various of the entrance statuses. In alternative embodiments, the door status is deduced from the actual images _;(t) taken with the series of sensors S(i) . Examples of door statuses and associated pre-defined conditions will be given below.

Typically, an automated door comprises a motor drive unit coupled with the control system and wherein the motor drive unit is configured for driving a motor for opening or closing the automated door.

The number of image sensors needed to image the entrance area strongly depends on the type of door that is used to open and close the entrance. A minimum of two sensors is at least needed to image both sides of the entrance, i.e. to be able to detect objects located in the first 14 and second 15 entrance zone. In Figures 3 to 7, a number of different type of doors are shown having various numbers of image sensors S ( i ) . In figures 3 to 5, entrance systems for controlling an entrance between a first area 3 (e.g. an area inside a building) and a second area 4 (e.g. an area outside a building) comprising different configurations of revolving doors are shown. The reference number 17 on the figures indicate a separation wall separating the first 3 and second 4 area. These entrance systems with revolving doors comprises a number of image sensors configured to image an entrance area wherein the entrance area comprise a first zone 14 in the first area 3 and a second zone 15 in the second area 4. In this way, persons walking from the inside of a building towards the revolving doors or persons walking from outside to the revolving door will both be detected by the image sensors .

In Fig. 3A and Fig.3B, a revolving door is shown having three rotatable compartments and an image sensor is placed in each of the compartments. As shown on Fig.3A, the sensors are configured to image the first 14 and second 15 entrance area such that objects arriving from either side of the entrance can be detected. In this example, as the sensors are rotating with the rotation of the door, the field of view of each sensor is varying with the angular position of the door. Hence, the first 14 and second 15 entrance area are imaged with various sensors depending on the rotational position of the revolving door. In Fig. 3B, for a given angular position of the revolving door, the field of views 7,8,9 of each of the three sensors are schematically represented using a different fill pattern on the drawing for each of the field of views . The combined field of views of each of the sensors are covering the entrance area with the first and second zone to be detected for any rotational position of the revolving door. In Fig. 4, an example of a revolving door comprising four compartments is shown wherein each compartment comprises an image sensor. The field of views 7,8,9,10 of each of the sensors, for the given rotational position shown, is illustrated on the drawing with a different fill pattern used for each field of view.

In alternative embodiments of revolving doors, additional image sensors attached to a non-rotating part of the door can be added to increase the size of the entrance area that can be detected with the image sensors.

A revolving door comprising two compartments is shown on Fig. 5 wherein each compartment comprises one image sensor S(l) and S(2) rotating with the rotation of the door and wherein two additional image sensors, S(3) and S(4), are attached to a non-rotating part of the door or to the ceiling in the entrance area. These additional sensors S(3) and S(4) are optimized for detecting objects on the pathway towards the entrance. In this way, for any angular rotation of the revolving door, the four image sensors are configured to image the entrance area comprising a first 14 and second 15 entrance area.

In Fig.6A an example of an entrance system between a first area 3 and a second area 4 comprising a single sliding door is shown. The first area is for example an area inside a building and the second area is an area outside the building. This entrance system comprises a fist sensor S(l) configured for imaging a first entrance zone 14 located in the first area 8 and a second sensor S(2) configured for imaging the second entrance zone 15 located in the second area 4. In this example, the second sensor S(2) is positioned on a side wall of the second entrance zone 15 such that it can detect a potential dangerous situation where a person could stand in the pathway of the sliding door. This dangerous area 18 is schematically indicated on Fig. 6A with a dotted line.

In Fig. 6B, an example is shown of an entrance system comprising two sliding doors wherein a first sensor S(l) images the first entrance zone 14 located in the first area 3, for example inside a building. A second S(2) and a third sensor S(3) are placed in the second entrance zone 15 located in the second area 4, for example at the exterior of the building .

As a further example of an entrance system, a system comprising a swing door is shown on Figures 7A, 7B and 7C . To each leaf of the door, two sensors are attached. These four sensors allow to image the entrance area for any position of the swing door as illustrated on the Figures . When in closed position as illustrated on Fig. 7A, the sensors S(l) and S(2) are imaging the first entrance zone 14 located on the first area 3 and sensors S(3) and S(4) are imaging the second entrance zone 15 located on the second area 4. As the sensors are mounted on the moving swing doors, the zone of the entrance imaged by a sensor can vary depending on the position of the door but the combined field of view of the sensors will result in imaging the entire entrance area with the first and second zone for any position of the door as illustrated in Figures 7A, 7B and 7C .

In embodiments according to the invention, the processor 28 that is processing the seguence of images _;(t) is using computer vision technology for detecting the occurrence of one or more objects on an image and determining the location of the detected object. Further, by tracking the objects in a seguence of images, the variation of the location of the objects can be observed and hence a trajectory of the objects can be determined. Such type of machine vision algorithms have been developed for example in the field of robotics.

With computer vision technology, different types of objects can be distinguished from each other, for example humans can be distinguished from a bicycle or a dog or any other object. These known algorithms for detecting objects on 2D images can be based on known technigues such as for example object shape recognition, object edge detecting approaches or blob detection. Typically, object recognition algorithms need to be trained using exemplary digital images in which objects have been outlined and labelled by hand. For the purpose of the current invention, the algorithm is optimized to recognize humans in order to be able to take the necessary actions for opening or closing the entrance in response to humans arriving in the entrance area.

In order to determine the location of the object in the entrance area imaged by the sensors S(i), the known computer stereo vision method can for example be applied. In this method, two images are taken from a different camera position and the 3D information can be extracted based on the relative positions of the object seen on the two images. As a result, three dimensional coordinates of the object with respect to a coordinate system associated to the entrance area are obtained. In embodiments, the image sensor is a dual image sensor having two lenses configured for taking stereo images.

In other embodiments, two single image sensors separated far apart from each other and configured for imaging an entrance side can be used in a wide stereo vision mode where a control system can combine the information taken from the two single image sensors to deduce 3D object information .

The processor is repetitively processing the actual images _;(t) acguired for each sensor S(i) and placing the data resulting from the computer vision analysis in a data file OD (t) .

In some embodiments, the object data OD(t) further comprise an object lateral size for each of the one or more objects (B(j)) identified from the images. This object lateral size is for example defined as a maximum width of the object measured in a plane parallel with the X,Y plane. Such a lateral size can for example be used to select a specific door control command for opening the door only partly, i.e. sufficiently to allow the object of a given size to pass the entrance.

As the image sensors are configured to image the entrance area on both sides of the entrance, i.e. covering the first 14 and second entrance zone 15, the object data are simultaneously acquired for all objects in the entrance area allowing to take a central decision based on the overall obtained information from objects occurring in the entrance area at a given moment in time. This process of acquiring actual images, processing the actual images for obtaining actual object data from each sensor is performed repetitively such that the object data OD(t) are repetitively updated and hence a function of the time. In this way, the control system can trigger immediately the necessary control commands for the motor driver unit when the object data correspond to pre-defined criteria requiring an action from the motor drive unit of the automated door. The object data OD(t) are for example updated every millisecond but the repetition of performing the processing can also be faster or slower.

The detailed implementation of the comparison of the object data OD(t) with pre-defined criteria defining conditions for opening or closing the entrance depends on the type of automated door. A number of examples are discussed here below.

The pre-defined criteria defining conditions for opening or closing the entrance have to be construed as criteria comprising a number of pre-defined scenarios or use cases. These scenarios allow to define criteria not only for the situation of a presence of a single object in the entrance area but also for situations of concurrent presence of multiple objects in the entrance area. In a first example, a single sliding door of the type as shown on Fig.l having a first sensor S(l) for detecting objects in the first detection zone 14 and a second sensor S(2) for detecting objects in the second detection zone 15 is considered. A number of exemplary scenarios are given in the second column of table 1 and the associated door control command signal is given in the third column. In the fourth column, some examples of object data OD(t) associated to a single object or associated to multiple objects are given. Remark that if the object data are associated to multiple objects, the object data of the multiple objects are compared in combination with the pre-defined criteria defined by the scenario given in column 2. This means that all the object data of all the objects detected at time t are compared together as a full set of data with the criteria defined in the scenario.

In the first pre-defined scenario, NO 1, the sliding door is in a closed status and one or more objects are located in the first 14 and/or second 15 entrance zone, respectively. In this scenario NO 1, these objects in the entrance zones have a speed egual or higher than lm/s along the Y axis and the objects are moving towards the entrance and the objects are located at a distance from the entrance, measured along the Y axis, between 0 m and 5 m. And, as a further pre-defined condition, no further object, other than those moving towards the entrance, is located in the entrance zones. When, at a given moment in time t, the object data OD(t) determined from the actual images fulfill these criteria of this scenario NO 1, then the associated door control command signal, as given in column 3, will be selected and a door control command signal will be triggered wherein this door control command signal corresponds to opening the door at high speed.

In the pre-defined scenario NO 2, a door control command signal for opening the door at slow speed is triggered when the object is moving towards the entrance at a slow speed below a pre-defined threshold (in this example set to 1 m/s ) while the object is at a distance from the entrance between 2m and 5m, as measured along the axis Y.

The scenario NO 3, is an example where the door is kept closed when objects are observed at slow speed and still being above a given pre-defined distance from the entrance.

The scenarios N04 and N05 are examples where the status of the door is an open status and where the scenarios define conditions to open the entrance. Scenario NO 4 is an example where the resulting door control command signal is to close the door at high speed, while scenario NO 5 is an example where the resulting door control command signal is to close the door at low speed. In scenario NO 5, one or more persons are walking away from the entrance and a further person is walking towards the entrance. In this scenario, if the further person is for example already close to the door and walking fast towards the door and hence, in such a situation, the door can already start closing slowly while still allowing this further person to cross the entrance. Scenario NO 5 is an example of pre-defined criteria that are only applicable for a concurrent presence of objects in the entrance area.

The more scenarios that are pre-defined, the more intelligent the automated door will be to recognize concrete entrance occupancy situations that can occur. In this example, four scenarios are defined but in practice 10 to hundreds of scenarios will be pre-defined such that the majority of the scenarios that are likely to occur in daily situations are covered. If no matching scenario is found, the door can either stay open or either stay closed. In the examples given, it is assumed that the object recognition of an object corresponds to the recognition of a human person. When for example a bird is detected, this will have no effect on the opening or closing of the automated door.

In the fourth column of table 1, one or more examples are given for each scenario of matching object data OD(t), i.e. actual object data determined from the actual images that are interpreted by the processor 28 as matching to one of the pre-defined scenarios and for which, as a result of the matching criteria, a door control command signal will be triggered. In this example, the object data of each sensor comprise four parameters, the first parameter is an ID given to the object identified (for example alphabetical letters are used in this example), the second parameter is the x coordinate, the third parameter is the y coordinate and the fourth parameter is the speed of the object in the y direction. The coordinate system used in this example corresponds to the coordinate system indicated on Fig. 1, i.e. a positive y value indicates that the object is located in the second entrance zone 15 and a negative y value indicates that the object is located in the first entrance zone 14. In these examples, the convention is that a positive speed indicates that the object is moving towards the entrance and a negative speed indicates that the object is moving away from the entrance. If no data are present in the object data file OD(t), this indicates that no object is identified on the images. When an object has left the first or second entrance zone, the object will disappear from the actual object data file.

NO Scenario Door OD(t) =

control (ID,x,y,dy/dt) command

signal

1 -Door status closed -open the Example 1

-In the first 14 entrance door at A, 0.2, 3.3, 1.2 zone or the second 15 high speed Example 2 entrance zone, one or A, 0.2, 3.3, 1.2 more objects are B, 0.8, -4.2, 1.3 identified having a speed 0,-1.1,-2.5,1.1 egual or higher than lm/ s

along the Y axis and

moving towards the

entrance and being at a

distance between 0 m and

5 m from the entrance,

and no further objects

detected .

2 -Door status closed -open the Example 1

-In the first 14 entrance door at A, 0.7, 2.5, 0.7 zone or the second 15 slow speed Example 2 entrance zone, one or A, 0.2, 3.3, 0.9 more objects are B, 0.8, -4.2, 0.8 identified having a speed 0,-1.1,-2.5,0.7 lower than lm/ s along the

Y axis moving towards the entrance and being at a

distance between 2m and

5m from the entrance, and

no further other objects

detected .

-Door status closed -keep the Example 1 -In the first 14 door A, 0.7, 2.5, 0.1 entrance zone or the closed Example 2 second 15 entrance zone, A, 0.7, 2.5, 0, 1 one or more objects are B, -1.0, -3.0, 0.0 identified having a

speed lower than 0.2 m/s

along the Y axis and

moving towards the

entrance and being at a

distance between 2 m and

5 m from the entrance,

and no other object

detected .

-Door status open -close the A, 0.7, 1.5, -0.8 - In the first 14 door at B, 1.5, 2.3, -1.2 entrance zone or the high speed C, -2,-2.1,-0, 9 second 15 entrance zone, D, 1,-1.2, -1,3 one or more objects are E, -1,-3, -1,5 identified moving away

from the entrance and

all being at a distance

larger than 1 m from the

entrance, and no further

objects detected.

-Door status open -close the A, 0.7, 1.5, -0.8 -In the first 14 door at B, 1.5, 2.3, -1.2 entrance zone or the low speed C, 1,1.3, 1.1 second 15 entrance zone, D, -2, -2.1,-0, 9 one or more objects are E, 1,-1.2, -1,3 identified moving away F, -1,-3, -1,5 from the entrance and

being at a distance

larger than 1 m from the

entrance, and one

further object is

detected in the first or

second entrance zone at

a distance between 0 m

and 2m from the entrance

having a speed egual or

larger than 1 m/ s and

moving towards the

entrance .

Table 1 examples of scenarios defining pre-defined conditions for triggering a control command for a single sliding door .

In some embodiments of the invention, specific safety zones in the entrance area, either in the first zone 14 or in the second zone 15, can be identified as being potential dangerous zones. For example, when a sliding door of the type shown on Fig. 6A and Fig. 6B is used, a person standing in the pathway of the door could be hit by a side of the door when the door is suddenly opening. Such a potential dangerous area 18 is schematically encircled with a dotted line on Fig. 6A and Fig. 6B. Another example of a potential dangerous zone is the zone covering the opening side of a leaf of a swing door, as shown in Fig. 7A, where the dangerous zone 19 is schematically encircled with a dotted line. A person standing in this zone could be hit by the leaf of the swing door if the door is suddenly opening due to another person arriving from the other side of the entrance triggering the opening of the door. To avoid these type of dangerous situations mentioned above, pre-defined criteria according to the invention comprise scenarios that cover these potential dangerous situations and the appropriate control commands associated with these scenarios are pre-defined.

In the table 2, here below, a few exemplary scenarios for a swing door defined to cope with the potential dangerous situations as discussed above, are provided. The dangerous zone 19 mentioned above is identified in this scenario as the swing door safety zone. In scenario NO 1 of table 2, the door is kept closed if one or more objects are detected in the second entrance zone, more specifically in the swing door safety zone 19. This scenario is independent of any other objects detected in the first 14 entrance zone. In scenario NO 2, again objects are detected in the swing door safety zone, but in this scenario these objects are identified as moving away from the entrance. In this second scenario when, at the same time, one or more objects are detected in the first zone 14 moving towards the entrance, a door control command signal is triggered to open the door at slow speed. In this way, both the persons leaving the safety zone 19 and the persons arriving at the entrance from the first entrance zone 14 will recognize that the intention of the door control system is to open the door.

NO Scenario Door control command signal

1 -Door status closed -keep the door

-One or more objects detected in closed

the second entrance zone 15 and being located in the swing door

safety zone 19.

2 -Door status closed -open the door

-One or more objects detected in at slow speed the second entrance zone 15 in the

swing safety zone 19 and each of

these objects is moving away from

the entrance, and one or more

objects detected in the first

entrance zone 14 moving towards the

entrance .

Table 2 examples of scenarios defining pre-defined conditions for triggering a control command for a swing door .

In table 3, some exemplary examples of scenarios are defined for a sliding door addressing situations where objects are located in or nearby the dangerous zone 18 discussed above. In scenario NO 1 of table 3, the automated door is in a closed status and an object is detected in the sliding door safety zone 18. When actual object data OD(t) are matching this scenario NOl, a door control command signal to keep the door closed is triggered. In the second scenario N02 of table 3, the door status corresponds to a moving door at a high speed from a closed to an open status. In this scenario in this door status, and one or more objects are moving towards the sliding door safety zone 18 and those objects are located at less than 0.5 m from this safety zone 18. When actual object data OD(t) would correspond to this scenario, then the processor of the control system will trigger a door control command signal corresponding to reduce the door speed from a high speed to a low speed. NO Scenario Door control command signal

1 -Door status : closed -keep the door -One or more objects detected in closed

the second entrance zone 15 and

being located in the sliding door

safety zone 19.

2 -Door status : moving at high speed -reduce the door from closed to open status speed to slow -One or more objects detected in

the second entrance zone 15 moving

towards the sliding door safety

zone 18 and located at less than

0.5 m from the safety zone 18.

Table 3 examples of scenarios defining pre-defined

conditions for triggering a control command for a sliding door .

As illustrated above by defining numerous specific scenarios for a specific door, an intelligent door entrance system is obtained. Some scenarios are addressing safety aspects while others can address factors that help to reduce the energy consumption for heating or cooling the building by reducing the number of opening and closing actions . Other scenarios can be dedicated to facilitate the people crossing the entrance in an efficient way. For each automated door, the specific set of scenarios is defining the pre-defined conditions for opening or closing the entrance with the automated door. As illustrated with the exemplary scenarios, the door control command signal triggered by the processor is a result of an overall evaluation of the occupancy situation in the entrance area of the door entrance system.

The present invention has been described in terms of specific embodiments, which are illustrative of the invention and not to be construed as limiting. It will be appreciated by persons skilled in the art that the present invention is not limited by what has been particularly shown and/or described above and that alternatives or modified embodiments could be developed in the light of the overall teaching of this disclosure.

Claims

1. An entrance system for controlling an entrance between a first area (3) and a second area (4), comprising

· an automated door (2) configured for opening and

closing said entrance,

• a control system (25) configured for controlling the automated door (2) for opening or closing the

entrance,

· a series of image sensors S(i), with i=l to NS with

NS≥2, configured for continuously or repetitively capturing images of an entrance area, wherein said entrance area comprises a first entrance zone (14) located within said first area (3) and a second entrance zone (15) located within said second area

(4), and wherein said series of image sensors S(i) are coupled with said control system (25) for

transmitting said images to the control system (25), characterized in that said control system (25) comprises a processor (28) programmed for repetitively performing steps of :

a) acguiring, from each image sensor S(i) of said series of image sensors, an actual image j(t), with t being a variable expressing a time of capturing the image, b) identifying one or more objects (B(j)) occurring on the actual images j(t) obtained from said sensors S(i), c) determining actual object data OD(t) associated with the one or more objects (B(j)) identified, wherein the actual object data OD(t) comprise at least position data defining a location (P(j)) and/or a variation of a location (AP(j)/At) in said entrance area of said one or more objects (B ( j )) identified, comparing the actual object data OD(t) associated with the one or more objects (B(j)) with pre-defined

criteria, as follows :

• if said actual object data OD(t) are associated

with a single object then comparing the actual object data associated with the single object with pre-defined criteria that are based on a presence of a single object in said entrance area and that are defining conditions for opening or closing said entrance, and

• if said actual object data OD(t) are associated

with multiple objects (B(j)), then comparing in combination the actual object data associated with the multiple objects with pre-defined criteria that are based on a concurrent presence of multiple objects in said entrance area and that are defining conditions for opening and closing said entrance, and wherein said comparing of the actual object data OD(t) associated to a single object or associated to multiple objects results in a selection of a door control command signal (26),

triggering of said door control command signal (26) for controlling the automated door (2) for opening or closing the entrance.

An entrance system according to claim 1 wherein said actual object data OD(t) further comprise an object type information for each object (B(j)) identified and wherein the object type information is determined by comparing a shape of the object identified with pre-defined shapes corresponding to various types of objects. An entrance system according to any of previous claims wherein said position data defining a location (P(j)) of the one or more objects B(j) comprise, for each object B(j), position coordinates x(j) and y(j) measured along coordinate axes X and Y, and wherein said coordinate axes X and Y are forming a plane parallel with a floor level of the entrance area.

An entrance system according to claim 3 wherein said position data defining a location (P(j)) of the one or more objects B(j) further comprises for each object (B(j)) a position coordinate z(j) measured along an axis Z, perpendicular to axes X and Y.

An entrance system according to any of previous claims wherein said position data defining a variation of a location (AP(j)/At) of the one or more objects (B(j)) identified, comprise for each object B(j) a speed information and/or an indication of whether the object is moving towards the entrance or the object is moving away from the entrance.

An entrance system according to any of claims 3 to 5 wherein said actual object data OD(t) further comprise an object lateral size for each of said one or more objects (B(j)) identified, and wherein the object lateral size is defined as a maximum width of the object measured in a plane parallel with said X,Y plane.

An entrance system according to any or previous claims wherein said automated door (2) comprises a revolving door assembly having two or more rotatable compartments.

8. An entrance system according to claim 7 wherein said series of image sensors S(i) or a subset of said series of images sensors S(i) are placed within said one or more rotatable compartments.

9. An entrance system according to any of claims 1 to 6 wherein said automated door (2) comprises a single or a double sliding door.

An entrance system according to claim 9 wherein said series of image sensors S(i) or a subset of said series of images sensors S(i) are attached to the single or double sliding door.

An entrance system according to any of claims 1 to 6 wherein said automated door (2) comprises a swing door having single or double leaves and wherein said series of image sensors S(i) or a subset of said series of images sensors S(i) are attached to said single or double leaves.

An entrance system according to any of previous claims wherein the pre-defined criteria that are based on a presence of a single object in the entrance area or that are based on a concurrent presence of multiple objects in the entrance area comprise a plurality of pre-defined object occupancy scenarios, each object occupancy scenario defining at least a location and/or a variation of the location of one or more exemplary objects in said entrance area, and wherein for each of the object occupancy scenarios, an associated door control command signal (26) is pre-defined.

13. An entrance system according to any of previous claims wherein said control system (25) is further configured for receiving a door status signal (27) from said automated door (2) or from a door status sensor, and wherein said status signal is indicating an entrance status such as for example an entrance closed status, an entrance open status or a transitional status between open and closed, and wherein said pre-defined criteria are defined for various of said entrance statuses .

An entrance system according to any of previous claims wherein said door control command signal (26) corresponds to a door operational instruction selected from the following non-limiting list of door operational instructions : open the door at nominal speed, close the door at nominal speed, open the door at reduced speed, open the door at high speed, close the door at reduced speed, close the door at high speed, keep the entrance closed or keep the entrance open.

An entrance system according to any of previous claims wherein said automated door comprises a motor drive unit configured for driving a motor for opening or closing said automated door (2) and wherein said motor drive unit is coupled with said control system (25) .