CN113793363A - Target tracking method and related device - Google Patents
Target tracking method and related device Download PDFInfo
- Publication number
- CN113793363A CN113793363A CN202111136528.8A CN202111136528A CN113793363A CN 113793363 A CN113793363 A CN 113793363A CN 202111136528 A CN202111136528 A CN 202111136528A CN 113793363 A CN113793363 A CN 113793363A
- Authority
- CN
- China
- Prior art keywords
- historical
- target object
- image
- target
- tracking
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/20—Analysis of motion
- G06T7/246—Analysis of motion using feature-based methods, e.g. the tracking of corners or segments
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Image Analysis (AREA)
Abstract
In the target tracking method and the related device provided by the application, after the data processing equipment determines the target object from the image to be recognized by using the historical characteristics, the tracking record of the target object is generated, and meanwhile, the historical characteristics are expanded by using the characteristic information of the target object in the image to be recognized. Therefore, the data processing equipment can simultaneously collect the characteristic information of the target object under various imaging effects in the process of tracking the target object by using the historical characteristics so as to enrich the historical characteristics, thereby realizing the improvement of the identification precision when tracking the target object.
Description
Technical Field
The present application relates to the field of data processing, and in particular, to a target tracking method and a related apparatus.
Background
In some scenarios, target objects need to be identified by the image capturing devices distributed throughout, so as to achieve target tracking across the image capturing devices.
The inventor researches and discovers that the target object can present different visual effects in the visual field range of different image acquisition devices, however, the existing target tracking method is difficult to adapt to the different visual effects of the target object, and then the problem of poor identification precision exists.
Disclosure of Invention
In order to overcome at least one of the deficiencies in the prior art, the present application provides a target tracking method and related apparatus, including:
in a first aspect, this embodiment provides a target tracking method, applied to a data processing device, where the data processing device is communicatively connected to a plurality of image capturing devices, and the method includes:
acquiring an image to be identified, which is acquired by the image acquisition device;
determining the target object from the image to be recognized according to the historical characteristics of the target object;
updating the historical characteristics through the characteristic information of the target object in the image to be recognized to obtain new historical characteristics;
and adding a new tracking record of the target object, wherein the tracking record represents that the target object is shot once by the image acquisition device.
In a second aspect, this embodiment provides a target tracking apparatus, which is applied to a data processing device, where the data processing device is communicatively connected to a plurality of image capturing devices, and the target tracking apparatus includes:
the image module is used for acquiring an image to be identified, which is acquired by the image acquisition device;
the identification module is used for determining the target object from the image to be identified according to the historical characteristics of the target object;
the identification module is further used for updating the historical characteristics through the characteristic information of the target object in the image to be identified to obtain new historical characteristics;
and the tracking module is used for newly adding a tracking record of the target object, and the tracking record represents that the target object is shot by the image acquisition device once.
In a third aspect, the present embodiment provides a data processing apparatus comprising a processor and a memory, wherein the memory stores a computer program, and the computer program, when executed by the processor, implements the target tracking method.
In a fourth aspect, the present embodiment provides a computer storage medium storing a computer program which, when executed by a processor, implements the object tracking method.
Compared with the prior art, the method has the following beneficial effects:
in the target tracking method and the related device provided by the embodiment of the application, after the data processing equipment determines the target object from the image to be recognized by using the historical characteristics, the tracking record of the target object is generated, and meanwhile, the historical characteristics are expanded by using the characteristic information of the target object in the image to be recognized. Therefore, the data processing equipment can simultaneously collect the characteristic information of the target object under various imaging effects in the process of tracking the target object by using the historical characteristics so as to enrich the historical characteristics, thereby realizing the improvement of the identification precision when tracking the target object.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are required to be used in the embodiments will be briefly described below, it should be understood that the following drawings only illustrate some embodiments of the present application and therefore should not be considered as limiting the scope, and for those skilled in the art, other related drawings can be obtained from the drawings without inventive effort.
Fig. 1 is a schematic view of a scenario provided in an embodiment of the present application;
fig. 2 is a schematic structural diagram of a data processing apparatus according to an embodiment of the present application;
FIG. 3 is a flowchart of a target tracking method provided in an embodiment of the present application;
FIG. 4 is a schematic flow chart diagram provided by an embodiment of the present application;
fig. 5 is a schematic structural diagram of a target tracking device according to an embodiment of the present application.
Icon: 120-a memory; 130-a processor; 140-a communication unit; 201-an image module; 202-an identification module; 203-tracking module.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments 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 of the present application, but not all embodiments. The components of the embodiments of the present application, generally described and illustrated in the figures herein, can be arranged and designed in a wide variety of different configurations.
Thus, the following detailed description of the embodiments of the present application, presented in the accompanying drawings, is not intended to limit the scope of the claimed application, but is merely representative of selected embodiments of the application. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
It should be noted that: like reference numbers and letters refer to like items in the following figures, and thus, once an item is defined in one figure, it need not be further defined and explained in subsequent figures.
In the description of the present application, the terms "first," "second," "third," and the like are used solely to distinguish one from another and are not to be construed as indicating or implying relative importance. Furthermore, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.
It should be understood that the operations of the flow diagrams may be performed out of order, and steps without logical context may be performed in reverse order or simultaneously. One skilled in the art, under the guidance of this application, may add one or more other operations to, or remove one or more operations from, the flowchart.
In some scenarios, target objects need to be identified by the image capturing devices distributed throughout, so as to achieve target tracking across the image capturing devices.
Taking a security scene as an example, assuming that the target object is a target suspect, image acquisition devices deployed at different positions need to be jointly used for tracking and predicting the motion track of the target suspect so as to facilitate subsequent capture.
Or taking the detection of the abnormal behavior as an example, assuming that the abnormal behavior represents a cross-regional loitering behavior, the image acquisition devices deployed at different positions need to be combined to track the trajectory of the object to be recognized, so as to determine whether the target solely shares the phenomenon of the cross-regional loitering behavior.
In the related art, the specific feature information of the target object is mostly used as a reference to be matched with the images acquired by the image acquisition devices, so as to identify the target object. However, the target object may exhibit a distinctive visual effect in the field of view of different image capturing devices. For example, some image capturing devices capture only the side face, the back head, or the side face of the body of the target object, however, the features of these regions are greatly different from the specific feature information of the target object, which may result in a failure in recognition.
In view of this, the present embodiment provides a target tracking method, which is applied to a data processing device. According to the method, the data processing equipment collects historical characteristics of a target object in different image acquisition device shooting ranges, and is used for matching the latest shot image to be identified and determining the target object. Thus, the recognition accuracy in tracking the target object is improved.
In some embodiments, the data processing device may be a server. For example, a Web server, an FTP (File Transfer Protocol) server, a data processing server, and the like. In addition, the server may be a single server or a server group. The set of servers can be centralized or distributed (e.g., the servers can be a distributed system). In some embodiments, the server may be local or remote to the user terminal. In some embodiments, the server may be implemented on a cloud platform; by way of example only, the cloud platform may include a private cloud, a public cloud, a hybrid cloud, a community cloud (community cloud), a distributed cloud, an inter-cloud, a multi-cloud, and the like, or any combination thereof. In some embodiments, the server may be implemented on an electronic device having one or more components.
As shown in fig. 1, when the data processing device is a server, the data processing device is in communication connection with image acquisition devices at different erection positions through a network to acquire images to be identified acquired by the image acquisition devices. Wherein, the image acquisition device can be a camera erected along a road or a camera erected in other ways.
In some embodiments, the data processing apparatus may also be an image acquisition device communicatively connected to the server. The server is also in communication connection with other image acquisition devices and is used for synchronizing the historical characteristics of the target object to each image acquisition device, so that each image acquisition device can determine the target object from the image to be recognized shot by the image acquisition device based on the historical characteristics of the target object.
The present embodiment also provides a hardware structure of the data processing apparatus, as shown in fig. 2, the data processing apparatus includes a memory 120, a processor 130, and a communication unit 140.
The memory 120, processor 130, and communication unit 140 are electrically connected to each other directly or indirectly to enable data transmission or interaction. For example, the components may be electrically connected to each other via one or more communication buses or signal lines.
The Memory 120 may be, but is not limited to, a Random Access Memory (RAM), a Read Only Memory (ROM), a Programmable Read-Only Memory (PROM), an Erasable Read-Only Memory (EPROM), an electrically Erasable Read-Only Memory (EEPROM), and the like. The memory 120 is used for storing a program, and the processor 130 executes the program after receiving the execution instruction.
The communication unit 140 is used for transceiving data through a network. The Network may include a wired Network, a Wireless Network, a fiber optic Network, a telecommunications Network, an intranet, the internet, a Local Area Network (LAN), a Wide Area Network (WAN), a Wireless Local Area Network (WLAN), a Metropolitan Area Network (MAN), a Wide Area Network (WAN), a Public Switched Telephone Network (PSTN), a bluetooth Network, a ZigBee Network, or a Near Field Communication (NFC) Network, or the like, or any combination thereof. In some embodiments, the network may include one or more network access points. For example, the network may include wired or wireless network access points, such as base stations and/or network switching nodes, through which one or more components of the service request processing system may connect to the network to exchange data and/or information.
Based on the above description about the data processing device and the usage scenario of the data processing device, in order to enable those skilled in the art to use the present disclosure, the steps of the target tracking method are described in detail below with reference to fig. 3. As shown in fig. 3, the method includes:
and S101, acquiring an image to be identified.
And S102, determining the target object from the image to be recognized according to the historical characteristics of the target object.
In some embodiments, the data processing apparatus may be a server communicatively connected to a plurality of image capture devices. The server obtains an image to be recognized collected by an image collecting device through a network, uses the historical characteristics of a target object to match with the object to be recognized in the image to be recognized, and takes the object to be recognized with the highest similarity and the similarity larger than a matching threshold value as the target object.
In other embodiments, the data processing apparatus may be a distributed deployment of image capture devices having mutually synchronized historical features between the image capture devices. The image acquisition device is used for matching an object to be recognized in an image to be recognized locally by using historical characteristics aiming at the image to be recognized acquired by the image acquisition device, and the object to be recognized with the highest similarity and the similarity larger than a matching threshold value is used as a target object.
And S103, updating the historical characteristics through the characteristic information of the target object in the image to be recognized, and obtaining new historical characteristics.
As one implementation, the history feature may include at least one reference feature, respectively obtained from different history images including the target object. The data processing device acquires feature information of a target object in an image to be recognized and adds the feature information to the historical features to obtain new historical features.
However, it is considered that the data processing apparatus matches at least one reference feature with each object to be recognized in the image to be recognized, and therefore, when the number of reference features is large, the recognition efficiency may be reduced.
In view of this, in order to improve the recognition accuracy while maintaining high recognition efficiency, the data processing apparatus acquires the similarity between the feature information and the history feature; and when the similarity meets the condition of being smaller than the updating threshold, updating the historical characteristics through the characteristic information to obtain new historical characteristics.
For example, the data processing device obtains the similarity between at least one reference feature and the object to be recognized respectively; selecting the maximum target similarity from the target similarity; and if the target similarity exceeds a similarity threshold, determining that the object to be identified is the target object.
And if the target similarity is greater than the similarity threshold and smaller than the update threshold, the data processing equipment adds the feature information of the target object in the image to be identified into the historical feature to update the historical feature.
Illustratively, assume that the similarity threshold is 60% and the update threshold is 80%; when the target similarity is between 60% and 80%, the target object exists in the image to be recognized, and when the target similarity exceeds 80%, the historical characteristic corresponding to the target similarity is extremely similar to the characteristic information of the target object in the image to be recognized, and the historical characteristic does not need to be added into the historical characteristic. Thus, the purpose of reducing the number of reference features in the historical features is achieved.
And S104, adding a tracking record of the target object, wherein the tracking record represents that the image acquisition device shoots the target object once.
Based on the above embodiment, after the data processing device determines the target object from the image to be recognized by using the historical features, the data processing device generates a tracking record of the target object and expands the historical features by using the feature information of the target object in the image to be recognized. Therefore, the data processing equipment can simultaneously collect the characteristic information of the target object under various imaging effects in the process of tracking the target object by using the historical characteristics so as to enrich the historical characteristics, thereby realizing the improvement of the identification precision when tracking the target object.
In this embodiment, in order to improve the recognition accuracy of the target object, in an optional embodiment, a hierarchical matching mode may be adopted to improve the matching success rate. In a hierarchical matching mode, the historical characteristics of the target object comprise historical human face characteristics and historical human body characteristics, and the similarity threshold comprises a human face threshold and a human body threshold; correspondingly, the similarity includes a human face similarity corresponding to a human face threshold value and a human body similarity corresponding to a human body threshold value.
In a corresponding embodiment, the data processing equipment detects a target object from an image to be recognized through historical human face characteristics; and if the target object with the human face similarity larger than the human face threshold value is not detected, determining the target object with the human body similarity larger than the human body threshold value from the image to be recognized through historical human body characteristics.
Of course, if neither of the two methods is matched with the target object, it indicates that the target object does not appear in the field of view of the image capturing device.
The inventor further researches and discovers that the historical human face features are firstly used for matching with the object to be recognized in the image to be recognized, if the target object is not determined, the historical human body features are used for matching with the object to be recognized in the image to be recognized, if the human face of the target object is shot, the historical human face features are used for matching, the accuracy is high, and therefore only the human body feature information of the target object needs to be collected.
In an optional embodiment, in order to further reduce the number of reference features in the historical features, if the data processing device detects a target object from the image to be recognized through the historical face features, human body feature information of the target object in the image to be recognized is acquired; and calculating the human body similarity between the human body characteristic information and the historical human body characteristics.
And if the human body similarity meets the condition that the human body similarity is smaller than the updating threshold, the data processing equipment updates the historical human body characteristics through the human body characteristic information.
Illustratively, the reference features are divided into reference face features and reference body features, wherein the reference face features are attributed to historical face features, and the reference body features are attributed to historical body features. Then, the data processing device compares the human body characteristics of the target object with each reference human body characteristic, and selects the maximum human body similarity.
And if the human body similarity is smaller than the update threshold, the data processing equipment acquires the human body characteristic information of the target object in the image to be recognized, and adds the human body characteristic information into the historical human body characteristics to acquire new historical characteristics.
It should be noted that, for a new image to be recognized, the data processing device uses the new history feature to recognize whether the new image to be recognized has a target object.
In addition, the pedestrian loitering detection is common abnormal behavior detection in the security field, however, the loitering detection in a single-camera scene is mostly adopted in the related technology; with the change of security and protection requirements, the loitering detection across areas or across cameras is required to be realized in some scenes.
In view of this, when the data processing device is a server communicatively connected to the plurality of image capturing apparatuses, the number of times of shooting of the target object in the first time range is acquired, wherein the number of times of shooting is acquired from at least two target devices of the plurality of image capturing apparatuses.
If the shooting times are larger than the first time threshold value, the server determines that the target object wanders between the shooting areas of the at least two target devices.
Illustratively, assume that the first time range is 1 hour and the first count threshold is 6 times; the server is in communication connection with 3 image acquisition devices, namely a camera A, a camera B and a camera C. In 1 hour, the camera A shoots the target object 4 times, the camera B shoots the target object 1 time, and the camera C shoots the target object 3 times; it may be determined that the target object wanders between the shooting areas of camera a and camera B.
Of course, in some embodiments, the server determines that the target object wandering behavior exists if the server detects that the target object appears in the shooting area of the single image capturing device within the second time range more than a second threshold number of times.
In other embodiments, the server counts the loitering time length of the target object in the shooting area of the single image acquisition device, and determines that the target object has loitering behavior if the loitering time length threshold value is exceeded.
The embodiment also considers that the time interval between two times of shooting of the target object needs to be less than the duration threshold value so as to be judged to have the association. For example, if the time interval between two shots exceeds 1 day, the wandering behavior is less likely to occur. Therefore, in order to reduce the number of trace records, the efficiency of data analysis is improved; the server determines a target tracking record with the latest shooting time from the historical tracking records; and if the time difference between the newly added tracking record and the target tracking record meets the condition that the time difference is greater than the time length threshold value, deleting the historical detection record.
For example, assuming that the time duration threshold is 6 hours, the historical trace records include 10 trace records, wherein the time of the latest trace record is 9:00, and the time information of the newly added trace record is 16:00, and the time difference between the two is more than 6 hours, the historical trace record is deleted, and the newly added trace record is reserved.
In order to enable those skilled in the art to use the solution provided by the present embodiment, it is assumed that the data processing apparatus is a server, and the following embodiments are given in conjunction with fig. 4. It will be apparent to those skilled in the art that the general principles defined herein may be applied to other embodiments and applications without departing from the spirit and scope of the application.
As shown in fig. 4, in order to facilitate monitoring 24 hours a day, the camera includes a light sensor, a lighting controller, and an auxiliary lighting device. The light sensor is used for receiving light signals and sending the light signals to the lighting controller, and the lighting controller controls the auxiliary lighting device to be turned on or turned off according to the intensity of the light.
And for each camera in the area range, the camera acquires the light brightness degree in the environment through a light sensor, and when the light brightness degree is smaller than a preset brightness threshold value, the auxiliary lighting device is turned on to control the auxiliary lighting device to start providing auxiliary lighting.
All cameras in the area range acquire pedestrian monitoring images in parallel, and the monitoring images are used as images to be identified and sent to a server together with the IDs of the cameras. After receiving the image to be identified, the server starts cross-domain parallel pedestrian detection and cross-domain parallel pedestrian tracking to generate a unique tracking identifier (TrackID) of the pedestrian, wherein the tracking identifier comprises the ID of the camera and the system time.
The following is a detailed description of the process of tracking the target object by the server with continued reference to fig. 4:
as shown in fig. 4, the present embodiment provides a face recognition database and a pedestrian re-identification (ReID) database for storing historical features and tracking records of different pedestrians, wherein the face recognition database is used for storing historical face features and tracking records generated based on the historical face features, and the different pedestrians are distinguished by face recognition id (face id). The pedestrian re-identification database is used for storing historical human body characteristics and tracking records generated based on the historical human body characteristics, and distinguishing different pedestrians by using pedestrian re-identification IDs (ReIDs); for the same pedestrian, the face recognition ID and the pedestrian re-recognition ID of the pedestrian are associated by a tracking identifier (TrackID).
Based on the design, the server cuts out the pedestrian image from the image to be recognized through a target detection algorithm aiming at each image to be recognized. And aiming at each pedestrian image, the server adopts a preset face detection model to detect whether the pedestrian image contains a face image with a confidence degree higher than the face confidence degree. According to the detection result, the following two cases can be classified:
the method comprises the following steps of:
the server outputs face coordinate information and judges whether the registered number of face identification IDs in the current face identification database is 0 or not.
If 0, it means that the face recognition database is empty, that is, the pedestrian image is the first image to be recognized, and therefore, the server sets the pedestrian in the pedestrian image as the target object, generates the TrackID of the target object, registers the face recognition ID of the target object with the TrackID, and records the face recognition ID in the face recognition data together with the registration time, thereby completing initialization of the face recognition data.
If not, indicating that the face identification data stores the face identification ID; the server takes the pedestrian in the pedestrian image as a target object and obtains the face characteristics of the target object in the pedestrian image; and then searching a face recognition database by using the face features, and if an object with the face similarity larger than a face threshold value is matched, indicating that the historical face features of the target object are recorded in the face recognition data, so that a tracking record is generated by using the face recognition ID of the target object.
If there is no object whose face similarity is greater than the face threshold, it means that the target object is detected for the first time, and therefore, a trackID of the target object is generated, and the face identification ID of the target object is registered with the trackID and recorded in the face identification data together with the registration time.
Referring to fig. 4 again, in this embodiment, when the target object cannot be determined from the image to be recognized through the historical human face features, the historical human body features are used for recognition again. In order to establish the relation between the historical human face characteristics and the historical human body characteristics of the target object, the server extracts the human body characteristics of the target object from the pedestrian image when a human face recognition ID is newly established on the basis of the human face characteristics or a tracking record of the target object is newly added on the basis of the human face characteristics, then the pedestrian re-recognition ID is registered by using the TrackID and is recorded into a pedestrian re-recognition database together with the registration time and the human body characteristics, and therefore the ReiD characteristic adding support is achieved.
Face images are not included:
the server first detects whether the number of the pedestrian re-identification IDs in the pedestrian re-identification database is 0.
If the number is 0, the pedestrian re-identification database is empty, the server generates the TrackID of the target object, registers the pedestrian re-identification ID of the target object by using the TrackID, and records the pedestrian re-identification ID and the registration time into the pedestrian re-identification database, thereby completing the initialization of the pedestrian re-identification database.
If not, the pedestrian re-identification database stores the pedestrian re-identification ID; the server extracts the human body features of the target object from the pedestrian image, searches the pedestrian re-identification database by the human body features, if an object with human body similarity larger than a human body threshold value is matched, the server indicates that the historical human body features of the target object are recorded in the human face identification data, and then the pedestrian re-identification ID of the target object generates a tracking record.
And if the similarity is smaller than the update threshold, extracting the human body characteristics of the target object from the pedestrian image and recording the human body characteristics in a pedestrian re-identification database.
And finally, the server carries out statistical analysis on the pedestrian re-identification ID in the pedestrian re-identification database and the face identification ID in the face identification data so as to identify a primary alarm with the wandering times exceeding a second time threshold value in the shooting range of a single camera or identify a secondary alarm with the transregional wandering times exceeding a first time threshold value. Thus, the process is repeated, and the loitering behavior is continuously detected.
Based on the same inventive concept as the target tracking method, the implementation also provides a related device of the method, which comprises the following steps:
the embodiment also provides a target tracking device which is applied to the data processing equipment. The target tracking device includes at least one functional module that may be stored in software in the memory 120. As shown in fig. 5, functionally divided, the target tracking apparatus may include:
the image module 201 is configured to acquire an image to be recognized.
In this embodiment, the image module 201 is used to implement step S101 in fig. 3, and for a detailed description of the image module 201, reference may be made to the detailed description of step S101.
The recognition module 202 is configured to determine a target object from the image to be recognized according to the historical features of the target object.
The identification module 202 is further configured to update the historical features according to the feature information of the target object in the image to be identified, so as to obtain new historical features.
In this embodiment, the identification module 202 is configured to implement steps S102 to S103 in fig. 3, and for a detailed description of the identification module 202, reference may be made to detailed descriptions of steps S102 to S103.
And the tracking module 203 is used for newly adding a tracking record of the target object, wherein the tracking record indicates that the target object is shot by the image acquisition device corresponding to the image to be identified once.
In this embodiment, the tracking module 203 is configured to implement step S104 in fig. 3, and for a detailed description of the tracking module 203, reference may be made to a detailed description of step S104.
It should be noted that the target tracking device may further include other software functional modules, which are used to implement other steps or sub-steps of the target tracking method; of course, the image module 201, the recognition module and the tracking module 203 may also be used to implement other steps or sub-steps of the target tracking method; the present embodiment does not specifically limit this, and those skilled in the art may appropriately adjust the module division criteria based on different module division criteria.
The embodiment further provides a data processing device, which includes a processor and a memory, where the memory stores a computer program, and when the computer program is executed by the processor, the target tracking method is implemented.
The present embodiment also provides a computer storage medium, which stores a computer program, and when the computer program is executed by a processor, the target tracking method is implemented.
In summary, in the target tracking method and the related apparatus provided in the embodiment of the present application, after the data processing device determines the target object from the image to be recognized by using the historical features, the tracking record of the target object is generated, and meanwhile, the historical features are extended by using the feature information of the target object in the image to be recognized. Therefore, the data processing equipment can simultaneously collect the characteristic information of the target object under various imaging effects in the process of tracking the target object by using the historical characteristics so as to enrich the historical characteristics, thereby realizing the improvement of the identification precision when tracking the target object.
In the embodiments provided in the present application, it should be understood that the disclosed apparatus and method may be implemented in other ways. The apparatus embodiments described above are merely illustrative, and for example, the flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of apparatus, methods and computer program products according to various embodiments of the present application. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
In addition, functional modules in the embodiments of the present application may be integrated together to form an independent part, or each module may exist separately, or two or more modules may be integrated to form an independent part.
The functions, if implemented in the form of software functional modules and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present application or portions thereof that substantially contribute to the prior art may be embodied in the form of a software product stored in a storage medium and including instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present application. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.
The above description is only for various embodiments of the present application, but the scope of the present application is not limited thereto, and any person skilled in the art can easily conceive of changes or substitutions within the technical scope of the present application, and all such changes or substitutions are included in the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.
Claims (10)
1. A target tracking method applied to a data processing apparatus, the method comprising:
acquiring an image to be identified;
determining the target object from the image to be recognized according to the historical characteristics of the target object;
updating the historical characteristics through the characteristic information of the target object in the image to be recognized to obtain new historical characteristics;
and adding a new tracking record of the target object, wherein the tracking record represents that the target object is shot once by an image acquisition device corresponding to the image to be identified.
2. The target tracking method according to claim 1, wherein the updating the historical features through the feature information of the target object in the image to be recognized to obtain new historical features comprises:
acquiring similarity between the characteristic information and the historical characteristics;
and when the similarity meets the condition of being smaller than an updating threshold, updating the historical characteristics through the characteristic information to obtain the new historical characteristics.
3. The target tracking method of claim 2, wherein the updating the historical features by the feature information to obtain the new historical features comprises:
and adding the characteristic information into the historical characteristics to obtain the new historical characteristics.
4. The target tracking method of claim 2, wherein the historical features comprise historical face features and historical body features;
the obtaining of the similarity between the feature information and the historical features includes:
if the target object is detected from the image to be recognized through the historical human face features, acquiring human body feature information of the target object in the image to be recognized;
acquiring human body similarity between the human body characteristic information and the historical human body characteristics;
when the similarity meets the condition that the similarity is smaller than an updating threshold, updating the historical characteristics through the characteristic information to obtain the new historical characteristics, wherein the method comprises the following steps:
and if the human body similarity meets the condition that the human body similarity is smaller than an updating threshold, updating the historical human body characteristics through the human body characteristic information.
5. The target tracking method according to claim 1, wherein the historical features include historical human face features and historical human body features, and determining the target object from the image to be recognized according to the historical features of the target object comprises:
detecting the target object from the image to be recognized through the historical human face features;
and if the target object is not detected, determining the target object from the image to be recognized through the historical human body characteristics.
6. The target tracking method of claim 1, further comprising:
determining a target tracking record with the latest shooting time from the historical tracking records;
and if the time difference between the newly added tracking record and the target tracking record meets the condition that the time difference is greater than a time length threshold value, deleting the historical detection record.
7. The object tracking method according to claim 1, wherein when the data processing apparatus is a server communicatively connected to a plurality of image capturing devices, the method further comprises:
acquiring shooting times of the target object in a first time range, wherein the shooting times are acquired from at least two target devices in the plurality of image acquisition devices;
if the shooting times are larger than a first time threshold value, determining that the target object wanders between the shooting areas of the at least two target devices.
8. An object tracking apparatus applied to a data processing device, the object tracking apparatus comprising:
the image module is used for acquiring an image to be identified;
the identification module is used for determining the target object from the image to be identified according to the historical characteristics of the target object;
the identification module is further used for updating the historical characteristics through the characteristic information of the target object in the image to be identified to obtain new historical characteristics;
and the tracking module is used for newly adding a tracking record of the target object, and the tracking record represents that the target object is shot once by the image acquisition device corresponding to the image to be identified.
9. A data processing device, characterized in that the data processing device comprises a processor and a memory, the memory storing a computer program which, when executed by the processor, implements the object tracking method of any one of claims 1-7.
10. A computer storage medium, characterized in that the computer storage medium stores a computer program which, when executed by a processor, implements the object tracking method of any one of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111136528.8A CN113793363A (en) | 2021-09-27 | 2021-09-27 | Target tracking method and related device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111136528.8A CN113793363A (en) | 2021-09-27 | 2021-09-27 | Target tracking method and related device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113793363A true CN113793363A (en) | 2021-12-14 |
Family
ID=79184621
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111136528.8A Pending CN113793363A (en) | 2021-09-27 | 2021-09-27 | Target tracking method and related device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113793363A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114758485A (en) * | 2022-04-21 | 2022-07-15 | 成都商汤科技有限公司 | Alarm information processing method and device, computer equipment and storage medium |
| TWI815351B (en) * | 2022-01-25 | 2023-09-11 | 威盛電子股份有限公司 | Target tracking method and target tracking device |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP2858008A2 (en) * | 2013-09-27 | 2015-04-08 | Ricoh Company, Ltd. | Target detecting method and system |
| CN109102531A (en) * | 2018-08-21 | 2018-12-28 | 北京深瞐科技有限公司 | A kind of target trajectory method for tracing and device |
| CN109766779A (en) * | 2018-12-20 | 2019-05-17 | 深圳云天励飞技术有限公司 | It hovers personal identification method and Related product |
| CN110175593A (en) * | 2019-05-31 | 2019-08-27 | 努比亚技术有限公司 | Suspect object processing method, wearable device and computer readable storage medium |
| CN110705478A (en) * | 2019-09-30 | 2020-01-17 | 腾讯科技（深圳）有限公司 | Face tracking method, device, equipment and storage medium |
| CN110852219A (en) * | 2019-10-30 | 2020-02-28 | 广州海格星航信息科技有限公司 | Multi-pedestrian cross-camera online tracking system |
| CN111709391A (en) * | 2020-06-28 | 2020-09-25 | 重庆紫光华山智安科技有限公司 | Human face and human body matching method, device and equipment |
| CN112084939A (en) * | 2020-09-08 | 2020-12-15 | 深圳市润腾智慧科技有限公司 | Image feature data management method and device, computer equipment and storage medium |
| WO2020259100A1 (en) * | 2019-06-28 | 2020-12-30 | 深圳市商汤科技有限公司 | Information processing method and apparatus, and storage medium |
| EP3796214A1 (en) * | 2019-09-17 | 2021-03-24 | Tsinghua University | Object recognition system and method |
| CN112686178A (en) * | 2020-12-30 | 2021-04-20 | 中国电子科技集团公司信息科学研究院 | Multi-view target track generation method and device and electronic equipment |
| CN113034550A (en) * | 2021-05-28 | 2021-06-25 | 杭州宇泛智能科技有限公司 | Cross-mirror pedestrian trajectory tracking method, system, electronic device and storage medium |
-
2021
- 2021-09-27 CN CN202111136528.8A patent/CN113793363A/en active Pending
Patent Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP2858008A2 (en) * | 2013-09-27 | 2015-04-08 | Ricoh Company, Ltd. | Target detecting method and system |
| CN109102531A (en) * | 2018-08-21 | 2018-12-28 | 北京深瞐科技有限公司 | A kind of target trajectory method for tracing and device |
| CN109766779A (en) * | 2018-12-20 | 2019-05-17 | 深圳云天励飞技术有限公司 | It hovers personal identification method and Related product |
| CN110175593A (en) * | 2019-05-31 | 2019-08-27 | 努比亚技术有限公司 | Suspect object processing method, wearable device and computer readable storage medium |
| WO2020259100A1 (en) * | 2019-06-28 | 2020-12-30 | 深圳市商汤科技有限公司 | Information processing method and apparatus, and storage medium |
| EP3796214A1 (en) * | 2019-09-17 | 2021-03-24 | Tsinghua University | Object recognition system and method |
| CN110705478A (en) * | 2019-09-30 | 2020-01-17 | 腾讯科技（深圳）有限公司 | Face tracking method, device, equipment and storage medium |
| CN110852219A (en) * | 2019-10-30 | 2020-02-28 | 广州海格星航信息科技有限公司 | Multi-pedestrian cross-camera online tracking system |
| CN111709391A (en) * | 2020-06-28 | 2020-09-25 | 重庆紫光华山智安科技有限公司 | Human face and human body matching method, device and equipment |
| CN112084939A (en) * | 2020-09-08 | 2020-12-15 | 深圳市润腾智慧科技有限公司 | Image feature data management method and device, computer equipment and storage medium |
| CN112686178A (en) * | 2020-12-30 | 2021-04-20 | 中国电子科技集团公司信息科学研究院 | Multi-view target track generation method and device and electronic equipment |
| CN113034550A (en) * | 2021-05-28 | 2021-06-25 | 杭州宇泛智能科技有限公司 | Cross-mirror pedestrian trajectory tracking method, system, electronic device and storage medium |
Non-Patent Citations (3)
| Title |
|---|
| K. SUSHEEL KUMAR等: "Multiple Cameras Using Real Time Object Tracking for Surveillance and Security System", 《2010 3RD INTERNATIONAL CONFERENCE ON EMERGING TRENDS IN ENGINEERING AND TECHNOLOGY》, pages 213 - 218 * |
| SHIPRA OJHA等: "Image processing techniques for object tracking in video surveillance- A survey", 《 2015 INTERNATIONAL CONFERENCE ON PERVASIVE COMPUTING (ICPC)》, pages 1 - 6 * |
| 罗浩: "基于深度学习的行人重识别算法研究 —从无遮挡到遮挡", 《中国优秀博士学位论文全文数据库 (信息科技辑)》, no. 1, pages 138 - 188 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| TWI815351B (en) * | 2022-01-25 | 2023-09-11 | 威盛電子股份有限公司 | Target tracking method and target tracking device |
| CN114758485A (en) * | 2022-04-21 | 2022-07-15 | 成都商汤科技有限公司 | Alarm information processing method and device, computer equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP7131843B2 (en) | Monitoring system, management device and monitoring method | |
| KR102104088B1 (en) | Uwb-based location tracking and ai combined intelligent object tracking video monitoring system | |
| CN107506684B (en) | Gait recognition method and device | |
| CN113793363A (en) | Target tracking method and related device | |
| KR101472077B1 (en) | Surveillance system and method based on accumulated feature of object | |
| JP4984728B2 (en) | Subject collation device and subject collation method | |
| US20080130949A1 (en) | Surveillance System and Method for Tracking and Identifying Objects in Environments | |
| CN104850563B (en) | Destination image compare retrieval device, destination image compares searching system and destination image compares search method | |
| CN108073577A (en) | A kind of alarm method and system based on recognition of face | |
| CN111325954B (en) | Personnel loss early warning method, device, system and server | |
| CN111010547A (en) | Target object tracking method and device, storage medium and electronic device | |
| EP3713225A1 (en) | Communication device, communication system, and mobile body tracking method | |
| JP6503079B2 (en) | Specific person detection system, specific person detection method and detection device | |
| CN109684494A (en) | A kind of scenic spot looking-for-person method, system and cloud server | |
| CN105869226A (en) | Face-recognition-based automatic roll-call system and method for prisons | |
| JP2019180024A (en) | Monitoring device | |
| JP2017063266A (en) | Information processing method, information processing apparatus, and program | |
| CN111294552A (en) | Image acquisition equipment determining method and device | |
| KR20160099289A (en) | Method and system for video search using convergence of global feature and region feature of image | |
| JP2010068466A (en) | Moving body tracking device and moving body tracking method | |
| CN110490104A (en) | A kind of monitoring system and monitoring method reducing tramp's conflict | |
| JP2021036646A (en) | Information collection system, camera terminal device, information collection method and information collection program | |
| KR101754995B1 (en) | Tracking criminals cctv apparatus | |
| CN108540759B (en) | Video monitoring method, device and system | |
| JP2020052822A (en) | Information processing apparatus, authentication system, control method thereof, and program |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |