CN113271440B

CN113271440B - Method, device, equipment and storage medium for obtaining action track of moving object

Info

Publication number: CN113271440B
Application number: CN202110552026.7A
Authority: CN
Inventors: 黄黎滨; 袁琪
Original assignee: Zhejiang Uniview Technologies Co Ltd
Current assignee: Zhejiang Uniview Technologies Co Ltd
Priority date: 2021-05-20
Filing date: 2021-05-20
Publication date: 2023-06-20
Anticipated expiration: 2041-05-20
Also published as: CN113271440A

Abstract

The application discloses a method for obtaining a movement track of a moving target, which comprises the following steps: determining entrance and exit information in a visual range of each image acquisition device in the monitoring area; determining access information between the entrances and exits of different image acquisition devices in the monitoring area; constructing a video road network according to the access information and the access information; and obtaining the action track of the moving target in the monitoring area based on the video road network. By using the technical scheme provided by the application, the video road network is constructed according to the access information of the image acquisition equipment and the access information of the access between the image acquisition equipment, so that the access condition of each access can be better embodied, and therefore, based on the video road network, the more accurate action track of the moving target in the monitoring area can be obtained, and the moving target can be effectively monitored, tracked and the like. The application also discloses a device, equipment and storage medium for obtaining the action track of the moving target, which have corresponding technical effects.

Description

Method, device, equipment and storage medium for obtaining action track of moving object

Technical Field

The present invention relates to the field of monitoring technologies, and in particular, to a method, an apparatus, a device, and a storage medium for obtaining a movement track of a moving target.

Background

In the scenes of video monitoring, target tracking and the like, image acquisition is often performed by using image acquisition equipment, and the acquired images are analyzed to obtain the action track of the moving target. If the moving object moves in the visible range of one image acquisition device, the action track of the moving object in the visible range of the image acquisition device is easily obtained by analyzing the image acquired by the image acquisition device. However, if the moving range of the moving object exceeds the visual range of the image capturing device, there is no better method for obtaining a more accurate action track of the moving object under the condition in the visual range of a plurality of image capturing devices.

Disclosure of Invention

The application aims to provide a method, a device, equipment and a storage medium for obtaining a movement track of a moving target so as to obtain a more accurate movement track of the moving target in a monitoring area.

In order to solve the technical problems, the application provides the following technical scheme:

A method for obtaining action tracks of moving targets comprises the following steps:

determining entrance and exit information in a visual range of each image acquisition device in the monitoring area;

determining access information between the entrances and exits of different image acquisition devices in the monitoring area;

constructing a video road network according to the access information and the access information;

and obtaining the action track of the moving target in the monitoring area based on the video road network.

In a specific embodiment of the present application, the determining the access information in the visible range of each image capturing device in the monitoring area includes:

aiming at each image acquisition device in the monitoring area, acquiring multi-frame images continuously acquired by the current image acquisition device in a set first time period;

respectively determining the moving positions of a plurality of first reference targets in each frame of image;

obtaining a movement track of each first reference target in the visual range of the current image acquisition equipment based on the moving position of each first reference target;

and performing aggregation processing on the starting points and the ending points of the action tracks of all the first reference targets to obtain the entrance and exit information in the visual range of the current image acquisition equipment.

In a specific embodiment of the present application, the determining access information between the entrances and exits of different image capturing devices in the monitored area includes:

obtaining access passage data of a plurality of second reference targets in the monitoring area in a set second time period;

and determining whether access is achieved between the entrances and exits of different image acquisition devices according to the entrance and exit passing data.

In a specific embodiment of the present application, the determining, according to the passageway passing data, whether the passageway is accessed between the passageway of the different image capturing devices includes:

respectively arranging the access passage data of each second reference target according to the time sequence;

determining the reachable times among the entrances and exits belonging to different image acquisition devices based on the arranged entrance and exit passing data;

for any two entrances and exits belonging to different image acquisition equipment, determining whether the current two entrances and exits are accessed according to the reachable times between the current two entrances and exits.

In a specific embodiment of the present application, the determining the number of reachable times between the gates belonging to different image capturing devices based on the aligned gate passing data includes:

Initializing the reachable times between the entrances and exits of different image acquisition devices;

for each second reference target, determining a time interval of the current second reference target sequentially passing through two entrances and exits belonging to different image acquisition devices according to the entrance and exit passing data of the current second reference target;

if the time interval is smaller than the set interval threshold, the number of times that is reachable between the corresponding two entrances and exits is increased.

In a specific embodiment of the present application, the determining, according to the number of times of accessibility between the current two entrances, whether the current two entrances are accessed includes:

and if the average reachable times between the current two entrances are larger than a preset time threshold, determining that the current two entrances are accessed.

In a specific embodiment of the present application, after the constructing a video road network, the method further includes:

for any two entrances and exits belonging to different image acquisition equipment, if the average reachable times between the current two entrances and exits are detected to have a falling trend and fall to be smaller than or equal to the times threshold, deleting reachable channels between the current two entrances and exits in the video road network.

repeatedly executing the steps of determining the access information in the visible range of each image acquisition device in the monitoring area and determining the access information between the access of different image acquisition devices in the monitoring area under the condition that the set updating condition is reached, and obtaining new access information and access information;

and updating the video road network based on the new access information and the access information.

In an embodiment of the present application, the obtaining, based on the video road network, a movement track of a moving target in the monitoring area includes:

obtaining an initial track of a moving target in the monitoring area;

determining missing passing points according to the video road network, wherein the missing passing points are passing points which are not included in the initial track and are passed by the moving target;

and adding the missing passing points in the initial track to obtain the action track of the moving target in the monitoring area.

In one specific embodiment of the present application, after the obtaining the initial trajectory of the moving object in the monitored area, the method further includes:

According to the video road network, determining abnormal passing points in the passing points contained in the initial track;

and outputting prompt information aiming at the abnormal passing points.

In one embodiment of the present application, after the obtaining the action track of the moving object in the monitored area, the method further includes:

predicting a next gateway in the monitoring area to which the moving target is to move based on the video road network and the action track of the moving target;

and controlling the image acquisition equipment corresponding to the next access opening to be in an open state so as to monitor the moving target.

and determining the current position of the moving target according to the obtained reachable duration between the entrances and exits in the video road network and the action track of the moving target.

An action trajectory acquisition device of a moving object, comprising:

the entrance and exit information determining module is used for determining entrance and exit information in the visible range of each image acquisition device in the monitoring area;

the access information determining module is used for determining access information between the entrances and exits of different image acquisition devices in the monitoring area;

The video road network construction module is used for constructing a video road network according to the access information and the access information;

and the action track obtaining module is used for obtaining the action track of the moving target in the monitoring area based on the video road network.

An action trajectory acquisition device of a moving object, comprising:

a memory for storing a computer program;

a processor for implementing the action trajectory acquisition method of the moving object according to any one of the above when executing the computer program.

A computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements the steps of the action trajectory acquisition method of a moving object of any one of the above.

By applying the technical scheme provided by the embodiment of the application, the access information of each image acquisition device in the monitoring area in the visible range is determined, the access information between the access of different image acquisition devices in the monitoring area is determined, the video road network is constructed according to the access information and the access information, and then the action track of the moving target in the monitoring area is obtained based on the video road network. The video road network is constructed according to the access information in the visible range of each image acquisition device and the access information between the access of different image acquisition devices, so that the access condition of each access of each image acquisition device in the monitoring area can be better reflected, and the more accurate action track of the moving target in the monitoring area can be obtained based on the video road network, thereby effectively monitoring, tracking and the like the moving target.

Drawings

In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described below, it being obvious that the drawings in the following description are only some embodiments of the present application, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.

FIG. 1 is a flowchart of a method for obtaining a motion trajectory of a moving object according to an embodiment of the present application;

fig. 2 is a schematic view of a visual range of the image capturing apparatus 1 in the embodiment of the present application;

fig. 3 is a schematic view of a visual range of the image capturing apparatus 2 according to the embodiment of the present application;

FIG. 4 is a schematic diagram of an example of a video network in an embodiment of the present application;

FIG. 5 is a schematic diagram illustrating determining doorway information according to an embodiment of the present application;

FIG. 6 is another schematic diagram of determining doorway information according to an embodiment of the present application;

fig. 7 is a schematic diagram of another example of a video network in an embodiment of the present application;

fig. 8 is a schematic structural diagram of a device for obtaining a movement track of a moving object according to an embodiment of the present application;

Fig. 9 is a schematic structural diagram of a movement track obtaining apparatus for a moving object in an embodiment of the present application.

Detailed Description

The core of the application is to provide a method for obtaining the action track of a moving target, which can be applied to equipment such as a server and the like capable of performing image analysis. The server can be in communication connection with each image acquisition device to obtain images in the corresponding visual range acquired by each image acquisition device.

The method comprises the steps of firstly determining the access information of each image acquisition device in the monitoring area in the visible range, determining the access information between the access of different image acquisition devices in the monitoring area, constructing a video road network according to the access information and the access information, and obtaining the action track of the moving target in the monitoring area based on the video road network. The video road network is constructed according to the access information in the visible range of each image acquisition device and the access information between the access of different image acquisition devices, so that the access condition of each access of each image acquisition device in the monitoring area can be better reflected, and the more accurate action track of the moving target in the monitoring area can be obtained based on the video road network, thereby effectively monitoring, tracking and the like the moving target.

In order to provide a better understanding of the present application, those skilled in the art will now make further details of the present application with reference to the drawings and detailed description. It will be apparent that the described embodiments are only some, but not all, of the embodiments of the present application. All other embodiments, which can be made by one of ordinary skill in the art without undue burden from the present disclosure, are within the scope of the present disclosure.

Referring to fig. 1, a flowchart of an implementation of a method for obtaining an action track of a moving object according to an embodiment of the present application may include the following steps:

s110: access information within the visual range of each image acquisition device in the monitored area is determined.

In this embodiment of the present application, the monitoring area is an area needing to be focused, and multiple image capturing devices may be deployed in advance in the monitoring area, where each image capturing device in the monitoring area has a respective visual range, and multiple inlets and outlets may exist in the visual range of each image capturing device.

The number of gates, the gate position, etc. of gates within the visual range of each image capturing device in the monitoring area can be determined first. Specifically, the technician can determine the entrance and exit information in the visual range of each image acquisition device through a manual judgment mode, and can also determine the entrance and exit information in the visual range of each image acquisition device by analyzing the image in the corresponding visual range acquired by each image acquisition device.

For example, fig. 2 is a schematic view of the visual range of the image capturing apparatus 1, and as can be seen from fig. 2, the image capturing apparatus 1 has three ports A, B, C, where port a is at the upper left part of the visual range of the image capturing apparatus 1, port B is at the upper middle part of the visual range of the image capturing apparatus 1, and port C is at the lower part of the visual range of the image capturing apparatus 1. For another example, fig. 3 is a schematic view of the visual range of the image capturing apparatus 2, and as can be seen from fig. 3, the image capturing apparatus 2 has D, E, F three inlets and outlets, where the inlet D is at the middle upper portion of the visual range of the image capturing apparatus 2, the inlet E is at the left lower portion of the visual range of the image capturing apparatus 2, and the inlet F is at the right upper portion of the visual range of the image capturing apparatus 2.

In this embodiment of the present application, for any one image capturing device, the visual range of the image capturing device may be accessed through the access opening within the visual range of the image capturing device, and even if the visual range of the image capturing device is an open area, the access information for accessing the visual range of the image capturing device may be obtained.

S120: access information between the entrances and exits of different image acquisition devices in the monitored area is determined.

In the embodiment of the present application, the entrances and exits of the same image capturing apparatus are considered to be in communication with each other. As shown in fig. 2, the access opening A, B, C in the visible range of the image capturing apparatus 1 communicates with each other, and as shown in fig. 3, the access opening D, E, F in the visible range of the image capturing apparatus 2 communicates with each other.

After determining the access information of each image acquisition device in the monitoring area in the visual range, the access information between the access of different image acquisition devices in the monitoring area can be further determined, namely whether the access between the access of different image acquisition devices is determined.

The access information between the entrances and exits of different image acquisition devices in the monitoring area can be determined according to the actual exploration condition, images in corresponding visible ranges acquired by the image acquisition devices can be analyzed, the access information between the entrances and exits of the different image acquisition devices is determined, the access information can comprise whether the access between the entrances and exits of the different image acquisition devices is achieved or not, and the access information can also comprise information such as the reachable duration between the entrances and exits of the different image acquisition devices. Access between the entrances and exits of different image acquisition devices means that one entrance and exit of one image acquisition device can reach one entrance and exit of another image acquisition device, and no other entrance and exit exists in the middle.

For any two image capturing apparatuses, whether access between the entrances and exits of the two image capturing apparatuses is possible may be determined according to a preset specification. For example, if the length of time from one entrance to one exit of one image capturing apparatus to one entrance of the other image capturing apparatus is short and within a set length of time, it is possible to determine that there is a communication between the two entrances of the two image capturing apparatuses. The time period from one entrance of one image capturing apparatus to one entrance of the other image capturing apparatus is an reachable time period.

The access information between the access opening of the image acquisition apparatus 1 and the access opening of the image acquisition apparatus 2 is as follows: the entrance B of the image capturing apparatus 1 communicates with the entrance D of the image capturing apparatus 2, that is, a pedestrian or a vehicle can reach the entrance D of the image capturing apparatus 2 from the entrance B of the image capturing apparatus 1 without passing through other entrances.

S130: and constructing a video road network according to the access information and the access information.

After determining the access information of each image acquisition device in the monitoring area within the visual range and the access information between the access of different image acquisition devices in the monitoring area, a video road network can be constructed according to the access information and the access information.

Specifically, the number of the entrances and exits in the visual range of each image acquisition device and the position of each entrance and exit can be determined according to the entrance and exit information and the access information, meanwhile, whether the entrances and exits of different image acquisition devices are accessed or not can be determined, and an accessible channel of the accessible entrances and exits is established, so that a video road network in a monitoring area is constructed.

For example, if four image capturing devices are disposed in the monitoring area, the image capturing device Cam1, the image capturing device Cam2, the image capturing device Cam3, and the image capturing device Cam4 are respectively disposed. Wherein, there are A, B, C three access points in the visual scope of Cam1, D, E, F three access points in the visual scope of Cam2, G, H, I, J four access points in the visual scope of Cam3, K, L two access points in the visual scope of Cam4. According to a certain rule, the access information between the entrances and exits of different image devices is determined as follows: the port B of Cam1 communicates with the port D of Cam2, the port C of Cam1 communicates with the port I of Cam3 and the port K of Cam4, respectively, and the port K of Cam4 also communicates with the port F of Cam2 and the port I of Cam3, respectively. It is possible to establish reachable passages between the gate B of Cam1 and the gate D of Cam2, between the gate C of Cam1 and the gate I of Cam3 and the gate K of Cam4, and between the gate K of Cam4 and the gate F of Cam2 and the gate I of Cam 3. From this, a video road network within the monitoring area can be constructed, as shown in fig. 4.

The number of the entrances and the exits in the visual range of each image acquisition device in the monitoring area and the positions of each entrance and exit can be clearly known through the constructed video road network, and meanwhile, the existence of the reachable channels between the entrances and exits of the image acquisition devices can be known.

S140: and obtaining the action track of the moving target in the monitoring area based on the video road network.

In the embodiment of the present application, the object of the pedestrian, the vehicle, or the like, which is currently focused on, may be determined as the moving target. After a video road network is constructed according to the access information of each image acquisition device in the monitoring area in the visible range and the access information of different image acquisition devices in the monitoring area, the action track of the moving target in the monitoring area can be obtained based on the video road network.

The access information in the visual range of each image acquisition device and the access information between the access points of different image acquisition devices can be clearly obtained through the video road network, so that when the moving object needs to be monitored, the characteristic recognition can be carried out in the images acquired by each image acquisition device through the identifiable characteristic of the moving object, whether the moving object appears in the visual range of each image acquisition device or not and the appearance time are searched, the action track of the moving object in the visual range of each image acquisition device is obtained, and then according to the access information between the access points of different image acquisition devices in the video road network, the access point of the moving object possibly reaching the visual range of other image acquisition devices through which access point in the visual range of the image acquisition device is obtained, and accordingly, the action track of the moving object in the visual range of each image acquisition device is integrated, and finally the action track of the moving object in the monitoring area is obtained accurately.

Specifically, after obtaining the action track of the moving object in the visible range of each image acquisition device, if any one image acquisition device in the monitoring area monitors that the moving object moves out of one access opening in the visible range of the image acquisition device, other access openings which are possibly multiple in number and are accessed to the access opening can be determined according to the video road network, and then the access opening of the other image acquisition device which is accessed to the moving object after the access opening in the visible range of the image acquisition device moves out is determined according to the appearance time of the moving object in the access opening and each other access opening which is accessed to the access opening. The operation is carried out on each image acquisition device, so that the moving target can enter the visual range of the image acquisition device after moving out of the visual range of the image acquisition device, and the action track of the moving target in the visual range of the corresponding image acquisition device can be integrated, so that the complete and accurate action track of the moving target in the monitoring area is obtained.

By the method provided by the embodiment of the application, the access information of each image acquisition device in the monitoring area in the visible range is determined, the access information between the access of different image acquisition devices in the monitoring area is determined, a video road network is constructed according to the access information and the access information, and then the action track of the moving target in the monitoring area is obtained based on the video road network. The video road network is constructed according to the access information in the visible range of each image acquisition device and the access information between the access of different image acquisition devices, so that the access condition of each access of each image acquisition device in the monitoring area can be better reflected, and the more accurate action track of the moving target in the monitoring area can be obtained based on the video road network, thereby effectively monitoring, tracking and the like the moving target.

In one embodiment of the present application, step S110 may include the steps of:

step one: aiming at each image acquisition device in the monitoring area, acquiring multi-frame images continuously acquired by the current image acquisition device in a set first time period;

step two: respectively determining the moving positions of a plurality of first reference targets in each frame of image;

step three: obtaining a movement track of each first reference target in the visible range of the current image acquisition equipment based on the moving position of each first reference target;

step four: and performing aggregation processing on the starting points and the ending points of the action tracks of all the first reference targets to obtain the entrance and exit information in the visual range of the current image acquisition equipment.

For ease of description, the steps described above are combined.

In the embodiment of the present application, in the case where each image capturing device in the monitoring area is in the on state, a multi-frame image continuously captured by each image capturing device in the set first period of time may be obtained. The first period of time may be set and adjusted according to the actual situation, such as 30 minutes, 1 hour, etc.

For each image acquisition device in the monitoring area, after multi-frame images continuously acquired by the current image acquisition device in a first time period are obtained, a plurality of first reference targets can be further determined from the previous several frames of images. The first reference object may be an object, such as a pedestrian, a vehicle, etc., that appears at the edge of the visual range of the current image capturing device in the previous several frames of images. Because the embodiment of the application is to perform the operation of determining the entrance and exit information in the corresponding visual range for each image acquisition device in the monitoring area, the image acquisition device for which the current operation is performed is the current image acquisition device. For example, when the first image acquisition device in the monitoring area is used for determining the entrance and exit information in the visual range, the current image acquisition device is the first image acquisition device, and when the second image acquisition device in the monitoring area is used for determining the entrance and exit information in the visual range, the current image acquisition device is the second image acquisition device.

The first reference object has a certain identifiable characteristic, if the first reference object is a pedestrian, the identifiable characteristic may be a face characteristic, a body characteristic, etc., and if the first reference object is a vehicle, the identifiable characteristic may be a license plate characteristic, a body characteristic, etc.

For any one first reference object, according to the identifiable characteristic of the first reference object, the obtained multi-frame images continuously collected by the current image collecting device in a set first time period are analyzed, so that the moving position of the first reference object in each frame of image can be determined.

After the moving positions of the plurality of first reference targets in each frame of image are respectively determined, the action track of each first reference target in the visible range of the current image acquisition device can be obtained based on the moving position of each first reference target. Because the multi-frame images continuously acquired by the current image acquisition device in the set first time period are obtained, the images have time continuity, and after the moving positions of the first reference target in each frame of images are obtained, the moving positions arrive in time sequence, so that the corresponding moving track of the first reference target in the visible range of the current image acquisition device can be obtained based on the moving positions of the same first reference target.

After the action track of each first reference target in the visual range of the current image acquisition device is obtained, the starting point and the ending point of the action track of all the first reference targets are subjected to aggregation processing, so that the entrance and exit information in the visual range of the current image acquisition device can be obtained.

Specifically, a minimum rectangular aggregation algorithm may be used to aggregate the start points and the end points of the action tracks of all the first reference targets. The minimum rectangular aggregation algorithm is: at least any two points are required to form a minimum rectangle, and the sum of all rectangular areas is minimum.

For example, each scheme of performing the aggregation processing on the start points and the end points of the action tracks of all the first reference targets may be listed, where the aggregation means that at least two points form a rectangle, and all the start points and the end points in each scheme are included in a certain rectangle. And then calculating the sum of the rectangular areas of each scheme, determining the scheme with the smallest sum of the rectangular areas as the final aggregation scheme, and obtaining the entrance and exit information in the visual range of the current image acquisition equipment according to the aggregation scheme.

The operation of the steps is carried out on each image acquisition device, so that the entrance and exit information in the visible range of each image acquisition device in the monitoring area can be accurately obtained, and the corresponding entrance and exit information can be stored in the database.

For ease of understanding, the determination of the doorway information is exemplified below.

Assume that the first reference target is a pedestrian. Through intelligent capturing of the moving human body, the moving position of the pedestrian in the image acquired by the corresponding image acquisition equipment can be obtained, and the lowest point selected by the human body frame can be regarded as the moving position of the pedestrian, such as the point P shown in fig. 5. The position where the pedestrian can reach is a road, and through analyzing continuous multi-frame images of the same image acquisition device, the moving positions of a plurality of pedestrians in different moments within the visible range of the image acquisition device can be obtained, so that the action track of each pedestrian in the visible range of the image acquisition device can be obtained.

As shown in fig. 5, three action tracks, track 1, track 2, and track 3, are obtained. The starting and ending points of the three tracks are the six points a, b, c, d, e, f, respectively. According to the minimum rectangle aggregation algorithm, when ab forms a rectangle, cd forms a rectangle, and ef forms a rectangle, the sum of the rectangular areas is minimized. And (5) performing aggregation treatment on the six points to obtain three aggregation points, namely A, B, C, so as to obtain the entrance and exit information in the visible range of the image acquisition equipment.

In this example, each of the schemes of aggregation processing for the six points a, b, c, d, e, f may be listed, for example, one scheme is to form ab into a rectangle, cd into a rectangle, and ef into a rectangle, one scheme is to form abcd into a rectangle, ef into a rectangle, and other schemes are possible, and are not listed. The sum of the rectangular areas of each scheme is calculated separately, taking the two schemes listed as examples, as shown in fig. 6, the sum of the rectangular areas of the former scheme is significantly smaller than the sum of the rectangular areas of the latter scheme, and the scheme with the smallest sum of the rectangular areas is determined as the final aggregation scheme. The final polymerization scheme is: ab forms a rectangle, cd forms a rectangle, ef forms a rectangle, and according to the aggregation scheme, entrance and exit information in the visual range of the image acquisition equipment is obtained, wherein A, B, C three entrances exist.

By the method, the problem of merging the entrances and exits in the visual range of the image acquisition equipment can be well solved, so that different road directions in the visual range of the same image acquisition equipment are obtained, and a refined video road network is formed conveniently.

In one embodiment of the present application, step S120 may include the steps of:

the first step: obtaining entrance and exit passing data of a plurality of second reference targets in a monitoring area in a set second time period;

and a second step of: and determining whether access is achieved between the entrances and exits of different image acquisition devices according to the entrance and exit passing data.

For ease of description, the two steps described above are combined.

In the embodiment of the present application, in the case where each image capturing device in the monitoring area is in the on state, a plurality of frame images continuously captured by each image capturing device in the set second period of time may be obtained. The second period of time may be set and adjusted according to the actual situation, such as 30 minutes, 1 hour, etc.

For each image acquisition device in the monitoring area, after multi-frame images continuously acquired by the current image acquisition device in a second time period are obtained, a plurality of second reference targets can be further determined from the previous several frames of images. The second reference object may be an object, such as a pedestrian, a vehicle, etc., that appears at the edge of the visual range of the current image capturing device in the previous several frames of images. The second reference target has a certain identifiable characteristic.

For any one second reference object, according to the identifiable characteristic of the second reference object, the obtained multi-frame images continuously collected by the current image collecting device in the set second time period are analyzed, so that the time when the second reference object passes through which gateway can be determined. So that the gateway passing data of the plurality of second reference targets in the monitoring area within the set second period of time can be obtained. The doorway passing data may include an identification of the doorway that passed and a time of passing.

According to the access passing data, whether access between the access of different image acquisition devices is achieved or not can be determined. And obtaining access information between the entrances and exits of different image acquisition devices in the monitoring area.

In a specific embodiment of the present application, the access passage data of each second reference object may be arranged according to the time sequence of the passage, then the number of times of accessibility between the access ports belonging to different image capturing devices is determined based on the arranged access passage data, and then, for any two access ports belonging to different image capturing devices, whether access between the current two access ports is performed is determined according to the number of times of accessibility between the current two access ports. The current two entrances refer to two entrances for which the current operation is performed. The doorway passing data may include data such as a doorway identification of the passage, a time of passage, and the like. The doorway identification may uniquely identify one doorway as which doorway is within the visual range of which image capturing apparatus.

In this embodiment of the present application, after obtaining the passage data of the entrances and exits in the monitoring area of the plurality of second reference targets in the set second time period, the passage data of the entrances and exits of each second reference target may be arranged according to the passage time sequence, so as to determine the time sequence of the entrances and exits of each second reference target.

Based on the arranged passageway passing data, the reachable times among the passageways belonging to different image acquisition devices can be determined. For any one second reference object, through the access passage data of the second reference object after arrangement, it can be known which access passage the second reference object passes through, and for two adjacent access passages which do not belong to the same image acquisition device, if the second reference object passes through the two access passages in a shorter time, it can be considered that the access passage is reachable between the two access passages. Such analysis and judgment are performed on the access passage data of each second reference object, so that the reachable times among the access passages belonging to different image acquisition devices can be determined.

For any two entrances and exits belonging to different image acquisition equipment, whether the current two entrances and exits are accessed or not can be determined according to the number of times of access between the current two entrances and exits. Specifically, if the average reachable times between the current two entrances is greater than the preset times threshold, determining that the current two entrances are accessed. The average reachable times can be calculated according to days or weeks, if the reachable times between the current two entrances and exits within a week are counted to be X, the average reachable times between the current two entrances and exits can be obtained to be X/7 by calculating according to days.

For any two ports, if there is a small number of times that the two ports are reachable, but the average number of times that the two ports are reachable is counted, it is considered that there is no communication between the two ports. If road damage, road repair and the like occur between two entrances and exits belonging to different image acquisition devices, fewer constructors may pass through the two entrances and exits, so that the average reachable times are fewer.

A threshold number of times may be preset, and only when the average number of times of access between the current two entrances is greater than the threshold number of times, the access between the current two entrances is determined. When the average reachable times between the current two entrances is smaller than or equal to the threshold of times, the average reachable times are considered to be smaller, and the current two entrances cannot be accessed. Thus, access information of the gateway between the image acquisition devices can be determined more accurately. The threshold number of times can be set and adjusted according to actual conditions.

In one embodiment of the present application, the reachable times between the gates belonging to different image capturing apparatuses may be determined based on the aligned gate passing data by:

Step one: initializing the reachable times between the entrances and exits of different image acquisition devices;

step two: for each second reference target, determining the time interval of the current second reference target passing through two entrances and exits belonging to different image acquisition devices according to the entrance and exit passing data of the current second reference target arranged according to the passing time sequence;

step three: if the time interval is less than the set interval threshold, the number of times reachable between the corresponding two gates is increased.

For ease of description, the steps described above are combined.

After the passage data of the entrances and exits of each second reference object are arranged according to the passage time sequence, the reachable times between the entrances and exits of different image acquisition devices can be initialized first, for example, the reachable times between the entrances and exits of different image acquisition devices can be initialized to 0.

For each second reference target, determining which of the entrances belonging to different image acquisition devices the current second reference target sequentially passes through according to the entrance-exit passing data of the current second reference target, and determining the time interval of the current second reference target sequentially passing through the two entrances belonging to different image acquisition devices. The current second reference target is the second reference target for the current operation. For example, when the corresponding gateway passing data is to be processed for the first second reference target, the current second reference target is the first second reference target, and when the corresponding gateway passing data is to be processed for the second reference target, the current second reference target is the second reference target.

For example, the gate passing data of a certain second reference target arranged according to the time sequence of passing is: from the fact that the second reference object sequentially passes through the port B, D, E from the port B (time 1) to the port D (time 2) to the port E (time 3), the port B, D belongs to different image capturing devices, and then the time interval between the two ports belonging to different image capturing devices is the time interval from time 1 to time 2.

An interval threshold may be set, and if the time interval that the current second reference object sequentially passes through two entrances and exits belonging to different image capturing devices is smaller than the interval threshold, the corresponding two entrances and exits may be considered to be closer, the possibility of access is greater, and the number of times of access of the corresponding two entrances and exits may be increased, for example, the number of times of access will be +1. If the time interval is greater than or equal to the interval threshold, the corresponding two entrances are considered to be far apart, the possibility of access is not great, and the number of times of access of the corresponding two entrances is not required to be increased. The interval threshold can be set and adjusted according to practical situations.

By comparing the time interval of sequentially passing through two entrances and exits belonging to different image acquisition devices with the interval threshold value, the reachable times between the entrances and exits belonging to different image acquisition devices can be accurately determined, so that the accessibility information between the entrances and exits of different image acquisition devices can be accurately determined later.

In one embodiment of the present application, after the video road network is constructed, the method may further include the steps of:

for any two entrances and exits belonging to different image acquisition equipment, if the average reachable times between the current two entrances and exits are detected to have a falling trend and fall to be smaller than or equal to a time threshold, deleting the reachable channels between the current two entrances and exits in the video road network.

In this embodiment of the present application, after determining the access information of each image capturing device in the monitoring area within the visible range and the access information of the access between the image capturing devices in the monitoring area, and constructing the video road network according to the access information and the access information, the access passing data of the second reference target in the monitoring area may be obtained continuously or according to the set time interval, and the average reachable times between any two accesses belonging to different image capturing devices may be counted.

For any two entrances and exits belonging to different image acquisition equipment, if the average reachable times between the current two entrances and exits are detected to have a falling trend, and fall to be smaller than or equal to a time threshold, the conditions such as road repair or urban road change can be considered to occur, so that the current two entrances and exits are not accessed any more, and a reachable channel between the current two entrances and exits can be deleted in the video road network, so that the accuracy of the follow-up action track of the moving target obtained based on the video road network is improved. The current two gates refer to the two gates for which the current operation is directed.

For example, in the video road network shown in fig. 4, the access port C of Cam1 and the access port I of Cam3 are accessed, if the average reachable times between the access port C and the access port I are found to have a falling trend due to road repair or urban road change, and fall to be less than or equal to the threshold of times, it is considered that the access between the two access ports is no longer accessed, the reachable channels between the two access ports can be deleted in the video road network, and the updated video road network is shown in fig. 7.

the first step: repeatedly executing the steps of determining the access information in the visual range of each image acquisition device in the monitoring area and determining the access information between the access of different image acquisition devices in the monitoring area under the condition that the set updating condition is reached, and obtaining new access information and access information;

and a second step of: based on the new access information and the access information, the video road network is updated.

In this embodiment of the present application, the update condition may be preset, for example, when the update instruction is received, the update condition is considered to be reached, or when the set time interval is reached, the update condition is considered to be reached, or further, when the change of the doorway is detected, the update condition is considered to be reached.

When the update condition is reached, the operations of steps related to step S110 and step S120 may be repeatedly performed, access information in the visible range of each image capturing device in the monitoring area and access information between the accesses of different image capturing devices in the monitoring area are redetermined, new access information and access information are obtained, and then the video road network is updated based on the new access information and access information. Therefore, the method can better cope with the conditions of road repair, urban road change and the like, and can improve the accuracy of the action track of the moving target obtained based on the video road network.

In one embodiment of the present application, step S140 may include the steps of:

step one: obtaining an initial track of a moving target in a monitoring area;

step two: determining missing path points according to the video road network, wherein the missing path points are path points which pass by a moving target and are not included in an initial track;

step three: and adding missing passing points in the initial track to obtain the action track of the moving target in the monitoring area.

For ease of description, the steps described above are combined.

In the embodiment of the application, after the video road network is constructed, when the monitoring requirement for the moving target exists, the initial track of the moving target in the monitoring area can be obtained first. Specifically, images in the corresponding visual range acquired by each image acquisition device in the monitoring area can be analyzed through the identifiable characteristics of the moving target, the entrance through which the moving target sequentially passes is determined, and the initial track of the moving target in the monitoring area is determined according to the entrance through which the moving target sequentially passes. If the initial track of the moving object is A- & gt B- & gt L, that is, the current identified result is that the passing point of the moving object is: the doorway a, the doorway B, and the doorway L do not detect the moving object at the other doorways.

The feature recognition algorithms adopted by different image acquisition devices can be the same or different, and the recognition accuracy degree is different.

After the initial track of the moving target in the monitoring area is obtained, missing passing points can be determined according to the video road network. The missing passing points are passing points that the moving object passes through, and are not included in the initial trajectory. Specifically, the determination of missing pass points can be performed according to a certain path finding algorithm. If the access gateway of each passing point contained in the initial track can be judged one by one according to the video road network, for each passing point contained in the initial track, whether the next passing point of the passing point can be reached through the access gateway of the passing point from the passing point can be further judged, and if the next passing point can be reached, the access gateway can be used as a missing passing point. Alternatively, the access gateway may be determined as a missing route point when the acquired image of the access gateway is analyzed again to determine that the moving object is actually passing.

For example, as shown in fig. 4, the pre-constructed video road network has an initial track of a moving object of a→b→l. That is, in the initial judgment, only the moving targets are known to appear in the visual range of Cam1 and Cam4, and similar objects are not compared in the visual range of other image acquisition devices. If the judgment is simply performed according to the reachable condition between the image acquisition devices, it is considered that both Cam2 and Cam3 may pass through, however, in the embodiment of the present application, after a refined video road network is constructed, the following information can be obtained based on the video road network:

Firstly, because the same image acquisition equipment has stronger reliability on target tracking, if a moving target disappears at the entrance B of the Cam1, the moving target is less likely to reach the Cam3 through the entrance C of the Cam 1;

next, according to the refined video road network, the access between the entrance B of Cam1 and the entrance D of Cam2 can be determined, so that the moving object passes through the entrance D of Cam2 after passing through the entrance B of Cam 1. In addition, since the moving object finally appears at the doorway L of Cam4, and the video road network can determine that the doorway K and the doorway L are included in the visual range of Cam4, the moving object has a high probability of reaching the doorway L through the doorway F of Cam 4. The related collected images of the entrance D and the entrance F of Cam4 in Cam2 can be re-analyzed, for example, the moving target can be re-identified by reducing the similarity threshold, and whether the entrance D and the entrance F are missing route points or not can be determined.

After the missing passing points are determined, the missing passing points can be added into the initial track, so that a more accurate action track of the moving target in the monitoring area can be obtained.

The initial track is obtained firstly, and then the missing passing points are determined according to the video road network, so that more accurate action tracks can be obtained, and meanwhile, the recognition efficiency can be improved.

In one embodiment of the present application, after obtaining the initial trajectory of the moving object within the monitored area, the method may further include the steps of:

the first step: determining abnormal passing points in the passing points contained in the initial track according to the video road network;

and a second step of: and outputting prompt information aiming at the abnormal passing points.

In this embodiment of the present application, after obtaining an initial track of a moving target in a monitoring area, an abnormal route point, that is, an entrance and exit that are unlikely to pass through, among route points included in the initial track may be further determined according to a video road network. Specifically, for each passing point included in the initial track, whether the passing point can be sent from the previous passing point of the passing point is determined according to the video road network, the next passing point of the passing point is not skipped from other entrances and exits, and if not, the passing point can be determined to be an abnormal passing point. After determining the abnormal passing points in the passing points contained in the initial track, the method can output prompt information aiming at the abnormal passing points so as to prompt the user of the current abnormal condition, and can reject the abnormal passing points in the initial track according to the instruction of the user so as to obtain a more accurate action track.

For example. The moving target is a pedestrian, the initial track of the pedestrian is A, C, E, K and L, and if the pedestrian is simply judged according to the reachable condition among the image acquisition devices, because the pedestrian can reach Cam4 through the visible range of Cam2 in the visible range of Cam1, the abnormal passing point can be considered to be absent in the initial track. However, in the embodiment of the present application, in the refined video road network, abnormality determination may be performed on each route point in the initial track separately. For example, for the third passing point E, it can be determined from the video road network that if the passing point K is reached from the passing point C through the passing point E, the entrance D must be passed, and the entrance D is not included in the initial track, so that the passing point E has an abnormality. That is, the pedestrian cannot pass through the entrance C and the exit D is jumped to reach the entrance E, and it is possible to determine that the passing point E among the passing points included in the initial trajectory is an abnormal passing point. The passing point and the entrance are equivalent concepts.

In one embodiment of the present application, after obtaining the action trajectory of the moving object within the monitored area, the method may further include the steps of:

step one: predicting the next gateway in the monitoring area to which the moving target moves based on the video road network and the action track of the moving target;

Step two: and controlling the image acquisition equipment corresponding to the next access opening to be in an open state so as to monitor the moving target.

For ease of description, the two steps described above are combined.

And after constructing a video road network according to the access information of each image acquisition device in the monitoring area in the visible range and the access information of different image acquisition devices in the monitoring area, acquiring the action track of the moving target in the monitoring area based on the video road network. The obtained action track is generated by the moving target, so that the moving target can be monitored more accurately, further actions of the moving target can be predicted based on the video road network and the action track of the moving target, and the moving target is predicted to move to the next entrance and exit in the monitored area.

If the action track of the moving object is a→c→i, as can be seen from the video road network and the action track shown in fig. 4, there are G, H, J and K ports communicating with the port I, so that it can be predicted that the moving object will move to the next port in the monitoring area as follows: G. h, J or K.

After the next access opening in the monitoring area to which the moving target is to be moved is predicted, the image acquisition equipment corresponding to the next access opening can be controlled to be in an on state, so that the moving target can be better monitored.

and determining the current position of the moving target according to the obtained reachable time length between the entrances and exits in the video road network and the action track of the moving target.

In the embodiment of the application, the reachable duration between the entrances and exits in the video road network can be obtained in advance, for example, the reachable duration can be obtained by manually measuring or analyzing the image acquired by the image acquisition equipment, the duration that the reference target reaches one entrance and another entrance and exit is obtained, and then the reachable duration is determined by a set algorithm. If the time length of a plurality of reference targets reaching from one access opening to the other access opening can be obtained, the average value or the median value is taken as the reachable time length of the two access openings.

After the action track of the moving object in the monitoring area is obtained, the passing time of the moving object at the last access point of the action track can be obtained, the length of the passing time from the current time is the length of continuous movement of the moving object after the moving object passes through the last access point of the action track, and the continuous movement distance of the moving object can be obtained according to the ratio of the continuous movement length to the last access point of the action track and the reachable length of the access point, so that the current position of the moving object is determined. This allows for efficient monitoring of moving objects. The current time refers to a time when determination of the position of the moving object is currently required.

Corresponding to the above method embodiments, the embodiments of the present application further provide an action track obtaining device of a moving object, where the action track obtaining device of the moving object described below and the action track obtaining method of the moving object described above may be referred to correspondingly to each other.

Referring to fig. 8, the apparatus may include the following modules:

an entrance information determining module 810 for determining entrance information in a visible range of each image capturing device in the monitoring area;

a accessibility information determining module 820 for determining accessibility information between the entrances and exits of different image capturing devices in the monitored area;

the video road network construction module 830 is configured to construct a video road network according to the access information and the access information;

the action track obtaining module 840 is configured to obtain an action track of the moving object in the monitoring area based on the video road network.

By using the device provided by the embodiment of the application, the access information of each image acquisition device in the monitoring area in the visible range is determined, the access information between the access of different image acquisition devices in the monitoring area is determined, the video road network is constructed according to the access information and the access information, and then the action track of the moving target in the monitoring area is obtained based on the video road network. The video road network is constructed according to the access information in the visible range of each image acquisition device and the access information between the access of different image acquisition devices, so that the access condition of each access of each image acquisition device in the monitoring area can be better reflected, and the more accurate action track of the moving target in the monitoring area can be obtained based on the video road network, thereby effectively monitoring, tracking and the like the moving target.

In a specific embodiment of the present application, the gateway information determining module 810 is configured to:

obtaining a movement track of each first reference target in the visible range of the current image acquisition equipment based on the moving position of each first reference target;

In one embodiment of the present application, the access information determining module 820 is configured to:

obtaining entrance and exit passing data of a plurality of second reference targets in a monitoring area in a set second time period;

for each second reference target, determining the time interval of the current second reference target sequentially passing through two entrances and exits belonging to different image acquisition devices according to the entrance and exit passing data of the current second reference target;

if the time interval is less than the set interval threshold, the number of times reachable between the corresponding two gates is increased.

if the average reachable times between the current two entrances is larger than the preset times threshold, determining that the current two entrances are accessed.

In a specific embodiment of the present application, the method further includes a first video road network updating module, configured to:

After the video road network is constructed, aiming at any two entrances and exits belonging to different image acquisition equipment, if the average reachable times between the current two entrances and exits are detected to have a falling trend and fall to be less than or equal to a time threshold, deleting the reachable channels between the current two entrances and exits in the video road network.

In a specific embodiment of the present application, the method further includes a second update module for a video road network, where the second update module is configured to:

after the video road network is constructed, repeatedly executing the steps of determining the access information in the visual range of each image acquisition device in the monitoring area and determining the access information between the access of different image acquisition devices in the monitoring area under the condition that the set updating condition is reached, and obtaining new access information and access information;

based on the new access information and the access information, the video road network is updated.

In one embodiment of the present application, the action trajectory obtaining module 840 is configured to:

obtaining an initial track of a moving target in a monitoring area;

determining missing passing points according to the video road network, wherein the missing passing points are passing points which are not included in the initial track and pass by the moving target;

And adding missing passing points in the initial track to obtain the action track of the moving target in the monitoring area.

In a specific embodiment of the present application, the method further includes an anomaly determination module, configured to:

after an initial track of a moving target in a monitoring area is obtained, determining abnormal passing points in passing points contained in the initial track according to a video road network;

and outputting prompt information aiming at the abnormal passing points.

In a specific embodiment of the present application, the method further includes a track prediction module, configured to:

after obtaining the action track of the moving object in the monitoring area, predicting the next gateway in the monitoring area to which the moving object will move based on the video road network and the action track of the moving object;

In a specific embodiment of the present application, the method further includes a current location determining module, configured to:

after the action track of the moving object in the monitoring area is obtained, the current position of the moving object is determined according to the reachable time length between the entrances and exits in the video road network and the action track of the moving object, which are obtained in advance.

Corresponding to the above method embodiment, the embodiment of the present application further provides an action track obtaining device of a moving target, including:

a memory for storing a computer program;

and a processor for implementing the steps of the action track obtaining method of the moving target when executing the computer program.

As shown in fig. 9, to schematically illustrate the composition structure of the action trajectory acquisition device of the moving object, the action trajectory acquisition device of the moving object may include: a processor 10, a memory 11, a communication interface 12 and a communication bus 13. The processor 10, the memory 11 and the communication interface 12 all complete communication with each other through a communication bus 13.

In the present embodiment, the processor 10 may be a central processing unit (Central Processing Unit, CPU), an asic, a dsp, a field programmable gate array, or other programmable logic device, etc.

The processor 10 may call a program stored in the memory 11, and in particular, the processor 10 may perform operations in an embodiment of an action trajectory obtaining method of a moving object.

The memory 11 is used for storing one or more programs, and the programs may include program codes, where the program codes include computer operation instructions, and in this embodiment, at least the programs for implementing the following functions are stored in the memory 11:

In one possible implementation, the memory 11 may include a storage program area and a storage data area, where the storage program area may store an operating system, and application programs required for at least one function (such as a communication interaction function, an information processing function), and the like; the storage data area may store data created during use, such as doorway information data, access information data, and the like.

In addition, the memory 11 may include high-speed random access memory, and may also include non-volatile memory, such as at least one magnetic disk storage device or other volatile solid-state storage device.

The communication interface 12 may be an interface of a communication module for interfacing with other devices or systems.

Of course, it should be noted that the configuration shown in fig. 9 does not constitute a limitation of the action trajectory acquisition device of the moving object in the embodiment of the present application, and the action trajectory acquisition device of the moving object may include more or less components than those shown in fig. 9 or may combine some components in practical applications.

Corresponding to the above method embodiments, the present application further provides a computer readable storage medium, on which a computer program is stored, which when executed by a processor, implements the steps of the action trajectory acquisition method of the moving object.

In this specification, each embodiment is described in a progressive manner, and each embodiment is mainly described in a different point from other embodiments, so that the same or similar parts between the embodiments are referred to each other.

Those of skill would further appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, computer software, or combinations of both, and that the various illustrative elements and steps are described above generally in terms of functionality in order to clearly illustrate the interchangeability of hardware and software. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present application.

The steps of a method or algorithm described in connection with the embodiments disclosed herein may be embodied directly in hardware, in a software module executed by a processor, or in a combination of the two. The software modules may be disposed in Random Access Memory (RAM), memory, read Only Memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art.

Specific examples are used herein to illustrate the principles and embodiments of the present application, and the description of the above examples is only for aiding in understanding the technical solution of the present application and its core ideas. It should be noted that it would be obvious to those skilled in the art that various improvements and modifications can be made to the present application without departing from the principles of the present application, and such improvements and modifications fall within the scope of the claims of the present application.

Claims

1. A method for obtaining a trajectory of a moving object, comprising:

acquiring a movement track of a moving target in the monitoring area based on the video road network;

the determining the access information between the entrances and exits of different image acquisition devices in the monitoring area comprises the following steps:

determining whether access is achieved between the entrances and exits of different image acquisition devices according to the entrance and exit passing data;

determining whether access is available between the entrances and exits of different image acquisition devices according to the entrance and exit passing data comprises the following steps:

for any two entrances and exits belonging to different image acquisition equipment, if the average reachable times between the current two entrances and exits is larger than a preset times threshold value, determining that the current two entrances and exits are accessed, wherein the average reachable times are calculated by day or by week.

2. The method according to claim 1, wherein determining the entrance and exit information in the visible range of each image capturing device in the monitored area, comprises:

3. The method according to claim 1, wherein the determining the number of times of accessibility between the entrances and exits belonging to different image capturing apparatuses based on the arranged entrance and exit passage data, comprises:

4. The method for obtaining an action trajectory of a moving object according to claim 1, further comprising, after said constructing a video road network:

5. The method for obtaining an action trajectory of a moving object according to claim 1, further comprising, after said constructing a video road network:

6. The method for obtaining an action trajectory of a moving object according to claim 1, wherein the obtaining an action trajectory of a moving object in the monitoring area based on the video road network comprises:

obtaining an initial track of a moving target in the monitoring area;

7. The action trajectory acquisition method of a moving object according to claim 6, further comprising, after the initial trajectory of the moving object within the monitored area is acquired:

and outputting prompt information aiming at the abnormal passing points.

8. The action trajectory acquisition method of a moving object according to claim 1, characterized by further comprising, after the acquisition of the action trajectory of the moving object within the monitoring area:

9. The action trajectory acquisition method of a moving object according to any one of claims 1 to 8, characterized by further comprising, after the acquisition of the action trajectory of a moving object within the monitoring area:

10. An action trajectory acquisition device of a moving object, comprising:

The action track obtaining module is used for obtaining the action track of the moving target in the monitoring area based on the video road network;

the access information determining module is specifically configured to obtain access passage data of a plurality of second reference targets in the monitoring area in a set second time period;

the access information determining module is specifically configured to arrange the access passing data of each second reference target according to the passing time sequence;

11. An action trajectory acquisition device of a moving object, characterized by comprising:

a memory for storing a computer program;

a processor for implementing the steps of the action trajectory acquisition method of a moving object according to any one of claims 1 to 9 when executing the computer program.

12. A computer-readable storage medium, characterized in that the computer-readable storage medium has stored thereon a computer program which, when executed by a processor, realizes the steps of the action trajectory acquisition method of a moving object according to any one of claims 1 to 9.