DE202017106794U1 - System for object detection and tracking on areas, in particular parking lots - Google Patents

Abstract

An object detection system for detecting and tracking objects, comprising: - at least one data acquisition unit for acquiring object information; A memory unit for storing the object information output by the at least one data acquisition unit, wherein the storage unit stores an object information record for each of the at least one data acquisition unit, each object information record storing a plurality of object information of the respective data acquisition unit at different times from the data acquisition unit has been issued; and a data processing unit configured to generate objects based on a temporal course of the object information of the at least one object information data record.

Description

BACKGROUND

The present invention relates to an object detection system which can be used to detect and track objects such as trucks in a large area such as a parking lot.

In conventional systems for monitoring large areas, on which movable objects, such as trucks, cars, containers, or the like. To temporarily stop, in order to determine whether there is still sufficient space available for further objects on the surface, in general the entrances and exits are monitored, as will be described below by way of example with reference to a truck parking lot.

In the case of a truck parking lot on a highway, for example, the number of arriving trucks and their length was monitored at the entrance. In addition, the number of departing trucks was monitored at the exit. By adding or subtracting the arriving or departing truck, the free capacity of the place could be determined.

However, such a system has to be calibrated at certain intervals in order to ensure a reliable detection of the free capacities over a longer period of time. For this purpose camera images of the entire parking lot are evaluated manually to calibrate the current number of parking vehicles. In addition to additional hardware usage, this requires significant personnel deployment, which causes additional costs.

Furthermore, it is not possible to monitor how the objects on the large area move and are parked. It is thus not recognizable, for example, whether an object, for example, blocks an entry zone to parking spaces, is parked over several parking spaces or rows or changes its position. A reliable estimate of the free space is therefore not possible.

On the other hand, if only conventional laser scanners were to be provided for each park row, they would only permit the detection of a last truck closest to the laser scanner in a park row. A reliable statement about the assignment of the rows or a statement about objects that move between rows can not be made.

SUMMARY OF THE INVENTION

It may thus be desirable to provide a system, in particular for monitoring parking spaces and other large areas, which enables reliable detection of arbitrarily positioned objects with low hardware and personnel deployment.

This object is achieved with an object detection system according to claim 1 and with a large area monitoring system according to claim 7. Advantageous developments are the subject of the dependent claims.

An object detection and tracking object detection system according to an embodiment comprises at least one data acquisition unit for acquiring object information, a storage unit for storing the object information output from the at least one data acquisition unit, the storage unit storing an object information record for each of the at least one data acquisition unit, wherein each object information record stores a plurality of object information of the respective data acquisition unit which has been output from the data acquisition unit at different times; and a data processing unit configured to generate objects based on a temporal course of the object information of the at least one object information data record.

Accordingly, such an object detection system makes it possible to store object information such as measurement points of a laser scanner in the storage unit for each data acquisition unit and at different acquisition times. From the time course. For example, from the movement of the measuring points and the movement of the measuring points to each other, it can be determined which measuring points belong to an object. That The data processing unit can generate objects from the course of the object information, for example, summarize object information or measurement points to objects. For example, a unique identifier can be assigned to the generated object generated, for example.

This allows the acquisition and tracking of the objects both in the detection area of a data acquisition unit, as well as across multiple data acquisition units.

It is also possible to detect and distinguish several objects lying behind one another already by a detection unit (for example several objects in a row), since the measurement points are determined by the method described above Objects can be assigned and thus the objects can be differentiated.

Accordingly, each of the at least one data acquisition units can be set up to acquire a plurality of objects and / or the data processing device can be set up to generate a plurality of objects based on a temporal course of the object information of an object information data record of a data acquisition unit.

Each of the data acquisition units can also be designed as a laser scanner, in particular as a 2D laser scanner.

In this case, the object information of the at least one data acquisition unit in each case comprises a multiplicity of data points.

The data processing device can then be set up to combine data points of the object information of the at least one data acquisition unit into an object in order to generate objects. This can be done based on a time course of the data points.

Accordingly, laser scanners can be used to record objects even at night or in bad weather. In the case of training as a 2D laser scanner, the detection of objects at a distance from the laser scanner along a detection line can take place here. This results in distance measuring points along a line for each laser scanner. These measuring points can then be evaluated by the data processing unit.

In this case, the time profile may include either a relative movement of at least two data points determined based on object information acquired at different times from the at least one data acquisition unit, and / or a time sequence of at least two data points of at least two data acquisition units.

The object detection system may further comprise at least two data acquisition units, each of which is designed as a laser scanner, in particular as a 2D laser scanner. The object information of the at least two data acquisition units in this case in each case comprises a multiplicity of data points. The data acquisition area of the at least two data acquisition units may be arranged adjacent to each other. The data processing device can be set up, based on the time profile of the data points of the data information of the at least two data acquisition units, to combine the data points into objects.

A summary of data points to objects or a generation of objects can thus be based on the data points of multiple data acquisition units. It is thus possible to move objects not only selectively in the data acquisition area of a data acquisition unit, but also via a plurality of data acquisition units and also to track over a plurality of data acquisition units.

Furthermore, the data processing device can be set up, based on the temporal course of the object information of the at least one object information data set, the extent of the at least one object; the length of the at least one object; the width of the at least one object; and / or to determine the position of the at least one object.

By the time course of the measuring points, the distance of measuring points with a relative speed of zero, and / or the distance between the data acquisition ranges, as well as other parameters, additional properties of the objects can thus be determined.

Furthermore, the data processing device can be set up to track the objects based on the temporal course of the object information of the at least one object information data record.

The generated objects or the measurement points combined into an object (to which a clear identifier has possibly been assigned) can thus be tracked in their movement over the total area covered by the data acquisition units.

A large area monitoring system for monitoring a large area may include an object detection system. The plurality of data acquisition units of the object detection system may be arranged such that data acquisition areas of the plurality of data acquisition units are arranged adjacent to each other. The data processing device of the object detection system may be configured to determine free areas between the objects based on the time history of the object information of a plurality of object information data sets of the plurality of data acquisition units.

Thus, it can be determined whether, for example, still parking spaces are available in a parking lot. The object detection system may include a transmitting unit for transmitting information regarding the free Include areas and the captured objects to at least one mobile unit.

Thus, the information can, for example, to drivers of trucks or cars on mobile phones, control systems, navigation devices o.Ä. to get redirected.

The mobile unit may include a reception unit and a data display unit and configured to receive information from the transmission unit via the reception unit and to output information to a user via the data display unit based on the received information about the free areas and the detected objects.

Thus, the user can on the free areas or in detail on to be driven parking lots, to be traveled park rows o.Ä. being taught.

The large area can be here, for example, a parking lot, a container handling center or a port.

Accordingly, the objects may be, for example, trucks, cars, buses, containers and / or ships.

BRIEF DESCRIPTION OF THE DRAWINGS

In the following, further exemplary embodiments of the present invention will be described with reference to the figures.

1 shows a schematic representation of an object detection system according to one aspect;

2 shows a large area with object detection system.

DETAILED DESCRIPTION OF EMBODIMENTS

The illustrations in the figures are schematic and not to scale.

In the following description of embodiments in different figures, the same reference numerals are used, they designate the same or similar elements. However, identical or similar elements can also be designated by different reference symbols.

The 1 shows an object detection system 100 according to an embodiment of the present invention.

The object detection system 100 here has one or more data acquisition units 101 , a storage unit 102 and a data processing unit 103 on.

The elements of the object detection system 100 will be described in detail below.

The one data acquisition unit 101 or the multiple data acquisition units 101 capture in a respective data collection area of the data acquisition units 101 located objects and output corresponding detection data, referred to here as object information. The data acquisition units 101 For example, it may be a laser scanner, preferably a 2D laser scanner, which is arranged, for example, elevated (on a mast or the like) in order to enable detection over a distance even across objects.

The object information may be, for example, data points or measuring points of the laser scanner.

The following is the formation of the data acquisition unit (s) 101 be described in more detail as a 2D laser scanner.

A 2D laser scanner, also known as LiDAR, emits laser pulses and uses the reflections to calculate the distance to obstacles. These laser pulses are output in a plane with a predetermined angular resolution over the opening angle of the LiDAR. For example. a scan is made over an aperture angle of 190 ° so that as the data output of the 2D laser scanner, i. Object information, data for the removal of measuring points (data points) in a plane (hereinafter referred to as data acquisition plane) over an angle of 190 ° (depending on the laser scanner used) are available.

The detection plane is preferably formed substantially at right angles to the surface to be monitored or large area.

In other words, the detection levels of the data acquisition unit 101 arranged substantially parallel to each other.

This results in an arrangement of the acquisition levels of the data acquisition units (scanners) 101 as in 2 shown as dotted lines (scan lines), wherein the detection planes are arranged substantially perpendicular to the surface to be monitored (here a parking lot) and therefore one-dimensional viewed from above.

Accordingly, the data acquisition unit 101 be arranged such that they along the removal of measuring points or objects a detection line detected, preferably by sampling or distance measurement using different scanning angle. The resulting measurement information is output as object information.

The object information of the one or more data acquisition units is in turn stored in the memory unit 102 saved.

The storage unit 102 stores for each of the at least one data acquisition units 101 one object information record each. Such an object information record in turn has a plurality of object information of the respective data acquisition unit 101 on, which at different times from the data acquisition unit 101 recorded and saved.

In other words, such a data record contains not only current measurement data of the respective data acquisition unit 101 but also a history of measurement data or object information. The data record thus has measurement data from different measurement times from the past.

Storing the object information of the respective data acquisition unit 101 can through the storage unit 102 itself, or for example by another unit such as the data processing unit 103 respectively.

The in the storage unit 102 stored object information or the object information stored in the object information data sets different measuring times is determined by the data processing unit 103 processed.

For this purpose, the data processing unit reads 103 Object information from at least one object information record from the storage unit 102 out.

Using the object information of various measurement times, the data processing unit analyzes 103 the course of the object information and generates objects based on the time course of the object information.

In other words, measurement points or data points of a data acquisition unit 101 or more data collection units 101 summarized as a function of the time course of the measuring points to an object.

The data processing unit 103 thus generates objects by combining several measurement points or data points of one or more data acquisition units into an object.

The detected or generated objects can then possibly in combination with their speed and / or movement direction and / or dimension and / or one by the data processing unit 103 assigned object identifier in the storage unit 102 be stored or saved.

A summary of the measuring points or data points can be carried out here preferably via the evaluation of a relative movement of two measuring points. If two measuring points have a similar or identical speed or a relative speed of zero or near zero over a specific time (at least two measuring times), then it can be assumed that the measuring points or data points belong to an object. Accordingly, an object can be generated which comprises the data points.

Such an evaluation or summary of objects can preferably be successively improved over a longer evaluation period, for example adaptively adapted. Accordingly, it can be checked for the generated objects in subsequent evaluation steps whether the movement of the data points or measuring points combined into the object is identical or below a predetermined threshold value. If this is the case, the assignment of the data points is confirmed, otherwise the assignment is adjusted.

An assignment of the measuring points to the objects or the generated objects can also be adapted if the position of the objects in the room changes.

If, for example, a truck on the monitored area changes its direction of travel, its extent in the direction of observation and the movement of the measuring points also change.

Due to the temporal evaluation of the time course of the measuring points as well as their relative speed to each other, it can be determined by the continuous monitoring and evaluation that a change of the movement has taken place in the room.

Accordingly, an adaptation of the generated objects can then take place. If the temporal course of the object information is evaluated by one or more data acquisition units, then not only the movement of the object can be detected, but also the extent of the object, such as length, width and height can be calculated.

If, for example, it is detected by the data acquisition unit that various Measuring points, which are at a distance of 10 m to each other, move relative to the same speed, it can be seen that the detected object must have at least a length of 10 m.

If, for example, measuring points at a distance x, x-1, x-2, x-3 meters are detected at time t0, and measuring points at a distance x-1, x-2, x-3, x-4 are detected at time t1, then can be determined that a) all measuring points due to their identical speed, for example, due to a relative speed of 0 belong to the same object, b) the object has a length of at least 3 meters and c) the object has a speed of 1 meter per unit time ,

Also, with the arrangement written above, the object can be tracked even if the object can not be detected in the meantime since it is behind another object, for example.

This can be done, for example, by having a data record with the detected or generated objects and their speeds and possibly directions of movement in the memory unit 102 through the data processing unit 103 is stored or stored and this object is assigned an object identifier.

Be by one or more detection units 101 Issued measuring points, which have no direct predecessor in the previous time course, it can be checked based on the objects stored in the memory unit, whether an object with similar relative speed and / or similar dimensions and / or similar position has been previously detected. The object can then be assigned the same object identifier.

It is thus possible to track objects, even if they are temporarily hidden by another object.

Regarding 2 the arrangement of the one or more data acquisition units shall be described on the basis of a parking space.

2 shows an example of a truck parking lot. Here, each park row of the parking lot is assigned at least one data acquisition unit. The data acquisition units 101 are here or raised on a mast or similar. (not shown) and / or mounted on a bridge (centered in the image) to allow detection over a greater distance and over other objects. The data acquisition area or the data acquisition line is arranged along the park row. Thus, each data acquisition device is 101 set up to measure the distance of objects or measurement points along the data collection line in a park row.

Already by a detection unit several consecutive objects (eg. Several objects in a row in 2 ) and distinguish them, since the measuring points can be assigned to the objects by the method described above and the objects can thus be distinguished.

As further shown, may also be at a distance from the first data acquisition units 101 (bottom of picture) a second row of data acquisition units 101 (in the middle of the picture) be provided. This is necessary if the park row has a length which can not be reliably covered by only one data acquisition unit.

It is thus covered or monitored using the multiple data acquisition units, the total surveillance area or the entire parking lot.

Such a system with multiple data acquisition units for monitoring a surface is referred to herein as a large area monitoring system.

As can be seen, the data acquisition areas of the data acquisition units can be arranged adjacent to each other. In particular, the data acquisition areas can be arranged substantially parallel to one another.

The adjacent arrangement of the data acquisition areas of the data acquisition units 101 allows an evaluation of the movement of measuring points relative to each other, not only in relation to a data acquisition unit but also over several data acquisition units. As a result, an object can be tracked over the measurement range of multiple data acquisition units.

Accordingly, the data processing unit 103 be configured to process the time course of measuring points of the multiple data acquisition units. In this case, the data processing unit can be set up in particular to evaluate the time profile of the measuring points of the data acquisition units and especially the course of measuring points between the data acquisition units.

Such detection of objects across multiple data acquisition units will be described below.

At a time t0, a measurement point becomes a distance of x from the data acquisition unit 1 detected. At a later time t1 is also by the adjacent data acquisition unit 2 a measuring point at a distance x detected. Again, at a later time t3, no more measuring point can be detected by the data acquisition unit 1 are detected at the distance x, but at this time by the adjacent data acquisition unit 2 the measuring point is still detected at the distance x. It can therefore be assumed that the object is at a distance x from the detection area of the data acquisition unit 1 into the measuring range of the data acquisition unit 2 has moved. As the distance between the data acquisition units 101 is known, and the object was detected by both data acquisition units at the same time, the length of the object can also be estimated by the data processing unit.

For example, the truck in the in 2 shown example when entering the parking lot already data acquisition areas of multiple data acquisition units 101 drive through, can already be made here an accurate estimation of the dimensions.

However, as already described above, it is also possible to estimate the dimensions when the lorries enter the lane assigned to the data acquisition unit, since there are already several measurement points in the area of the data acquisition unit which move at the same speed. If, for example, measuring points in the range between x and y meters are detected by the data acquisition unit, which move uniformly or synchronously over time, it can be concluded that the object of y-x has a length.

Furthermore, the data processing unit can preferably be designed to extract from the object information data sets of the plurality of data acquisition units regions that are not occupied by objects, and preferably to determine the extent and / or position of the free regions. For this purpose, the data processing device can be set up to determine free regions between the objects based on the temporal course of the object information of a plurality of object information data sets of the plurality of data acquisition units.

Thus, for example, free parking spaces for trucks or cars in a parking lot, free spaces for containers in a container port or free berths in a port can be determined.

Data of the data processing unit can furthermore preferably be transmitted to terminals (not shown) by means of a transmission unit (not shown).

By means of the terminals, which may have a display as data display unit, for example, users of the system can be informed about a current occupancy of the area.

The term "unit" in the preceding description is not limited to a physical unit. The above-mentioned units, such as the data acquisition unit, can also be, for example, software or routines which can be implemented on a common chip or distributed over several chips or computers.

In addition, it should be noted that "comprising" and "having" does not exclude other elements or steps, and "a" or "an" does not exclude a multitude. It should also be appreciated that features or steps described with reference to any of the above embodiments may also be used in combination with other features or steps of other embodiments described above. Reference signs in the claims are not to be considered as limiting.

Claims (10)

An object detection system for detecting and tracking objects, comprising: - at least one data acquisition unit for acquiring object information; A memory unit for storing the object information output by the at least one data acquisition unit, wherein the storage unit stores an object information record for each of the at least one data acquisition unit, each object information record storing a plurality of object information of the respective data acquisition unit at different times from the data acquisition unit has been issued; and a data processing unit which is set up objects based on a temporal course of the object information of the at least one object information record to generate.  An object detection system according to any one of the preceding claims, wherein - Each of the at least one data acquisition unit in each case as a laser scanner, in particular as a 2D laser scanner is formed; - The object information of the at least one data acquisition unit each comprises a plurality of data points; - The data processing device is set up for generating objects to summarize data points of the object information of the at least one data acquisition unit to an object, based on a time course of the data points.  An object detection system according to any one of the preceding claims, wherein - the timeline of at least one of: A relative movement of at least two data points, determined based on object information, recorded at different times by the at least one data acquisition unit; A time sequence of at least two data points of at least two data acquisition units.  An object detection system according to any one of the preceding claims, wherein - The system at least two data acquisition units, which are each designed as a laser scanner, in particular as a 2D laser scanner; - The object information of the at least two data acquisition units each comprise a plurality of data points; - Data acquisition ranges of the at least two data acquisition units are arranged adjacent to each other; - The data processing device is set up, based on the time profile of the data points of the data information of the at least two data acquisition units, the data points to summarize objects.  An object detection system according to any one of the preceding claims, wherein The data processing device is set up, based on the time profile of the object information of the at least one object information data set, to determine at least one of: The extent of the at least one object; The length of the at least one object; The width of the at least one object; - the height of the at least one object; The position of the at least one object.  An object detection system according to any one of the preceding claims, wherein - The data processing device is set up to track the objects based on the time course of the object information of the at least one object information record.  Large area monitoring system for monitoring a large area, comprising an object detection system according to one of the preceding claims, wherein - A plurality of data acquisition units is arranged such that data acquisition ranges of the plurality of data acquisition units are arranged adjacent to each other; - The data processing device is configured to determine free areas between the objects based on the time history of the object information of a plurality of object information data set of the plurality of data acquisition units.  An object detection system according to any one of the preceding claims, further comprising A transmitting unit for transmitting information regarding the free areas and the detected objects to at least one mobile unit.  An object detection system according to claim 8, further comprising - the at least one mobile unit, where - The mobile unit has a receiving unit and a data display unit and is adapted to receive information from the transmitting unit by means of the receiving unit and output information based on the received information regarding the free areas and / or the detected objects to a user via the data display unit.  An object detection system according to any one of the preceding claims, wherein - the large area is one of - a parking lot; - a container handling center; or - a port; and - the objects of at least one are of: - trucks; - cars; - buses; - containers; and or - ships.

Images