CN114898307A

CN114898307A - Object tracking method and device, electronic equipment and storage medium

Info

Publication number: CN114898307A
Application number: CN202210807780.5A
Authority: CN
Inventors: 朱益铭; 唐邦杰; 潘华东; 殷俊
Original assignee: Zhejiang Dahua Technology Co Ltd
Current assignee: Zhejiang Dahua Technology Co Ltd
Priority date: 2022-07-11
Filing date: 2022-07-11
Publication date: 2022-08-12
Anticipated expiration: 2042-07-11
Also published as: CN114898307B

Abstract

The application relates to the technical field of computers, in particular to an object tracking method, an object tracking device, electronic equipment and a storage medium, which are used for more accurately realizing object tracking when targets are worn similarly in an object tracking scene by adopting a plurality of camera equipment. The method comprises the following steps: acquiring the appearance characteristics of an object to be matched and the appearance characteristics of a plurality of matched objects; screening at least two candidate objects from the plurality of matched objects based on the feature similarity between the appearance features of the object to be matched and each matched object; screening a target candidate object from the at least two candidate objects based on the communication relation between the at least two candidate objects and the camera equipment corresponding to the object to be matched; and determining a tracking identification result of the object to be matched based on the track similarity between each target candidate object and the object to be matched. According to the method and the device, the objects are comprehensively identified according to the feature similarity, the topological network and the track similarity, and different objects with similar appearances can be effectively distinguished.

Description

Object tracking method and device, electronic equipment and storage medium

Technical Field

The present application relates to the field of computer technologies, and in particular, to an object tracking method and apparatus, an electronic device, and a storage medium.

Background

With the development of science and technology and the improvement of living standard, people's requirement to security protection system is also promoting constantly, and more supervisory equipment is installed in each region. The intelligent monitoring system can automatically identify the target and save a great deal of manpower besides completing recording and playing the current area situation, so that the intelligent monitoring system is very important. The multi-object tracking system of the cross-camera equipment is an important research content in the field of intelligent monitoring systems, and aims to realize continuous tracking of moving targets under different camera equipment.

Taking cross-camera tracking as an example, in the related art, a cross-camera tracking system often implements cross-camera object tracking based on target expressions (such as appearance features of color, geometry, texture, and the like). However, when the scheme is applied to the situation that similar targets are worn (for example, certain factories require workers to wear uniform work clothes), the difference between different individuals cannot be identified and distinguished better by means of the appearance characteristics of the human body alone, and the cross-camera object tracking effect is influenced.

In summary, in the object tracking scene using a plurality of image capturing apparatuses, object tracking cannot be more accurately achieved when the targets are worn similarly.

Disclosure of Invention

The application provides an object tracking method, an object tracking device, electronic equipment and a storage medium, so as to at least improve the accuracy of tracking a target object which is similar to a target object.

An object tracking method provided by an embodiment of the present application includes:

the method comprises the steps of obtaining appearance characteristics of an object to be matched and appearance characteristics of a plurality of matched objects; the object to be matched is an object in a current video picture acquired by first camera equipment, and the matched object is an object in a historical video picture acquired by second camera equipment;

screening at least two candidate objects from the plurality of matched objects based on the feature similarity between the appearance features of the object to be matched and each matched object;

screening a target candidate object from at least two candidate objects based on the communication relation between the at least two candidate objects and the camera equipment corresponding to the object to be matched;

and determining a tracking identification result of the object to be matched based on the track similarity between each target candidate object and the object to be matched.

An embodiment of the present application provides an object tracking apparatus, including:

the characteristic acquisition unit is used for acquiring the appearance characteristics of the object to be matched and the appearance characteristics of the matched objects; the object to be matched is an object in a current video picture acquired by first camera equipment, and the matched object is an object in a historical video picture acquired by second camera equipment;

the first filtering unit is used for screening out at least two candidate objects from the plurality of matched objects based on the feature similarity between the appearance features of the object to be matched and each matched object;

the second filtering unit is used for screening out a target candidate object from the at least two candidate objects based on the communication relation between the at least two candidate objects and the camera equipment corresponding to the object to be matched;

and the first matching unit is used for determining the tracking identification result of the object to be matched based on the track similarity between each target candidate object and the object to be matched.

In some optional embodiments, the second filter unit is specifically configured to:

respectively determining the communication relation between the first camera equipment and each second camera equipment according to the camera equipment topology network;

determining a candidate object corresponding to a second image pickup apparatus which is communicated with the first image pickup apparatus, from the at least two candidate objects, as the target candidate object based on the determined respective communication relationships;

the camera equipment topological network represents the connection relation among camera equipment in a specified area; each of the image pickup apparatuses within the designated area includes the first image pickup apparatus and the second image pickup apparatus.

In some optional embodiments, the first filter unit is specifically configured to:

and based on the feature similarity between the object to be matched and each matched object, taking at least two matched objects with the feature similarity not less than a feature similarity threshold as the candidate objects.

In some optional embodiments, the apparatus further comprises:

the second matching unit is used for marking the object to be matched as a new object if the feature similarity of all matched objects is smaller than the feature similarity threshold;

and if the feature similarity of only one matched object is not smaller than the feature similarity threshold, taking the matched object as the tracking identification result of the object to be matched.

In some optional embodiments, the first matching unit is specifically configured to:

determining the track similarity between the object to be matched and each target candidate object according to the overlapping information of the track of the object to be matched and the track of each target candidate object in time;

taking the corresponding target candidate object with the track similarity smaller than the track similarity threshold as a target object;

and determining the tracking identification result of the object to be matched from each target object.

for each target candidate, the following operations are respectively performed:

if the track of the object to be matched and the track of one target candidate object have overlapped tracks in time, taking a first similarity associated with the overlapped tracks as the track similarity;

if the track of the object to be matched and the track of one target candidate object do not have overlapping tracks in time, taking a second similarity between a reference object corresponding to the target candidate object and the track of the object to be matched as the track similarity; the reference objects are: and other objects with the third similarity between the tracks of the target candidate object and the other objects meeting the similarity condition.

In some optional embodiments, the other objects are: except the object to be matched, an object in a current video picture acquired by any camera equipment in a designated area;

the similarity condition is as follows: the sorted results of the third similarity are in the specified order.

if only one target object exists, taking the target object as a tracking identification result of the object to be matched;

and if a plurality of target objects exist, taking the sequencing result of the feature similarity of the target objects to be matched and the target objects in the specified sequence as the tracking identification result of the target objects to be matched.

In some optional embodiments, the first matching unit is further configured to:

and if the object with the track similarity smaller than the track similarity threshold value does not exist, marking the object to be matched as a new object.

An electronic device provided in an embodiment of the present application includes a processor and a memory, where the memory stores a computer program, and when the computer program is executed by the processor, the processor is caused to execute any one of the steps of the object tracking method.

An embodiment of the present application provides a computer-readable storage medium, which includes a computer program, when the computer program runs on an electronic device, the computer program is configured to enable the electronic device to execute any one of the steps of the object tracking method described above.

An embodiment of the present application provides a computer program product, which includes a computer program, the computer program being stored in a computer-readable storage medium; when the processor of the electronic device reads the computer program from the computer-readable storage medium, the processor executes the computer program, so that the electronic device performs the steps of any one of the object tracking methods described above.

The beneficial effect of this application is as follows:

the embodiment of the application provides an object tracking method and device, electronic equipment and a storage medium. According to the method, the method for further tracking and identifying the object according to the topological network of the camera equipment and the motion track of the object is added on the basis that tracking and identifying are realized according to the appearance characteristics in the related technology, and the method can further screen the matched object of the tracked object according to the communication relation before the track similarity between the objects is calculated, wherein the communication relation of the camera equipment can reflect the position relation between the acquisition regions corresponding to the camera equipment, and the position of the acquisition regions can influence the judgment of the track of the object, so that the matched object with low track similarity calculation significance can be effectively filtered by combining the communication relation. Compared with a mode of identifying according to appearance characteristics, the method can solve the problem that different individuals with similar characteristics cannot be effectively identified according to the appearance in some special scenes. In addition, for the method for identifying and tracking the object only through the map track, the method is matched on the basis of the appearance characteristics, and the appearance characteristics are easy to obtain and have strong stability when the target posture is changed, so that the method is higher in robustness.

Additional features and advantages of the application will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by the practice of the application. The objectives and other advantages of the application may be realized and attained by the structure particularly pointed out in the written description and claims hereof as well as the appended drawings.

Drawings

The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this application, illustrate embodiment(s) of the application and together with the description serve to explain the application and not to limit the application. In the drawings:

fig. 1 is a schematic view of an application scenario of an object tracking method according to an embodiment of the present application;

fig. 2 is an overall flowchart of an object tracking method according to an embodiment of the present application;

fig. 3 is a schematic diagram of a display screen of a terminal according to an embodiment of the present disclosure;

fig. 4 is a flowchart of screening recognition objects based on feature similarity according to an embodiment of the present disclosure;

fig. 5 is a flowchart of screening an identification object based on a camera device topology network according to an embodiment of the present disclosure;

fig. 6 is a schematic diagram of an object track provided in an embodiment of the present application;

fig. 7 is a flowchart for determining an object to be matched based on a trajectory similarity according to an embodiment of the present disclosure;

fig. 8 is a schematic structural diagram of an object tracking apparatus according to an embodiment of the present disclosure;

fig. 9 is a schematic diagram of a hardware component structure of an electronic device according to an embodiment of the present application;

fig. 10 is a schematic hardware component structure diagram of another electronic device according to an embodiment of the present application.

Detailed Description

In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are some embodiments, but not all embodiments, of the technical solutions of the present application. All other embodiments obtained by a person skilled in the art without any inventive step based on the embodiments described in the present application are within the scope of the protection of the present application.

Some concepts related to the embodiments of the present application are described below.

Object: the target is a traceable target in a video picture, and can be a movable target such as a person, an animal, a driving tool and the like; in the embodiment of the application, the method is divided into two categories, namely an object to be matched and an object already matched according to the time when the object appears in a video picture and whether an identification result exists, wherein the object to be matched refers to: and the first camera device acquires a certain object in the video picture at the current moment, wherein the object is in a state to be identified. Matched objects refer to: the second camera device captures an object in the video picture before the current time, that is, at the historical time, the object having the recognition result. The first image pickup apparatus and the second image pickup apparatus may be the same or different in the embodiment of the present application.

Feature similarity: the similarity between appearance features of the object to be matched and the matched object is referred to, in the embodiment of the present application, the appearance features represent appearance characteristics of the object, such as color, geometric pattern, and the like, and on this basis, the feature similarity is the similarity between appearance color and geometric pattern of the object to be matched and the matched object.

The camera equipment topology network: the camera device topology network refers to a connectivity relationship between different camera devices. The communication relation represents the position relation between the shooting areas of the image pickup devices, and if the communication relation between the two image pickup devices is communication, the position relation of the shooting areas of the two image pickup devices is intersected or closely adjacent. For example, the image pickup apparatus A _Ⅰ Only with the image pickup apparatus a _Ⅱ Image pickup apparatus A _Ⅲ There is a connected relation, i.e. when a certain object is moving from the image pickup apparatus a _Ⅰ After disappearance, if the object appears again, it is certain that it is in the image pickup apparatus a _Ⅰ Or A _Ⅱ Or A _Ⅲ In (because there may be an overlap of the fields of view of the two image capturing apparatuses, there is a possibility that an object appears in the screens of the two image capturing apparatuses at the same time, and the image capturing apparatuses basically achieve no dead-angle coverage within the specified area). And if the communication relation between the two camera devices is not communicated, the position relation of the shooting areas of the two camera devices is separated.

Candidate and target candidate: the candidate object refers to the matched object obtained by screening the matched object through the characteristic similarity; the target candidate object is obtained by further screening the candidate object based on the camera equipment topological network, and the candidate object corresponding to the second camera equipment communicated with the first camera equipment is the target candidate object.

Reference object: when the track of a target candidate object does not have an overlapped part with the track of an object to be matched, the system calculates the track similarity of the target candidate object and other objects in a current video picture acquired by all the camera equipment except the object to be matched in a specified area, and the other objects meeting the similarity condition are reference objects.

Track similarity: the similarity between the motion trajectories of the objects on the map, which is the similarity between the object to be matched and the target candidate object on the trajectories in the present application, may be a first similarity or a second similarity. The first similarity is the similarity calculated according to the time overlapping part of the object track to be matched and the target candidate object track when the object track to be matched and the target candidate object track have the overlapping part in time; the second similarity is a similarity calculated from the reference object trajectory and the object trajectory to be matched when there is no overlapping portion in time between the object trajectory to be matched and the target candidate object trajectory. Wherein the track overlapping part refers to: on the map, two tracks exist simultaneously in a period of time, which indicates that the two tracks have a time overlapping part, so that the track overlapping part refers to a track section in which the two tracks exist simultaneously in the same period of time, and the overlapping part of the two tracks in time can be referred to as an overlapping track of the two tracks in time.

Target object: and selecting target candidate objects, of which the track similarity with the object to be matched is smaller than a track similarity threshold value, from the target candidate objects subjected to the feature similarity and topological network twice screening.

Third phase similarity: and calculating the similarity according to the track of the target candidate object and the tracks of other objects in the current video picture collected by all the camera equipment except the object to be matched in the designated area, wherein the similarity is a third similarity.

The preferred embodiments of the present application will be described below with reference to the accompanying drawings of the specification, it should be understood that the preferred embodiments described herein are merely for illustrating and explaining the present application, and are not intended to limit the present application, and that the embodiments and features of the embodiments in the present application may be combined with each other without conflict.

Fig. 1 is a schematic view of an application scenario in the embodiment of the present application. The application scene diagram includes a terminal device 110 and a server 120.

In the embodiment of the present application, the terminal device 110 includes, but is not limited to, a mobile phone, a tablet computer, a notebook computer, a desktop computer, a camera, a video camera, an intelligent appliance, a vehicle-mounted terminal, and other devices; the terminal device may have a related client installed thereon, where the client may be software (e.g., security software, video recording software, etc.), or may be a web page, an applet, etc., and the server 120 is a background server corresponding to the software, or the web page, the applet, etc., or a server specially used for object tracking, which is not limited in this application. The server 120 may be an independent physical server, a server cluster or a distributed system formed by a plurality of physical servers, or a cloud server providing basic cloud computing services such as a cloud service, a cloud database, cloud computing, a cloud function, cloud storage, a Network service, cloud communication, a middleware service, a domain name service, a security service, a Content Delivery Network (CDN), a big data and artificial intelligence platform, and the like.

It should be noted that, the method for tracking an object in the embodiments of the present application may be performed by an electronic device, which may be the terminal device 110 or the server 120 as shown in fig. 1, that is, the method may be performed by the terminal device 110 or the server 120 alone, or may be performed by both the terminal device 110 and the server 120.

Taking an example of the server 120 executing alone, for example, in an artificial intelligence scene, the designated area is an office building of a certain company, the area is a closed area (or a semi-closed area), no dead angle coverage of the camera equipment can be realized inside the area, and the staffs of the company dress uniform staffs. In this scenario, after a certain image capturing device captures an object to be matched at the current time, the server 120 obtains the object matched at the historical time (referred to as the matched object for short), and appearance characteristics of the object to be matched and the matched object, performs feature similarity calculation, and performs first screening on all the objects to be matched. After the first screening is completed, the server 120 sets the retained matched object as a candidate object, and checks whether the candidate object is connected with the image pickup device where the object to be matched appears according to the topology network between the image pickup devices, based on which, the server 120 performs a second screening on the candidate object, and sets the retained candidate object as a target candidate object. After the second screening is completed, the server 120 calculates the trajectory similarity between the target candidate object and the object to be matched again, performs the third screening on the target candidate object according to whether the trajectory similarity meets the requirement of the trajectory similarity threshold, and sets the target candidate object meeting the requirement as the target object. Finally, the server 120 finds the recognition result of the object to be matched from the target object and sends the result to the terminal 110.

In addition, the method and the device can also be applied to the traffic field, such as real-time detection and tracking of passing vehicles. The object to be matched is a vehicle, the server 120 extracts the features of the vehicle to be matched and the matched vehicle through the picture shot by the monitoring cameras, the matched vehicle is screened based on the feature similarity, the server 120 sets the reserved matched vehicle as a candidate vehicle, and performs secondary screening on the candidate vehicle based on the topological network among the monitoring cameras, and sets the reserved candidate vehicle as a target candidate vehicle. After the second screening is completed, the server 120 calculates the trajectory similarity between the target candidate vehicle and the vehicle to be matched again, performs the third screening on the target candidate vehicle based on the trajectory similarity, sets the target candidate vehicle meeting the requirements as the target vehicle, sends the result to the terminal 110, and finally realizes the identification and tracking of the vehicle.

In an alternative embodiment, terminal device 110 and server 120 may communicate via a communication network.

In an alternative embodiment, the communication network is a wired network or a wireless network.

It should be noted that fig. 1 is only an example, and the number of the terminal devices and the servers is not limited in practice, and is not specifically limited in the embodiment of the present application.

In the embodiment of the present application, when the number of the servers is multiple, the multiple servers may be grouped into a blockchain, and the servers are nodes on the blockchain.

In addition, the embodiment of the application can be applied to various scenes, including but not limited to cloud technology, artificial intelligence, intelligent traffic, driving assistance and the like.

The object tracking method provided by the exemplary embodiment of the present application is described below with reference to the accompanying drawings in conjunction with the application scenarios described above, it should be noted that the application scenarios described above are only shown for the convenience of understanding the spirit and principles of the present application, and the embodiments of the present application are not limited in this respect.

Referring to fig. 2, which is a flowchart illustrating an implementation of an object tracking method according to an embodiment of the present application, the method is implemented by a server alone, and the following steps S201 to S204 are specifically implemented:

s201: and acquiring the appearance characteristics of the object to be matched and the appearance characteristics of the matched objects.

The object to be matched is an object in a current video picture acquired by a first camera device, the matched object is an object in a historical video picture acquired by a second camera device, and the first camera device and the second camera device are the same camera device or different camera devices.

In the embodiment of the present application, the image capturing device refers to a device capable of recording and capturing a video or an image in a designated area, such as a video camera, a monitoring camera, a camera, and the like, which is not specifically limited herein, and the camera is taken as an example to be described in detail below.

The object refers to a trackable object in a video picture, and may be a movable object such as a person, an animal, a driving tool, etc., and in the following, for example, the object refers to a human object, and the designated area is an office building of a company.

In the embodiment of the application, the object refers to a traceable target, and each traceable target has certain appearance characteristics which specifically describe appearance characteristics of the object. By way of example, appearance characteristics include, but are not limited to, garment color, type, pattern, geometric pattern on the garment, character style, and the like.

For example, as shown in fig. 3, a terminal display screen is shown, the current time is 8 o' clock today in the morning, and the object to be matched is Y ₁ The historical time is any time before 8 am today, and as shown in FIG. 3, there is Y for the matched object corresponding to 7:59 am today ₂ Y is the matched object corresponding to the moment of 7:57 in the morning of this day ₃ … …, the matched object corresponding to this morning at 7:58 am has Y ₁₀ 。

Before the appearance characteristics of the object to be matched and the matched object are obtained, the server also obtains the identifier of the first image pickup device with the object to be matched and the global unique identifications of the plurality of matched objects.

Wherein each camera device possesses a unique identifier for determining the position of the shooting area of the camera device, and the identifier is a combination of letters and roman numerals. For example, the identifier of the first image pickup apparatus listed above is a _Ⅰ 。

The global unique identifier means that the same object has one and only one global unique identifier no matter the same object appears in any image pickup device picture of the designated area. The global means that a plurality of cameras are under a region for a period of time.

In the embodiment of the application, the designated area is a closed or semi-closed area, and a designated access opening is formed instead of an open scene, such as an office building instead of a square. In the embodiment of the application, the plurality of camera devices in the designated area can realize no dead angle coverage in the whole range.

Specifically, the server matches the object to be matched with the matched object, and after the matching is successful, a globally unique identifier (for example, a number, a string of characters, or the like) is assigned to the matched object.

Matched object Y as enumerated above ₂ Has a global unique identifier of 2, and has matched object Y ₃ Is 3, … …, matched object Y ₁₀ Is 10.

In the embodiment of the application, after the image pickup device identifier and the globally unique identifier of the matched object are acquired, the server may further acquire appearance characteristics of the object to be matched and all matched objects.

The appearance features can be extracted by utilizing a Person Re-identification (ReID) neural network, and the appearance features of the object are expressed by vectors formed by 1024 numbers.

It should be noted that the above-listed methods for representing the appearance characteristics are only provided as a possible embodiment, and are not specifically limited herein.

After the appearance characteristics of the object to be matched and the plurality of matched objects are obtained, the server can execute the following processes:

step S202: and screening at least two candidate objects from the plurality of matched objects based on the feature similarity between the appearance features of the object to be matched and each matched object.

In the embodiment of the application, the server calculates the feature similarity between the object to be matched and each matched object, counts the matched objects with the feature similarity more than or equal to the feature similarity threshold value, obtains the matched objects after filtering, sets the matched objects as candidate objects, and stores the candidate objects in the candidate object set Y _filter1 In (3), further obtaining the number N of the candidate objects _filter1 。

If the server can screen at least two matched objects with the feature similarity not less than the feature similarity threshold based on the feature similarity between the object to be matched and the matched objects, the screened matched objects can be used as candidate objects.

That is, if N _filter1 If the number of the candidate objects is more than 1, the server filters the candidate objects again through the topological network of the camera equipment to obtain target candidate objects, and counts the number N of the target candidate objects _filter2 。

The number of matched objects satisfying the feature similarity not less than the feature similarity threshold may be 1 or 0 in addition to the cases enumerated in step S202. An alternative embodiment is:

if the feature similarity of all the matched objects is smaller than the feature similarity threshold, it is indicated that no matched object capable of being matched with the appearance features of the object to be matched exists, and the object to be matched is firstly appeared under the camera equipment in the specified area, so that the object to be matched can be marked as a new object;

that is, if N _filter1 And if the value is 0, a new global unique identifier is allocated to the object to be matched.

If the feature similarity of only one matched object is not smaller than the feature similarity threshold, it indicates that only the appearance feature of one matched object can be matched with the object to be matched, so that the corresponding matched object can be directly used as the tracking identification result of the object to be matched.

That is, if N _filter1 If 1, the global unique identifier of the corresponding matched object that meets the threshold requirement is assigned to the object to be matched.

For example, as shown in fig. 4, it is a flowchart of a method for a server to filter matched objects according to feature similarities of an object to be matched and the matched object in this embodiment of the application, and includes the following steps 401 and 406:

step 401: and calculating the feature similarity of the object to be matched and each matched object.

Step 402: and according to a preset feature similarity threshold, deleting the corresponding matched objects with the feature similarity smaller than the feature similarity threshold, and taking at least two matched objects with the feature similarity not smaller than the feature similarity threshold as candidate objects.

In addition, after the candidate object screening is finished, the server further executes the following steps:

step 403: and reading the number of the candidate objects.

If the number of the candidate objects is more than two, the server executes step 404; if the number of the candidate objects is 1, that is, the server can only screen out a matched object whose feature similarity is not less than the feature similarity threshold, the server executes step 405; if the number of the candidate objects is 0, that is, the feature similarity of all the matched objects is smaller than the feature similarity threshold, the server executes step 406.

Step 404: continuing the subsequent screening (see steps S203 and S204);

step 405: and directly distributing the global unique identification of the candidate object to the object to be matched, and finishing the tracking and identifying process.

Step 406: and distributing a new global unique identifier to the object to be matched, and ending the tracking identification process.

Taking a specific object as an example, assume that an object Y to be matched ₁ A matched object Y appearing in the picture acquired by the first camera ₂ -Y ₁₀ The image is acquired by the second camera equipment; object Y to be matched acquired by server ₁ Has an appearance characteristic of F ₁ Matched object Y ₂ Has an appearance characteristic of F ₂ Matched object Y ₃ Has an appearance characteristic of F ₃ Matched object Y ₄ Has an appearance characteristic of F ₄ Matched object Y ₅ Has an appearance characteristic of F ₅ … …, matched object Y ₁₀ Has an appearance characteristic of F ₁₀ 。

Following with the object Y to be matched ₁ And matched object Y ₂ Calculating the similarity of the characteristics as an example, and the server calculates the similarity according to the appearance characteristics F ₁ And F _2， Calculating an object Y to be matched ₁ With matched object Y ₂ Characteristic similarity of (S) ₁₂ 。

In the embodiment of the present application, the cosine similarity is used for calculating the feature similarity, and the specific formula is as follows:

wherein, the value range of the feature similarity is 0-1, and the larger the value is, the more similar the features are.

Similarly, the server can calculate the object Y to be matched by referring to the formula ₁ The feature similarity S of each of the matched objects ₁₃ ，S ₁₄ ，S ₁₅ ……S ₁₁₀ 。

Assume that the feature similarity threshold is set to 0.6, S ₁₂ To S ₁₁₀ If all the similarity thresholds are less than 0.6, the server allocates a new global unique identifier to the object to be matched and ends the tracking identification.

In another case, it is still assumed that the feature similarity threshold is 0.6, S ₁₂ To S ₁₁₀ In, only S ₁₈ Is not less than 0.6, the server directly matches the matched object Y ₈ The global unique identifier is distributed to the object to be matched and the tracking identification is finished.

In the third case, it is still assumed that the feature similarity threshold is 0.6, S ₁₂ To S ₁₁₀ In, has S ₁₃ 、S ₁₅ 、S ₁₈ 、S ₁₉ Is not less than 0.6, the server will match the object Y ₃ 、Y ₅ 、Y ₈ 、Y ₉ Set as candidates. And the server continues to perform subsequent screening because the number of the candidate objects is more than 1.

It should be noted that, for the above-listed calculation manner of feature similarity, the present application only provides a feasible embodiment, and the specific method is not limited herein.

In the embodiment of the present application, in order to further identify the tracking result of the object to be matched in the candidate objects, the server may perform the following processes:

step S203: and screening out target candidate objects from the at least two candidate objects based on the communication relation between the at least two candidate objects and the image pickup equipment corresponding to the object to be matched.

Wherein, the connectivity relationship between the camera devices in the designated area can be obtained by the camera device topological network, and the connectivity relationship represents the position relationship between the camera device shooting areas. If the two camera devices are communicated, the position relations of the shooting areas of the two camera devices are intersected or closely adjacent; and if the communication relation between the two camera devices is not communicated, the position relation of the shooting areas of the two camera devices is separated.

The camera device topology network is pre-labeled, such as manually labeled.

In the embodiment of the application, the server respectively determines the communication relation between the first camera equipment and each second camera equipment according to the camera equipment topology network; and then based on the determined communication relations, deleting the candidate object corresponding to the second camera shooting equipment which is not communicated with the first camera shooting equipment, and screening out the target candidate object. I.e. to the candidate set Y _filter1 Filtering to obtain a target candidate object set Y _filter2 。

For example, as shown in fig. 5, it is a flowchart of a method for a server to further filter candidate objects according to a camera device topology network in an embodiment of the present application, and the method includes the following steps 501 and 507:

step 501: identifiers of the second image pickup apparatuses corresponding to the respective candidate objects are acquired.

Step 502: and judging whether the first camera equipment is communicated with each second camera equipment or not according to the camera equipment topology network marked in advance. If so, go to step 503, otherwise go to step 504.

Step 503: the candidate object corresponding to this second image pickup apparatus is retained and set as a target candidate object.

Step 504: and deleting the candidate object corresponding to the second image pickup device.

Step 505: and acquiring the number of the target candidate objects, if the number is greater than 1, executing the step 506 by the server, and if the number is equal to 1, executing the step 507.

Step 506: and continuing the subsequent screening.

Step 507: and directly distributing the global unique identifier of the target candidate object to the object to be matched, and finishing the tracking and identifying process.

Taking the specific object as an example, the matched object Y after feature similarity screening is described above ₃ 、Y ₅ 、Y ₈ 、Y ₉ Candidate Y obtained for filtering ₃ 、Y ₅ 、Y ₈ 、Y ₉ The server acquires identifiers of second camera devices of the four candidate objects, and judges each second camera device and the object Y to be matched according to the camera device topology network ₁ Whether the first camera devices are communicated with each other or not. In the embodiment of the present application, the image pickup apparatus topology network indicates a connectivity relationship between different image pickup apparatuses. For example, the image pickup apparatus A _Ⅰ Only with the image pickup apparatus a _Ⅱ Image pickup apparatus A _Ⅲ There is a connected relation, i.e. when a certain object is from the image pickup apparatus a _Ⅰ After disappearance, if the object appears again, it is certain that it is in the image pickup apparatus a _Ⅰ Or A _Ⅱ Or A _Ⅲ In (because there may be an overlap of the fields of view of the two image capturing apparatuses, there is a possibility that an object appears in the screens of the two image capturing apparatuses at the same time, and the image capturing apparatuses basically achieve no blind spot coverage within the specified area).

Suppose candidate Y ₅ 、Y ₈ The second image pickup apparatus appearing in communication with the first image pickup apparatus, candidate object Y ₃ 、Y ₉ If the second image pickup apparatus does not communicate with the first image pickup apparatus, the server deletes the candidate object Y ₃ 、Y ₉ Keeping the candidate Y ₅ 、Y ₈ And set it as a target candidate Y ₅ 、Y ₈ 。

After the screening by the camera device topology network, the server executes step S204 to further identify the tracking result of the object to be matched in the target candidate object:

s204: and the server determines a tracking identification result of the object to be matched based on the track similarity between the target candidate object and the object to be matched.

Specifically, the server determines the track similarity between the object to be matched and each target candidate object according to the overlapping information of the track of the object to be matched and the track of each target candidate object in time; taking the corresponding target candidate object with the track similarity smaller than the track similarity threshold as a target object; and determining a tracking identification result of the object to be matched from each target object.

In the embodiment of the present application, the trajectory of the object refers to a map trajectory that the object leaves in a map of a specified area. The map refers to a map obtained by observing the whole closed area from an overlooking angle; the map trajectory is a series of two-dimensional coordinate points (x, y) with reference to the horizontal and vertical coordinates of the planar map. After an object starts moving, a series of coordinate points (x, y) of the position of the object on a map are mapped under an image pickup device, and the series of coordinate points form the map track of the object.

The overlapping information represents whether two tracks have overlapping parts in time, and if two tracks have overlapping parts in time, that is, two tracks exist at the same time in a certain time period, a track section (time overlapping part) corresponding to the overlapping time is an overlapping track of the two tracks, that is, the overlapping information includes the overlapping track.

The overlap information may also indicate the degree of overlap of the two tracks in time, i.e. the time length of the overlap of the two tracks. For example, if the degree of overlap of two tracks is 0, it means that the two tracks do not exist simultaneously in time, and there is no overlapping portion; if the overlapping degree of the two tracks is greater than 0, it indicates that the two tracks exist at the same time in a certain period of time, and an overlapping part exists, that is, an overlapping track exists, and the like.

In an optional implementation manner, the server first obtains the map tracks of the object to be matched and each target candidate object, and determines whether there is an overlapping portion in time between the tracks of the object to be matched and each target candidate object, where the overlapping portion refers to that there is an overlap in time between the two tracks. The following two cases can be specifically distinguished:

in case of overlapping, the server takes the first similarity associated with the overlapping track as the track similarity between the object to be matched and the corresponding target candidate object. The first similarity associated with the overlapped tracks is the similarity of the tracks of the overlapped parts in the tracks of the object to be matched and the target candidate object.

Continuing with the specific assumptions above, assume target candidate Y ₅ With the object Y to be matched ₁ If the two tracks overlap in time, the server calculates Y ₅ And Y ₁ The first similarity of the tracks of the overlapping part in the tracks is taken as Y ₅ And Y ₁ The trajectory similarity of (1).

And in the second situation, if the overlapped part does not exist, the server acquires a reference object corresponding to one target candidate object, calculates the second similarity of the track of the reference object and the target object to be matched, and takes the second similarity as the track similarity of the target object to be matched and the corresponding target candidate object.

When the track of a target candidate object does not have an overlapped part with the track of an object to be matched, the system calculates the track similarity of the target candidate object and other objects in a current video picture acquired by all the camera equipment except the object to be matched in a specified area, and the other objects meeting the similarity condition are reference objects.

Continuing with the above specific assumptions, referring to FIG. 6, assume target candidate Y ₈ With the object Y to be matched ₁ If the map track does not appear at the same time, that is, the map track does not have an overlapping part, the division object Y is obtained ₁ Besides, all objects in the current video picture collected by all the camera devices in the designated area are taken as other objects Y ₁₁ 、Y ₁₂ 、Y ₁₃ … … (except for the object to be matched Y) ₁ ) Calculating a target candidate Y ₈ Third similarity with other object tracks is obtained, and other objects with the highest third similarity are obtained as target candidate objects Y ₈ Assuming that the reference object is Y ₁₁ 。

Calculating Y ₁₁ And Y ₁ A second similarity of the trajectories and comparing the second similarity with the first similaritySimilarity as target candidate Y ₈ With the object Y to be matched ₁ Track similarity T of ₁₈ 。

After the work is finished, the server sets the target candidate object with the track similarity smaller than the track similarity threshold as the target object to obtain a target object set Y _filter3 And according to the number N of the target objects _filter3 Continuing the next step, three specific cases are distinguished:

case one, if there is only one target object, i.e. N _filter3 And 1, taking one target object as a tracking identification result of the object to be matched.

Case two, if there are multiple target objects, i.e. N _filter3 And if the result is more than 1, the sequencing result of the feature similarity of the object to be matched is positioned in the target object in the specified sequence, and the target object is used as the tracking identification result of the object to be matched.

The designated sequence refers to a certain position in a preset sequencing sequence, for example, each target object is sequenced from large to small according to the feature similarity between the target object and the object to be matched, and the first sequencing position, namely the feature similarity is the highest and is selected as the tracking identification result of the object to be matched.

Case three, if there is no object with the track similarity smaller than the track similarity threshold, that is, N _filter3 And marking the object to be matched as a new object if the number is 0.

Referring to fig. 7, which is a flowchart illustrating determining an object to be matched based on track similarity in the embodiment of the present application, the server performs steps 701-715:

step 701: and acquiring the map tracks of the object to be matched and each target candidate object.

The map track is obtained by using an object detection frame of an object detection network and obtaining the track of the object on the map based on track conversion.

Step 702: and judging whether the track of the object to be matched and the track of each target candidate object have an overlapping part or not. If so, go to step 703, otherwise go to step 705.

Step 703: and calculating the first similarity of the overlapping part of the object to be matched and the target candidate object track.

Step 704: and taking the first similarity as the track similarity between the target candidate object and the object to be matched.

Step 705: and calculating a third similarity of the corresponding target candidate object and other object tracks.

The other objects are objects in the current video picture collected by any one of the camera devices in the designated area except the object to be matched.

Step 706: and acquiring other objects with the highest third similarity as reference objects of the target candidate object.

Step 707: and calculating a second similarity between the reference object and the track of the object to be matched.

Step 708: and taking the second similarity as the track similarity between the target candidate object corresponding to the reference object and the object to be matched.

Step 709: and judging whether the track similarity is smaller than a preset track similarity threshold value. If so, go to step 710, otherwise go to step 711.

Step 710: the target candidate object is set as a target object.

Step 711: the target candidate is deleted.

Step 712: and acquiring the number of the target objects. If the number is greater than 1, step 713 is executed, if the number is equal to 1, step 714 is executed, and if the number is 0, step 715 is executed.

Step 713: and allocating the global unique identification of the target object with the highest feature similarity to the object to be matched.

Step 714: and assigning the global unique identification of the unique target object to the object to be matched.

Step 715: and assigning a new global unique identifier to the object to be matched.

Specifically, continuing to use the above assumption, the server obtains the target candidate object Y filtered by the feature similarity and the camera topology network ₅ 、Y ₈ With the object Y to be matched ₁ And determining whether there is overlapIn part, due to target candidate Y ₅ With the object Y to be matched ₁ Has an overlapping portion of the map track, target candidate Y ₈ With the object Y to be matched ₁ There is no overlapping portion of the map track. Then for the target candidate Y ₅ The server calculates the distance between the server and the object Y to be matched according to the Hausdorff distance ₁ The first similarity of the tracks of the overlapping parts is specifically represented as follows:

wherein A and B respectively represent an object Y to be matched ₁ And target candidate Y ₅ The trajectory of the overlapping portion, h (a, B), is specifically calculated as follows:

it should be noted that the range of the calculation result of the above formula is 0 to positive infinity, and a smaller value represents a higher similarity.

Where hausdorff distance refers to the distance between two subsets in the metric space. If each point of one set is close to a point of the other set, then the two sets are close in terms of hausdorff distance. The hausdorff distance refers to the longest distance an adversary has to select a point in one of the two groups and then has to reach the other group from there. In other words, it is the largest of all distances from a point in one set to the nearest point in the other set.

After the calculation is finished, the first similarity is taken as a target candidate object Y ₅ With the object Y to be matched ₁ The track similarity is recorded as T ₁₅ 。

For target candidate Y ₈ Obtaining the object to be matched Y ₁ Besides, all objects in the current video picture collected by all the image pickup devices in the designated area are used as other objects, and a target candidate object Y is calculated ₈ Third similarity with other object tracks and obtaining a third phaseThe other objects with the highest similarity are used as target candidate objects Y ₈ The reference object of (1).

Calculating a reference object Y ₁₁ And taking the second similarity as a target candidate object Y ₈ With the object Y to be matched ₁ The track similarity is recorded as T ₁₈ 。

Suppose T ₁₅ And T ₁₈ Are all smaller than a preset track similarity threshold value, and the server enables the target candidate object Y ₅ 、Y ₈ Set as a target object Y ₅ 、Y ₈ Since the number of target objects is greater than Y ₁ The server acquires the target object Y again ₅ 、Y ₈ With the object Y to be matched ₁ Characteristic similarity of (S) ₁₅ And S ₁₈ And comparing the magnitudes, assuming S ₁₅ Less than S ₁₈ Then the server will target object Y ₈ Is assigned to the object Y to be matched ₁ And the tracking recognition is finished.

Based on the same inventive concept, the embodiment of the application also provides an object tracking device. As shown in fig. 8, which is a schematic structural diagram of an object tracking apparatus 800, the object tracking apparatus may include:

a feature obtaining unit 801, configured to obtain an appearance feature of an object to be matched and an appearance feature of each of multiple matched objects; the object to be matched is an object in a current video picture acquired by first camera equipment, and the matched object is an object in a historical video picture acquired by second camera equipment;

a first filtering unit 802, configured to filter out at least two candidate objects from the multiple matched objects based on feature similarity between appearance features of the object to be matched and each matched object;

a second filtering unit 803, configured to screen out a target candidate object from the at least two candidate objects based on a communication relationship between the at least two candidate objects and the image capturing apparatus corresponding to the object to be matched;

the first matching unit 804 is configured to determine a tracking identification result of the object to be matched based on the trajectory similarity between each target candidate object and the object to be matched.

In some alternative embodiments, the second filtering unit 803 is specifically configured to:

determining a candidate object corresponding to a second image pickup apparatus which is communicated with the first image pickup apparatus, from among the at least two candidate objects, as a target candidate object based on the determined respective communication relationships;

the camera equipment topological network represents the connection relation among all camera equipment in a specified area; each of the image pickup apparatuses in the designated area includes the first image pickup apparatus and the second image pickup apparatus.

In some optional embodiments, the first filtering unit 802 is specifically configured to:

and based on the feature similarity between the object to be matched and each matched object, taking at least two matched objects with the feature similarity not less than a feature similarity threshold as candidate objects.

In some optional embodiments, the apparatus further comprises:

a second matching unit 805, configured to mark the object to be matched as a new object if the feature similarity of all the matched objects is smaller than the feature similarity threshold;

and if the feature similarity of only one matched object is not smaller than the feature similarity threshold, taking the matched object as a tracking identification result of the object to be matched.

In some optional embodiments, the first matching unit 804 is specifically configured to:

determining the track similarity between the object to be matched and each target candidate object according to the time overlapping information of the track of the object to be matched and the track of each target candidate object;

and determining a tracking identification result of the object to be matched from each target object.

for each target candidate, the following operations are respectively performed:

if the track of the object to be matched and the track of one target candidate object have an overlapping part in time, taking the first similarity of the tracks of the overlapping part as the track similarity;

if the track of the object to be matched and the track of one target candidate object do not have an overlapping part in time, taking a second similarity between a reference object corresponding to one target candidate object and the track of the object to be matched as a track similarity; the reference objects are: and other objects with the third similarity between the tracks of the target candidate object and the tracks of the target candidate object, wherein the third similarity meets the similarity condition.

In some alternative embodiments, the other objects are: except for the object to be matched, the object in the current video picture acquired by any camera equipment in the designated area;

the similarity conditions are as follows: the sorted results of the third similarity are in the specified order.

and if a plurality of target objects exist, the target objects with the sequencing result of the feature similarity of the objects to be matched positioned in the specified sequence are used as the tracking identification result of the objects to be matched.

In some optional embodiments, the first matching unit 804 is further configured to:

Having described the object tracking method and apparatus of the exemplary embodiments of the present application, an electronic device according to another exemplary embodiment of the present application is next described.

As will be appreciated by one skilled in the art, aspects of the present application may be embodied as a system, method or program product. Accordingly, various aspects of the present application may be embodied in the form of: an entirely hardware embodiment, an entirely software embodiment (including firmware, microcode, etc.) or an embodiment combining hardware and software aspects that may all generally be referred to herein as a "circuit," module "or" system.

The electronic equipment is based on the same inventive concept as the method embodiment, and the embodiment of the application also provides the electronic equipment. In one embodiment, the electronic device may be a server, such as server 120 shown in FIG. 1. In this embodiment, the electronic device may be configured as shown in fig. 9, and include a memory 901, a communication module 903, and one or more processors 902.

A memory 901 for storing computer programs executed by the processor 902. The memory 901 may mainly include a program storage area and a data storage area, where the program storage area may store an operating system, a program required for running an instant messaging function, and the like; the storage data area can store various instant messaging information, operation instruction sets and the like.

Memory 901 may be a volatile memory (volatile memory), such as a random-access memory (RAM); the memory 901 may also be a non-volatile memory (non-volatile memory), such as a read-only memory (rom), a flash memory (flash memory), a hard disk (HDD) or a solid-state drive (SSD); or the memory 901 is any other medium that can be used to carry or store a desired computer program in the form of instructions or data structures and that can be accessed by a computer, but is not limited to such. The memory 901 may be a combination of the above memories.

The processor 902 may include one or more Central Processing Units (CPUs), a digital processing unit, and the like. A processor 902 for implementing the above object tracking method when calling the computer program stored in the memory 901.

The communication module 903 is used for communicating with terminal equipment and other servers.

The embodiment of the present application does not limit the specific connection medium among the memory 901, the communication module 903, and the processor 902. In the embodiment of the present application, the memory 901 and the processor 902 are connected through the bus 904 in fig. 9, the bus 904 is depicted by a thick line in fig. 9, and the connection manner between other components is merely illustrative and is not limited. The bus 904 may be divided into an address bus, a data bus, a control bus, and the like. For ease of description, only one thick line is depicted in fig. 9, but only one bus or one type of bus is not depicted.

The memory 901 stores a computer storage medium, in which computer-executable instructions are stored, and the computer-executable instructions are used for implementing the object tracking method according to the embodiment of the present application. The processor 902 is configured to perform the object tracking method described above, as shown in FIG. 2.

In another embodiment, the electronic device may also be other electronic devices, such as the terminal device 110 shown in fig. 10. In this embodiment, the structure of the electronic device may be as shown in fig. 9, including: a communications component 1010, a memory 1020, a display unit 1030, a camera 1040, a sensor 1050, audio circuitry 1060, a bluetooth module 1070, a processor 1080, and the like.

The communication component 1010 is configured to communicate with a server. In some embodiments, a Wireless Fidelity (WiFi) module may be included, the WiFi module being a short-range Wireless transmission technology, through which the electronic device may help the user to transmit and receive information.

Memory 1020 may be used to store software programs and data. Processor 1080 performs various functions of terminal device 110 and data processing by executing software programs or data stored in memory 1020. The memory 1020 may include high speed random access memory and may also include non-volatile memory, such as at least one magnetic disk storage device, flash memory device, or other volatile solid state storage device. The memory 1020 stores an operating system that enables the terminal device 110 to operate. The memory 1020 may store an operating system and various application programs, and may also store computer programs for executing the object tracking method according to the embodiment of the present application.

The display unit 1030 may also be used to display information input by the user or information provided to the user and a Graphical User Interface (GUI) of various menus of the terminal device 110. Specifically, the display unit 1030 may include a display screen 1032 disposed on the front surface of the terminal device 110. The display 1032 may be configured in the form of a liquid crystal display, a light emitting diode, or the like.

The display unit 1030 may also be configured to receive input numeric or character information and generate signal input related to user settings and function control of the terminal device 110, and specifically, the display unit 1030 may include a touch screen 1031 disposed on the front surface of the terminal device 110 and configured to collect touch operations by a user thereon or nearby, such as clicking a button, dragging a scroll box, and the like.

The touch screen 1031 may cover the display screen 1032, or the touch screen 1031 and the display screen 1032 may be integrated to implement the input and output functions of the terminal device 110, and the integrated function may be referred to as a touch display screen for short. In the present application, the display unit 1030 may display the application program and the corresponding operation steps.

The camera 1040 may be used to capture still images, and the user may post the images captured by the camera 1040 through an application. The number of the cameras 1040 may be one or plural. The object generates an optical image through the lens and projects the optical image to the photosensitive element. The photosensitive element may be a Charge Coupled Device (CCD) or a complementary metal-oxide-semiconductor (CMOS) phototransistor. The light sensing elements convert the light signals into electrical signals which are then passed to a processor 1080 for conversion into digital image signals.

The terminal device may further comprise at least one sensor 1050, such as an acceleration sensor 1051, a distance sensor 1052, a fingerprint sensor 1053, a temperature sensor 1054. The terminal device may also be configured with other sensors such as a gyroscope, barometer, hygrometer, thermometer, infrared sensor, light sensor, motion sensor, and the like.

Audio circuitry 1060, speaker 1061, microphone 1062 may provide an audio interface between a user and terminal device 110. The audio circuit 1060 may transmit the electrical signal converted from the received audio data to the speaker 1061, and convert the electrical signal into a sound signal by the speaker 1061 and output the sound signal. Terminal device 110 may also be configured with a volume button for adjusting the volume of the sound signal. On the other hand, the microphone 1062 converts the collected sound signals into electrical signals, which are received by the audio circuit 1060 and converted into audio data, which is then output to the communication module 1010 for transmission to, for example, another terminal device 110, or to the memory 1020 for further processing.

The bluetooth module 1070 is used for exchanging information with other bluetooth devices having a bluetooth module through a bluetooth protocol. For example, the terminal device may establish a bluetooth connection with a wearable electronic device (e.g., a smart watch) having a bluetooth module via the bluetooth module 1070, so as to perform data interaction.

The processor 1080 is a control center of the terminal device, connects various parts of the entire terminal device using various interfaces and lines, and performs various functions of the terminal device and processes data by running or executing software programs stored in the memory 1020 and calling data stored in the memory 1020. In some embodiments, processor 1080 may include one or more processing units; processor 1080 may also integrate an application processor, which primarily handles operating systems, user interfaces, applications, etc., and a baseband processor, which primarily handles wireless communications. It is to be appreciated that the baseband processor described above may not be integrated into processor 1080. In the present application, the processor 1080 may run an operating system, an application program, a user interface display, a touch response, and an object tracking method according to the embodiments of the present application. Further, processor 1080 is coupled to a display unit 1030.

In some possible embodiments, the aspects of the object tracking method provided in the present application may also be implemented in the form of a program product including a computer program for causing an electronic device to perform the steps in the object tracking method according to various exemplary embodiments of the present application described above in this specification when the program product is run on the electronic device, for example, the electronic device may perform the steps as shown in fig. 2.

The program product may employ any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. A readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the foregoing. More specific examples (a non-exhaustive list) of the readable storage medium include: an electrical connection having one or more wires, a portable disk, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.

The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

It will be apparent to those skilled in the art that various changes and modifications may be made in the present application without departing from the spirit and scope of the application. Thus, if such modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is intended to include such modifications and variations as well.

Claims

1. An object tracking method, comprising:

2. The method of claim 1, wherein the screening out target candidate objects from the at least two candidate objects based on a connection relationship between the at least two candidate objects and the image capturing apparatus corresponding to the object to be matched comprises:

3. The method of claim 1, wherein the screening out at least two of the plurality of matched objects as candidate objects based on feature similarity between the object to be matched and the appearance features of the respective matched objects comprises:

4. The method of claim 3, wherein the method further comprises:

if the feature similarity of all matched objects is smaller than the feature similarity threshold, marking the object to be matched as a new object;

5. The method according to any one of claims 1 to 4, wherein the determining the tracking recognition result of the object to be matched based on the track similarity between each target candidate object and the object to be matched comprises:

6. The method according to claim 5, wherein the overlapping information includes overlapping tracks, and the determining the track similarity between the object to be matched and each of the target candidate objects according to the overlapping information of the track of the object to be matched and the respective track of each of the target candidate objects in time includes:

for each target candidate, the following operations are respectively performed:

7. The method of claim 6, wherein the other objects are: except the object to be matched, an object in a current video picture acquired by any camera equipment in a designated area;

8. The method as claimed in claim 5, wherein said determining the tracking identification result of the object to be matched from the respective target objects comprises:

9. The method according to claim 5, wherein the determining the tracking recognition result of the object to be matched from the respective target objects further comprises:

and if the target object with the track similarity smaller than the track similarity threshold does not exist, marking the object to be matched as a new object.

10. An object tracking apparatus, comprising:

11. An electronic device, comprising a processor and a memory, wherein the memory stores a computer program which, when executed by the processor, causes the processor to carry out the steps of the method of any of claims 1 to 9.

12. A computer-readable storage medium, characterized in that it comprises a computer program for causing an electronic device to carry out the steps of the method according to any one of claims 1 to 9, when said computer program is run on said electronic device.

13. A computer program product, comprising a computer program stored in a computer readable storage medium; when a processor of an electronic device reads the computer program from the computer-readable storage medium, the processor executes the computer program, causing the electronic device to perform the steps of the method of any of claims 1-9.