Object determination method and apparatus, storage medium, and electronic device

Abstract

This invention provides a method, apparatus, storage medium, and electronic device for determining objects. The method includes: performing object detection on a current image frame to obtain a current detection result; when the current detection result includes first information of a first object, determining whether a target platform includes a target file for the first object, wherein the target file is obtained by aggregating information of the first object over a predetermined time period, and the target file stores a set of first information of the first object acquired over the predetermined time period; when the target platform includes the target file, storing the first information into the target file to obtain an updated file, wherein the updated file includes a second set of information of the first object; and determining whether the first object is an object of a specified type based on the updated file. This invention solves the problem of low efficiency in determining specified types of objects in related technologies.

Description

Technical Field

[0001] The embodiments of the present invention relate to the field of video surveillance technology, and more specifically, to a method, apparatus, storage medium, and electronic device for determining an object. Background Technology

[0002] Video surveillance technology has been widely applied in many fields. For example, it can be used to monitor vehicles, people, or restricted areas. The video or image resources obtained through surveillance can be used to determine whether a vehicle is the cause of an accident, or to determine whether a child is a missing or abducted child reported by their parents. In other words, it can be used to confirm whether a target object is a specified type of object. The following example illustrates how to determine if a child is abducted. Currently, most towns and cities have large-scale video surveillance coverage, but cases of child abduction and trafficking are still rampant and difficult to prevent. After a child is abducted, traffickers usually transfer the child to another location as quickly as possible. If the transfer is successful, it is very difficult to find the child again in a vast sea of ​​people.

[0003] In related technologies, photos and related identity information of missing children are uploaded to a remote management center. A mobile client displays these photos to users in real time. Users can take photos of suspected missing or abducted children using the mobile client and upload them to the remote management center for matching. Once a match is found, the result is uploaded to the relevant department's management system. It is evident that these technologies require users to actively upload photos of suspected missing or abducted children. The management system is relatively passive in determining whether a child is abducted, meaning it cannot autonomously analyze whether an object belongs to a specified target type. Therefore, the methods used in these technologies to determine whether an object belongs to a specified type suffer from low efficiency.

[0004] There is currently no effective solution to the problem of low efficiency in determining objects of a specific type in related technologies. Summary of the Invention

[0005] The present invention provides a method, apparatus, storage medium, and electronic device for determining objects, so as to at least solve the problem of low efficiency in determining a specified type of object in the related art.

[0006] According to an embodiment of the present invention, a method for determining an object is provided, comprising: performing object detection on a current image frame to obtain a current detection result; if the current detection result includes first information of a first object detected in the current image frame, determining whether a target platform includes a target file of the first object, wherein the first object is an object of a predetermined age group, and the target file is a file obtained by performing an archive operation on the information of the first object within a predetermined time period in the past, and the target file stores a set of first information of the first object obtained within the predetermined time period; if the target platform includes the target file, storing the first information in the target file to obtain an updated file, wherein the updated file includes a second set of information of the first object, the second set of information including the first set of information and the first information; and determining whether the first object is an object of a specified type based on the updated file.

[0007] In an exemplary embodiment, determining whether the first object is an object of a specified type based on the updated archive includes: determining a first duration for which the first object appears in the target area based on the updated archive, wherein the first information set is information about the first object obtained by object detection of a set of image frames, the set of image frames being image frames captured by a target camera device in the target area before the current image frame; and determining whether the first object is an object of a specified type if the first duration is greater than or equal to a predetermined duration threshold.

[0008] In an exemplary embodiment, determining the first duration of the first object's appearance in the target area based on the updated archive includes: determining the difference between a first moment and a second moment as the first duration, wherein the first moment is the capture moment of the current image frame, the second moment is the capture moment of a target image frame in the set of image frames, the target image frame is the earliest captured image frame in the set of image frames, the updated archive stores the first moment corresponding to the first information and the second moment corresponding to the target information, the target information being information about the first object obtained by object detection of the target image frame.

[0009] In an exemplary embodiment, determining whether the first object is an object of a specified type includes: comparing the second information set of the first object with the information of a group of objects included in a pre-acquired target comparison library to obtain a comparison result, wherein the group of objects included in the target comparison library are all objects of the specified type; and determining that the first object is an object of the specified type when the comparison result indicates that the similarity between the second information of the second object in the information of the group of objects and some or all of the information in the second information set is greater than or equal to a preset similarity threshold.

[0010] In an exemplary embodiment, determining whether the first object is an object of a specified type includes: obtaining a set of monitored objects associated with the first object, wherein the set of monitored objects are the set of image frames and objects detected in the current image frame that have a predetermined contact method with the first object; obtaining family member information of each monitored object in the set of monitored objects to obtain a set of family member information; and determining whether the first object is an object of the specified type based on the set of family member information.

[0011] In an exemplary embodiment, obtaining the family member information of each of the monitored objects in the group of monitored objects includes: searching for files corresponding to each of the monitored objects in the group of monitored objects in an existing file set on the target platform to obtain a set of files, wherein each file in the existing file set stores the family member information of the corresponding object; or sending an information query request to a target device, wherein the information query request carries the information of the group of monitored objects, the information query request is used to request the acquisition of the family member information of each of the monitored objects in the group of monitored objects, the information of the group of monitored objects is the information of each monitored object obtained by object detection of a second set of image frames, the second set of image frames is the image frames of the group of image frames and the image frames in the current image frame in which each monitored object is detected; and acquiring the family member information of each of the monitored objects in the group of monitored objects sent by the target device, wherein the target device is used to find the family member information of each of the monitored objects in the group of monitored objects that matches the information of the group of monitored objects.

[0012] In an exemplary embodiment, determining whether the first object is an object of the specified type based on the set of family member information includes: determining the first object as an object of the specified type when none of the family member information includes objects of the predetermined age group; or, determining the first object as an object of the specified type when the family member information includes objects of the predetermined age group but none of the family member information includes the first object.

[0013] In one exemplary embodiment, the method further includes: issuing a warning notification when it is determined that the first object is an object of the specified type, wherein the warning notification is used to indicate that the first object is an object of the specified type.

[0014] According to another embodiment of the present invention, an object determination apparatus is also provided, comprising: a detection module, configured to perform object detection on a current image frame to obtain a current detection result; a first determination module, configured to determine whether a target platform includes a target file of the first object when the current detection result includes first information of a first object detected in the current image frame, wherein the first object is an object of a predetermined age group, the target file is a file obtained by performing an archive operation on the information of the first object within a predetermined time period in the past, and the target file stores a set of first information of the first object obtained within the predetermined time period; an acquisition module, configured to store the first information in the target file when the target platform includes the target file to obtain an updated file, wherein the updated file includes a second set of information of the first object, the second set of information including the first set of information and the first information; and a second determination module, configured to determine whether the first object is an object of a specified type based on the updated file.

[0015] According to yet another embodiment of the present invention, a computer-readable storage medium is also provided, wherein a computer program is stored therein, wherein the computer program is configured to perform the steps in any of the above method embodiments when executed.

[0016] According to yet another embodiment of the present invention, an electronic device is also provided, including a memory and a processor, wherein the memory stores a computer program and the processor is configured to run the computer program to perform the steps in any of the above method embodiments.

[0017] This invention performs object detection on the current image frame to obtain a current detection result. If the current detection result includes first information about a first object detected in the current image frame, it determines whether the target platform includes a target file for the first object. The first object is an object within a predetermined age group, and the target file is a file obtained by aggregating information about the first object over a predetermined time period. The target file stores a set of first information about the first object acquired over the predetermined time period. If the target platform includes a target file, the first information is stored in the target file to obtain an updated file. The updated file includes a second set of information about the first object, which includes the first information set and the first information. Based on the updated file, it is determined whether the first object is an object of a specified type. In other words, by performing object detection on the acquired current image frame and aggregating the first information of the first object based on the first information included in the current detection result, when the target platform includes a target file for the first object, the first information of the first object is stored in the target file to obtain an updated file. Then, based on the updated file, it can be determined whether the first object is an object of a specified type. This approach avoids the problem of low efficiency in identifying specific object types caused by relying primarily on manual photo uploads and matching in related technologies. Therefore, it solves the problem of low efficiency in identifying specific object types in related technologies, achieving the effect of improving the efficiency of identifying specific object types. Attached Figure Description

[0018] Figure 1 This is a mobile terminal hardware structure block diagram of the object determination method according to an embodiment of the present invention;

[0019] Figure 2 This is a flowchart of a method for determining an object according to an embodiment of the present invention;

[0020] Figure 3 This is an example diagram of the environmental topology according to an embodiment of the present invention;

[0021] Figure 4 This is a schematic diagram of the personnel information collection process according to a specific embodiment of the present invention;

[0022] Figure 5 This is a schematic diagram illustrating the judgment of a guardian according to a specific embodiment of the present invention;

[0023] Figure 6 This is a flowchart of a child safety early warning system according to a specific embodiment of the present invention;

[0024] Figure 7 This is a structural block diagram of an object determination device according to an embodiment of the present invention. Detailed Implementation

[0025] The embodiments of the present invention will be described in detail below with reference to the accompanying drawings and examples.

[0026] It should be noted that the terms "first," "second," etc., in the specification, claims, and drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence.

[0027] The methods and embodiments provided in this application can be executed on a mobile terminal, computer terminal, or similar computing device. Taking running on a mobile terminal as an example, Figure 1 This is a mobile terminal hardware structure block diagram of the object determination method according to an embodiment of the present invention. Figure 1 As shown, a mobile terminal may include one or more ( Figure 1 Only one is shown in the diagram. A processor 102 (which may include, but is not limited to, a microprocessor MCU or a programmable logic device FPGA, etc.) and a memory 104 for storing data are also shown. The mobile terminal may further include a transmission device 106 for communication functions and an input / output device 108. Those skilled in the art will understand that... Figure 1 The structure shown is for illustrative purposes only and does not limit the structure of the mobile terminal described above. For example, the mobile terminal may also include components that are more... Figure 1 The more or fewer components shown, or having the same Figure 1 The different configurations shown.

[0028] The memory 104 can be used to store computer programs, such as application software programs and modules, like the computer program corresponding to the object determination method in this embodiment of the invention. The processor 102 executes various functional applications and data processing by running the computer programs stored in the memory 104, thereby implementing the above-described method. The memory 104 may include high-speed random access memory and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 104 may further include memory remotely located relative to the processor 102, and these remote memories can be connected to the mobile terminal via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0029] The transmission device 106 is used to receive or send data via a network. Specific examples of the network described above may include a wireless network provided by the mobile terminal's communication provider. In one example, the transmission device 106 includes a Network Interface Controller (NIC), which can connect to other network devices via a base station to communicate with the Internet. In another example, the transmission device 106 may be a Radio Frequency (RF) module used for wireless communication with the Internet.

[0030] This embodiment provides a method for determining an object. Figure 2 This is a flowchart of a method for determining an object according to an embodiment of the present invention, such as... Figure 2 As shown, the process includes the following steps:

[0031] Step S202: Perform object detection on the current image frame to obtain the current detection result;

[0032] Step S204: If the current detection result includes the first information of the first object detected in the current image frame, determine whether the target platform includes a target file of the first object, wherein the first object is an object of a predetermined age group, the target file is a file obtained by performing a file aggregation operation on the information of the first object within a predetermined time period in the past, and the target file stores a set of the first information of the first object obtained within the predetermined time period.

[0033] Step S206: If the target platform includes the target file, the first information is stored in the target file to obtain an updated file, wherein the updated file includes a second information set of the first object, and the second information set includes the first information set and the first information;

[0034] Step S208: Determine whether the first object is an object of a specified type based on the updated file.

[0035] Through the above steps, object detection is performed on the current image frame to obtain the current detection result. If the current detection result includes the first information of the first object detected in the current image frame, it is determined whether the target platform includes a target file for the first object. Here, the first object is an object within a predetermined age group, and the target file is a file obtained by archiving the information of the first object over a predetermined period of time. The target file stores a set of first information about the first object acquired within the predetermined period of time. If the target platform includes a target file, the first information is stored in the target file to obtain an updated file. The updated file includes a second set of information about the first object, which includes the first information set and the first information. Based on the updated file, it is determined whether the first object is an object of a specified type. In other words, by performing object detection on the acquired current image frame and archiving the first information of the first object based on the first information included in the current detection result, when the target platform includes a target file for the first object, the first information of the first object is stored in the target file to obtain an updated file. Then, based on the updated file, it can be determined whether the first object is an object of a specified type. This approach avoids the problem of low efficiency in identifying specific object types caused by relying primarily on manual photo uploads and matching in related technologies. Therefore, it solves the problem of low efficiency in identifying specific object types in related technologies, achieving the effect of improving the efficiency of identifying specific object types.

[0036] The entity executing the above steps may be a platform, a cloud platform, a server, a terminal device, a processor with human-computer interaction capabilities configured on a storage device, or a processing device or processing unit with similar processing capabilities, but is not limited to these.

[0037] In step S202 above, object detection is performed on the current image frame to obtain the current detection result; optionally, in practical applications, the front-end camera device can perform object detection on the image frame and then upload the detection result to the target platform, or the front-end camera device can acquire the image frame and upload the image frame to the target platform, and then the target platform can perform object detection on the image frame to obtain the detection result;

[0038] As an optional implementation of step S204 above, if the current detection result includes the first information of the first object detected in the current image frame, it is determined whether the target platform includes a target file of the first object. The first object is an object within a predetermined age range, and the target file is a file obtained by aggregating the information of the first object over a predetermined period of time. The target file stores a set of the first information of the first object acquired within the predetermined period of time. For example, the first object is a child (e.g., a child aged 3-7). The current image frame can be an image captured by a camera device in a certain area (e.g., town T) and uploaded to the target platform. In practical applications, this area (e.g., town T) may include multiple camera devices, such as 50, 100, or others. Each camera device uploads the captured image frames or videos to the target platform. Optionally, the camera device can upload image frames or videos in real time or at certain intervals. In practical applications, the cameras in this area can capture image frames or videos of all objects. The image is uploaded to the target platform, where it can perform image aggregation based on the characteristic information of each object. For example, the characteristic information may include facial and / or body feature information and / or age information. After aggregation, each object's snapshots taken at different times and spaces can be stored as natural persons, meaning each object has a corresponding file. Optionally, if the object is an adult in the local area (such as the aforementioned town T), the object's identity information and / or family member information can be further clarified based on information from relevant departments or platforms. For example, the object's ID card, name, and family members can be confirmed based on photos and other information in the public security system. When the first object is a child, for example, transferred from another place to the local area (such as the aforementioned town T), the target platform archives the first object's information after obtaining it, forming a file. If, before the current image frame, the camera equipment in this area may have captured the first object once or multiple times and uploaded the first object's information to the target platform for aggregation, then the target file may already exist on the target platform.

[0039] As an optional implementation of step S206 above, if the target platform includes a target file, the first information is stored in the target file to obtain an updated file. That is, the updated file includes both the previous set of first information and the first information of the first object obtained by detecting the current frame image. Optionally, the updated file contains information about the first object acquired each time, such as the feature information of the first object. Optionally, the file may also include information such as the time and location of each capture of the first object.

[0040] In step S208 above, it is determined whether the first object is a specified type of object based on the updated file. Taking the first object as an example, there may be many different situations in which the first object appears in the local area, such as visiting relatives or tourists from other places, or being trafficked to the local area. In this embodiment, the updated file can be used to further determine whether the first object is a specified type of object. For example, the updated file contains records of the time and location information of each time the first object is captured in the local area (such as the town T mentioned above). The duration of the first object's presence in the local area can be determined based on the updated file, such as the duration of presence being one month or more than two months. Optionally, the profile of the first object (such as the target profile mentioned above, or the updated profile) may also include information about other objects associated with the first object. For example, children are usually accompanied by an adult (such as object A) when they go out, or they may be accompanied by a guardian. When the front-end camera captures the first object and determines that object A may be a guardian, the information of object A can be stored in the profile of the first object. Since the first object may have one or more guardians (or suspected guardians), the target platform can establish or generate a list of guardians. The list can record the time, location, and information of the guardians when they appear at the same time as the first object. In this way, the updated profile can determine whether the first object is a specified type of object, such as whether the first object mentioned above is a trafficked child.

[0041] Through the above embodiments, object detection is performed on the acquired current image frame, and the first information of the first object included in the current detection result is archived. When the target platform includes a target file of the first object, the first information of the first object is stored in the target file to obtain an updated file. Then, based on the updated file, it can be determined whether the first object is an object of a specified type. This avoids the problem of low efficiency in determining specified types of objects caused by the method of mainly relying on manual uploading of photos and then matching and confirming in related technologies. Therefore, the problem of low efficiency in determining specified types of objects in related technologies is solved, and the effect of improving the efficiency of determining specified types of objects is achieved.

[0042] In an optional embodiment, determining whether the first object is an object of a specified type based on the updated archive includes: determining a first duration of the first object's appearance in the target area based on the updated archive, wherein the first information set is information about the first object obtained by object detection of a set of image frames, the set of image frames being image frames captured by a target camera device in the target area before the current image frame; and determining whether the first object is an object of a specified type if the first duration is greater than or equal to a predetermined duration threshold. In this embodiment, the first duration of the first object's appearance in the target area can be determined based on the updated archive. For example, the updated archive records information such as the time and location of each appearance of the first object in the target area. Therefore, the first duration can be determined based on the updated archive. For example, the first duration is 20 days (or 42 days, or 500 hours, or other durations), and the predetermined duration threshold can be set to 30 days. When the first duration is greater than or equal to the predetermined duration threshold, the first object is then determined to be an object of a specified type based on the updated archive. That is, the type of the first object is determined only after the first object appears for more than a predetermined time threshold, thus avoiding the problem of a large workload caused by determining it every time it appears.

[0043] In an optional embodiment, determining the first duration of the first object's appearance in the target area based on the updated archive includes: determining the difference between a first moment and a second moment as the first duration, wherein the first moment is the capture time of the current image frame, the second moment is the capture time of the target image frame in the group of image frames, the target image frame is the earliest captured image frame in the group of image frames, the updated archive stores the first moment corresponding to the first information and the second moment corresponding to the target information, the target information being the information of the first object obtained by object detection of the target image frame. In this embodiment, when determining the first duration, the shooting time of the current image frame can be used as the first moment, the shooting time corresponding to the earliest image frame that captured the first object can be used as the second moment, and the difference between the first moment and the second moment can be used as the first duration. In practical applications, the updated archive may include information such as the shooting time and location of the first object each time it is captured. The updated archive may include one or more records, each record corresponding to the information of the first object obtained after detecting the image frame that includes the first object collected by the front-end camera device each time. The second moment may also be the information of a certain capture in the middle of the updated archive. Through this embodiment, the purpose of determining the first duration of the first object appearing in the target area is based on the updated archive.

[0044] In an optional embodiment, determining whether the first object is an object of a specified type includes: comparing the second information set of the first object with the information of a group of objects included in a pre-acquired target comparison library to obtain a comparison result, wherein the group of objects included in the target comparison library are all objects of the specified type; if the comparison result indicates that the similarity between the second information of the second object in the group of objects and some or all of the information in the second information set is greater than or equal to a preset similarity threshold, then the first object is determined to be an object of the specified type. In this embodiment, the second information set of the first object can be compared with the information of a group of objects included in the target comparison library. For example, taking the determination of whether the first object is a trafficked child as an example, the target comparison library can be a missing children database provided by the public security system or a certain platform. The missing children database can include information on missing children in various regions of the country. When the similarity between some or all of the information in the second information set and the second information of the second object in the target comparison library is greater than or equal to a preset similarity threshold (such as 85% or 90%, or others), then the first object can be determined to be an object of the specified type. The second object can be any object in the target comparison library. This embodiment achieves the goal of determining whether the first object is an object of a specified type by comparing the information of the first object with the information of objects included in the target comparison library.

[0045] In an optional embodiment, determining whether the first object is an object of a specified type includes: obtaining a set of monitored objects associated with the first object, wherein the set of monitored objects are the set of image frames and objects detected in the current image frame that have a predetermined contact method with the first object; obtaining family member information of each monitored object in the set of monitored objects to obtain a set of family member information; and determining whether the first object is an object of the specified type based on the set of family member information. In this embodiment, a set of monitored objects associated with the first object can be obtained. These monitored objects can be objects detected from the aforementioned set of image frames and the current image frame that have a predetermined contact relationship with the first object. For example, in the aforementioned set of image frames or the current image frame, the image frame may include both the first object and one or more adults (such as adult A). The image frames can then be analyzed to determine whether a predetermined contact relationship exists between the adult and the first object. For example, the predetermined contact relationship might be adult A holding the first object or adult A hugging the first object. When this predetermined contact relationship exists, adult A can be temporarily designated as a monitored object of the first object. The first object may have been held or hugged by different adults in the target area within a predetermined time period and captured by the front-end camera. That is, there may be one or more monitored objects in the aforementioned set of monitored objects. Generally, adult A is a local resident (such as the aforementioned town T). As long as they have appeared locally once or multiple times (meaning they have been captured by the local front-end camera), and the target platform also has a file for adult A, the family member information of each monitored object in the set of monitored objects can be obtained. For example, this can be done through a third party. The platform searches for the identity information or family member information of adult A. Optionally, upon first obtaining information about the guardian of the first object (e.g., adult A), a list of guardians of the first object can be generated on the target platform. When information about the guardians of the first object (e.g., adult A, B, or others) is subsequently obtained, the information of the guardians is added to the list of guardians. Optionally, this list of guardians can be placed in the file corresponding to the first object. In practical applications, when an image frame containing adult A is first captured, after detecting the image frame to obtain information about adult A, the target platform will create a file about adult A. At this time, the identity information and / or family member information of adult A can be searched through a third-party platform (e.g., the public security system). Optionally, the identity information and / or family member information of adult A can also be stored in the file of adult A. In this way, the family member information of each guardian in a group of guardians can be obtained, resulting in a group of family member information. Each family member in the group of family member information corresponds to the family member information of a guardian. Then, based on the group of family member information, it is determined whether the first object is an object of a specified type.For example, if based on a set of family member information, it is determined that none of the guardians have children, then the first subject can be identified as a specific type of subject, meaning that the first subject is very likely a trafficked child. In this embodiment, family member information of a set of guardians is obtained to determine whether the first subject is a specific type of subject.

[0046] In an optional embodiment, obtaining the family member information of each of the monitored objects in the group of monitored objects includes: searching for files corresponding to each of the monitored objects in the group of monitored objects in an existing file set on the target platform to obtain a set of files, wherein each file in the existing file set stores the family member information of the corresponding object; or sending an information query request to a target device, wherein the information query request carries the information of the group of monitored objects, the information query request is used to request the acquisition of the family member information of each of the monitored objects in the group of monitored objects, the information of the group of monitored objects is the information of each monitored object obtained by object detection of a second set of image frames, the second set of image frames is the image frames of the group of image frames and the image frames in the current image frame in which each monitored object is detected; and acquiring the family member information of each of the monitored objects in the group of monitored objects sent by the target device, wherein the target device is used to find the family member information of each of the monitored objects in the group of monitored objects that matches the information of the group of monitored objects. In this embodiment, files corresponding to each guardian in a group of guardians can be searched in the existing file set on the target platform. Each file in the existing file set stores the family member information corresponding to each object (such as the aforementioned guardian). The family member information stored in each file in the existing file set is obtained from the target device and corresponds to the information of the object stored in each file. Optionally, after determining the files of each guardian in the target platform, an information query request can be sent to the target device to request the acquisition of the family member information of each guardian in the group of guardians. For example, the information query request can carry the information of each guardian. Then, the family member information of each guardian in the group of guardians sent by the target device is acquired. The target device can be a citizen identity information management device, for example, the target device stores the identity information and family member information of all local permanent residents. Through this embodiment, the purpose of acquiring the family member information of the guardian is achieved.

[0047] In an optional embodiment, determining whether the first object is an object of the specified type based on the set of family member information includes: determining the first object as an object of the specified type when none of the family member information includes objects of the predetermined age group; or, determining the first object as an object of the specified type when the family member information includes objects of the predetermined age group but does not include the first object. In this embodiment, when none of the family member information includes objects of the predetermined age group (e.g., children aged 3-7 or 3-5), the first object can be determined to be an object of the specified type; optionally, when the family member information includes objects of the predetermined age group but does not include the first object, the first object can also be determined to be an object of the specified type. This embodiment achieves the purpose of determining whether a first object is an object of the specified type based on the family member information of a group of guardians.

[0048] In an optional embodiment, the method further includes: issuing a warning notification when it is determined that the first object is an object of the specified type, wherein the warning notification is used to indicate that the first object is an object of the specified type. In this embodiment, when it is determined that the first object is an object of the specified type, the target platform can trigger a warning notification, for example, by issuing a warning notification to a third-party platform (such as a public security system or a missing children management platform).

[0049] Obviously, the embodiments described above are only some embodiments of the present invention, and not all embodiments. The present invention will be specifically described below with reference to the embodiments. The following description uses the analysis and determination of whether a child is a trafficked child as an example. This is only an example and does not constitute a limitation on the present solution. The embodiments of the present invention are also applicable to the determination of other objects, such as determining whether an object is an object of a specified type.

[0050] This invention provides a method for assessing the safety of newly arrived children based on facial recognition data, primarily addressing the difficulty of locating missing persons across geographical regions. The method involves screening and assessing newly arrived children in the region, and analyzing and judging both the child and guardian based on information such as the child's guardian's situation and the duration of the child's residence, to determine whether the child is a trafficked child.

[0051] Figure 3 This is an example diagram of the system environment topology according to an embodiment of the present invention, such as... Figure 3As shown, the system includes front-end cameras and a large-scale cloud platform (or cloud platform). The front-end cameras may include multiple cameras, which are mainly used to collect information on children and their companions. The cloud platform mainly performs cluster analysis on children (corresponding to the aforementioned first object) and their companions (corresponding to the aforementioned guardians) to obtain a list of companions for each individual. The system can be deployed throughout the city, and all data collected by the cameras can be recorded by the cloud platform, without being limited by the shooting access of a single camera, and has full coverage.

[0052] The overall concept of this invention is as follows:

[0053] This invention is applicable to regions where the clustering algorithm has been running successfully for a certain period (e.g., one year), and primarily targets the safety assessment of children (e.g., children aged 3-7). The purpose of this solution is mainly to screen whether stable local residents have engaged in child trafficking and to investigate and analyze the safety of newly arrived children in the region.

[0054] The front-end intelligent camera captures all people in the scene, and uses the cloud platform's facial recognition algorithm to create personnel profiles. If a new child's profile is found, it is marked, and the duration of the child's appearance in the profile and the guardian's information are analyzed to determine whether the child has been abducted.

[0055] The embodiments of this invention mainly include three parts of the algorithm, as follows:

[0056] (1) Collection of information on social personnel

[0057] Social surveillance cameras collect facial and body information of individuals, including facial and body attributes such as personal characteristics and age, and provide this information to a cloud platform for image aggregation. After aggregation, each individual's snapshots taken at different times and locations can be stored as natural persons. Simultaneously, for adults within the system, their identities can be traced back to their registered addresses within the public security system, clearly identifying each person's specific information.

[0058] like Figure 4 As shown, the process for collecting information on individuals in the general public includes:

[0059] S402, Collect facial and body features and information of people in the image;

[0060] S404, personnel are landed based on their ID card information.

[0061] (2) Judgment by the child's guardian

[0062] When children aged 3 to 7 go out, their guardians are more likely to hold or carry them. Based on this characteristic, we can define adults who have these two types of physical contact with the child (corresponding to the aforementioned predetermined contact methods) as their guardians, and record the time periods during which the guardians accompany the child. Figure 5 This is a schematic diagram of the judgment of guardians according to a specific embodiment of the present invention. When it is determined that an adult (such as adult A) and a child have the above two types of physical contact, adult A is added to the list of guardians of the child.

[0063] (3) Child safety assessment

[0064] Generally, the appearance of children in another town is mainly due to the following reasons:

[0065] 1) Left-behind children come to the town to visit their parents during holidays;

[0066] 2) Left-behind children return to their parents' homes;

[0067] 3) Traveling, studying, seeking medical treatment, or handling other matters with family;

[0068] 4) During the holiday, stay temporarily at a relative's house;

[0069] 5) Moving with family;

[0070] 6) Trafficked.

[0071] Analyzing the above situations, children in situations 1) and 4) mainly come during winter and summer vacations and will return to their hometowns after the start of the new school year; children in situations 3) and 5) will appear in the new town with their guardians; for situation 2), confirming the information of their parents reveals that they have children under their names, which can distinguish it from situation 6).

[0072] Based on the duration of the child's presence and the circumstances of the guardian, we can analyze whether the child is suspected of being abducted.

[0073] The overall process for determining whether a child is at risk of being abducted is as follows:

[0074] a) The newly generated child profiles (corresponding to the aforementioned target profiles) are marked using the facial recognition system.

[0075] b) For children's files established for more than one month, determine whether the occurrence of such incidents will continue frequently within the next period (e.g., one month). Given the high frequency of children's outings and the complexity of their situations during winter and summer vacations, the determination period will be extended if the incidents occur during winter or summer vacation.

[0076] c) For children who have been missing for an extended period, determine whether their later guardians are local residents. If so, determine if that person / family has any children. If the guardian has no children, then the child is at risk of being abducted.

[0077] Figure 6 This is a flowchart of a child safety early warning system according to a specific embodiment of the present invention. The process includes:

[0078] S602, captures an image of person A. For example, the front-end camera captures an image (or video) of person A and uploads the image (or video) containing person A to the cloud platform.

[0079] S604, Determine whether person A is a child;

[0080] S606, If the judgment result of step S604 above is negative, the process ends;

[0081] S608, if the judgment result of the above step S604 is yes, further determine whether the person A already has a file, that is, determine whether there is a file of person A in the cloud platform; S610, if the judgment result of the above step S608 is no, create a new file, that is, person A may appear in this area for the first time (such as the aforementioned town T). At this time, compare the person A's facial image information with other existing facial image files in the cloud platform. There are no other files of people who are the same as or similar to person A. Therefore, it is necessary to create a new file for person A.

[0082] S612, mark the newly created file of person A as a child to be confirmed, that is, person A is to be confirmed as a trafficked child;

[0083] S614, end the current process. At this point, you can return to step S602 to continue acquiring images (or videos) of person A.

[0084] S616, If the judgment result of the above step S608 is yes, further determine whether the file of person A is a "child file to be confirmed";

[0085] S618, If the judgment result of step S616 is negative, the current judgment process ends. At this time, it means that person A may be a child in this area.

[0086] S620, if the judgment result of step S618 is yes, further determine whether the person A has been appearing for more than a month (or other durations). For example, the cloud platform can determine whether the person A has been appearing in the area for more than a month based on the information of the person A obtained from the images or videos collected by the front-end camera in a certain area (such as town T) and uploaded to the cloud platform in the past period of time. For example, it can be determined based on the difference between the earliest appearance time and the appearance time in the current frame image.

[0087] S622, If the judgment result of step S620 is negative, the process ends; this means that person A may only be in the area briefly, and further judgment can only be made based on subsequent monitoring results.

[0088] S624, if the judgment result of step S620 is yes, further determine whether it is currently winter or summer vacation;

[0089] If the result of step S624 is yes, continue to execute step S624 until the current date is no longer during winter or summer vacation.

[0090] S626, If the judgment result of step S624 is negative, that is, the current date does not belong to the winter or summer vacation, such as October-November, or March-April, or other dates, then the child (i.e., the aforementioned person A) is judged.

[0091] S628, determine whether a child can be matched with a person in the missing children database;

[0092] The database of missing children can be provided by the public security system; alternatively, the cloud platform can query the database of missing children online.

[0093] S630, if the judgment result of the above step S628 is negative, the direct judgment of the child can be terminated, that is, the process of judging whether the child is a trafficked child based on the child's own information can be stopped.

[0094] If the judgment result of the above step S628 is yes, a child safety warning can be triggered. For example, the cloud platform can send a warning to the public security system.

[0095] S632, If the judgment result of the above step S624 is negative, optionally, the guardian of the child (i.e., the above-mentioned person A) can be judged at this time.

[0096] S634, determine whether the guardian (such as guardian G) is a local resident;

[0097] If the judgment result of step S634 above is negative, the process can be terminated;

[0098] S636, If the judgment result of the above step S634 is yes, it can be further determined whether the guardian (such as the above guardian G) has children;

[0099] It should be noted that in the above embodiments, a child (such as person A mentioned above) may correspond to one or more guardians;

[0100] Normally, if the guardian G is a local resident, his or her identity information and family member information, such as whether he or she has children, can be obtained through the public security system.

[0101] S638. If the judgment result of the above step S636 is negative, it can be determined that the above child (i.e., person A) may be at risk of being abducted, and a child safety warning can be triggered at this time.

[0102] In the above embodiments, based on the facial recognition data aggregation function, newly added child files are screened for relevant personnel. A list of child guardians and their accompanying periods is automatically generated based on physical contact between the child and accompanying persons. The guardian list is updated daily based on captured data. By screening child files according to their creation time, large-scale, blanket searches are avoided. Suspicious children are targeted and automatically compared with individuals in the missing children database, significantly reducing the frequency of comparisons and improving their effectiveness.

[0103] Compared with related technologies, the child safety early warning scheme based on clustering analysis adopted in this invention has the following advantages: 1) High targeting: This scheme only makes arrangements and judgments for newly added children in the current area, avoiding the waste of resources from large-scale investigation and comparison. 2) Automatic maintenance of file information: Based on daily social monitoring information, personnel data can be updated in a timely manner, and one child can support multiple guardians. 3) Wide coverage: It is not affected by the field of view of a single camera; as long as a child appears under any networked camera, their information can be recorded. 4) Solving cross-regional problems: This scheme is mainly designed to address the difficulty of locating abducted persons across regions. By analyzing the status of children and guardians, suspicious children can be screened, which can greatly reduce the difficulty of locating and investigating abducted persons.

[0104] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods according to the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods described in the various embodiments of the present invention.

[0105] This embodiment also provides an object determination device. Figure 7 This is a structural block diagram of an object determination device according to an embodiment of the present invention, such as... Figure 7 As shown, the device includes:

[0106] Detection module 702 is used to perform object detection on the current image frame and obtain the current detection result;

[0107] The first determining module 704 is used to determine whether the target platform includes a target file of the first object when the current detection result includes the first information of the first object detected in the current image frame. The first object is an object of a predetermined age group, and the target file is a file obtained by performing a file aggregation operation on the information of the first object within a predetermined time period in the past. The target file stores a set of the first information of the first object obtained within the predetermined time period.

[0108] The module 706 is configured to, when the target platform includes the target file, store the first information into the target file to obtain an updated file, wherein the updated file includes a second information set of the first object, and the second information set includes the first information set and the first information;

[0109] The second determining module 708 is used to determine whether the first object is an object of a specified type based on the updated file.

[0110] In an optional embodiment, the second determining module 708 includes: a first determining submodule, configured to determine a first duration of the first object's appearance in the target area based on the updated archive, wherein the first information set is information about the first object obtained by object detection of a set of image frames, the set of image frames being image frames captured by a target camera device in the target area before the current image frame; and a second determining submodule, configured to determine whether the first object is an object of a specified type if the first duration is greater than or equal to a predetermined duration threshold.

[0111] In an optional embodiment, the first determining submodule includes: a first determining unit, configured to determine the difference between a first time and a second time as the first duration, wherein the first time is the time when the current image frame was captured, the second time is the time when the target image frame in the group of image frames was captured, the target image frame is the earliest captured image frame in the group of image frames, the updated archive stores the first time corresponding to the first information and the second time corresponding to the target information, the target information being the information of the first object obtained by object detection of the target image frame.

[0112] In an optional embodiment, the second determining submodule includes: a comparison unit, configured to compare the second information set of the first object with the information of a group of objects included in a pre-acquired target comparison library to obtain a comparison result, wherein the group of objects included in the target comparison library are all objects of the specified type; and a second determining unit, configured to determine that the first object is an object of the specified type when the comparison result indicates that the similarity between the second information of the second object in the information of the group of objects and some or all of the information in the second information set is greater than or equal to a preset similarity threshold.

[0113] In an optional embodiment, the second determining submodule includes: a first acquiring unit, configured to acquire a set of monitored objects associated with the first object, wherein the set of monitored objects are the set of image frames and objects detected in the current image frame that have a predetermined contact method with the first object; a second acquiring unit, configured to acquire family member information of each of the monitored objects in the set of monitored objects, to obtain a set of family member information; and a third determining unit, configured to determine whether the first object is an object of the specified type based on the set of family member information.

[0114] In an optional embodiment, the second acquisition unit includes: a search subunit, configured to search for files corresponding to each of the monitored objects in the group of monitored objects in an existing file set on the target platform, to obtain a group of files, wherein each file in the existing file set stores family member information of the corresponding object; or, a request subunit, configured to send an information query request to the target device, wherein the information query request carries information of the group of monitored objects, the information query request being used to request the acquisition of family member information of each of the monitored objects in the group of monitored objects, wherein the information of the group of monitored objects is the information of each monitored object obtained by object detection of a second set of image frames, the second set of image frames being the image frames in the group of image frames and the current image frame in which each monitored object is detected; and an acquisition subunit, configured to acquire the family member information of each of the monitored objects in the group of monitored objects sent by the target device, wherein the target device is used to search for family member information of each of the monitored objects in the group of monitored objects that matches the information of the group of monitored objects.

[0115] In an optional embodiment, the third determining unit includes: a first determining subunit, configured to determine the first object as the specified type of object when none of the objects in the set of family member information are objects in the predetermined age range; or, a second determining subunit, configured to determine the first object as the specified type of object when the set of family member information includes objects in the predetermined age range but does not include the first object.

[0116] In an optional embodiment, the above apparatus further includes: an issuing module, configured to issue a warning notification when it is determined that the first object is an object of the specified type, wherein the warning notification is used to indicate that the first object is an object of the specified type.

[0117] It should be noted that the above modules can be implemented by software or hardware. For the latter, they can be implemented in the following ways, but are not limited to: all the above modules are located in the same processor; or, the above modules are located in different processors in any combination.

[0118] Embodiments of the present invention also provide a computer-readable storage medium storing a computer program, wherein the computer program is configured to perform the steps in any of the above method embodiments when executed.

[0119] In one exemplary embodiment, the aforementioned computer-readable storage medium may include, but is not limited to, various media capable of storing computer programs, such as a USB flash drive, read-only memory (ROM), random access memory (RAM), portable hard disk, magnetic disk, or optical disk.

[0120] Embodiments of the present invention also provide an electronic device including a memory and a processor, the memory storing a computer program and the processor being configured to run the computer program to perform the steps in any of the above method embodiments.

[0121] In one exemplary embodiment, the electronic device may further include a transmission device and an input / output device, wherein the transmission device is connected to the processor and the input / output device is connected to the processor.

[0122] Specific examples in this embodiment can be found in the examples described in the above embodiments and exemplary implementations, and will not be repeated here.

[0123] It is obvious to those skilled in the art that the modules or steps of the present invention described above can be implemented using general-purpose computing devices. They can be centralized on a single computing device or distributed across a network of multiple computing devices. They can be implemented using computer-executable program code, and thus can be stored in a storage device for execution by a computing device. In some cases, the steps shown or described can be performed in a different order than those described herein, or they can be fabricated as separate integrated circuit modules, or multiple modules or steps can be fabricated as a single integrated circuit module. Thus, the present invention is not limited to any particular combination of hardware and software.

[0124] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, or improvements made within the principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for determining an object, characterized in that, include: Perform object detection on the current image frame to obtain the current detection result; If the current detection result includes the first information of the first object detected in the current image frame, determine whether the target platform includes a target file of the first object, wherein the first object is an object of a predetermined age group, the target file is a file obtained by performing a file aggregation operation on the information of the first object within a predetermined time period in the past, and the target file stores a set of the first information of the first object obtained within the predetermined time period. If the target platform includes the target file, the first information is stored in the target file to obtain an updated file, wherein the updated file includes a second information set of the first object, and the second information set includes the first information set and the first information; Determine whether the first object is an object of a specified type based on the updated file; The step of determining whether the first object is an object of a specified type based on the updated archive includes: determining a first duration for which the first object appears in the target area based on the updated archive, wherein the first information set is information about the first object obtained by object detection of a set of image frames, the set of image frames being image frames captured by a target camera device in the target area before the current image frame; and determining whether the first object is an object of a specified type when the first duration is greater than or equal to a predetermined duration threshold. Determining whether the first object is an object of a specified type includes: obtaining a set of monitored objects associated with the first object, wherein the set of monitored objects are objects detected in the set of image frames and the current image frame that have a predetermined contact method with the first object; obtaining family member information of each monitored object in the set of monitored objects to obtain a set of family member information; and determining whether the first object is an object of the specified type based on the set of family member information.

2. The method according to claim 1, characterized in that, Determining the first duration of the first object's presence in the target area based on the updated archive includes: The difference between the first moment and the second moment is determined as the first duration, wherein the first moment is the shooting time of the current image frame, the second moment is the shooting time of the target image frame in the group of image frames, the target image frame is the earliest image frame captured in the group of image frames, the updated file stores the first moment corresponding to the first information and the second moment corresponding to the target information, the target information is the information of the first object obtained by object detection of the target image frame.

3. The method according to claim 1, characterized in that, Determining whether the first object is an object of a specified type includes: The second information set of the first object is compared with the information of a group of objects included in the pre-acquired target comparison library to obtain a comparison result, wherein the group of objects included in the target comparison library are all objects of the specified type; If the comparison result indicates that the similarity between the second information of the second object in the set of objects and some or all of the information in the second information set is greater than or equal to a preset similarity threshold, then the first object is determined to be an object of the specified type.

4. The method according to claim 1, characterized in that, The step of obtaining the family member information of each of the monitored persons in the group of monitored persons includes: Search the existing file set on the target platform for files corresponding to each of the guardians in the group of guardians, to obtain a set of files, wherein each file in the existing file set stores the family member information of the corresponding guardian; or A query request is sent to a target device, wherein the query request carries information about the group of monitored objects. The query request is used to request information about the family members of each of the monitored objects in the group of monitored objects. The information about the group of monitored objects is obtained by object detection of each monitored object in a second set of image frames. The second set of image frames consists of the first set of image frames and the current image frame in which each monitored object is detected. The target device then retrieves the family member information of each of the monitored objects in the group of monitored objects sent by the target device, wherein the target device is used to find family member information of each of the monitored objects in the group of monitored objects that matches the information of the first set of monitored objects.

5. The method according to claim 1, characterized in that, Determining whether the first object is an object of the specified type based on the set of family member information includes: If none of the family member information includes individuals within the predetermined age range, then the first individual is determined to be an individual of the specified type; or... If the set of family member information includes objects of the predetermined age group but does not include the first object, then the first object is determined to be an object of the specified type.

6. The method according to any one of claims 1 to 5, characterized in that, The method further includes: If the first object is determined to be an object of the specified type, an early warning notification is issued, wherein the early warning notification is used to indicate that the first object is an object of the specified type.

7. An object determining device, characterized in that, include: The detection module is used to perform object detection on the current image frame and obtain the current detection result; The first determining module is used to determine whether the target platform includes a target file of the first object when the current detection result includes the first information of the first object detected in the current image frame. The first object is an object of a predetermined age group, and the target file is a file obtained by performing a file aggregation operation on the information of the first object within a predetermined time period in the past. The target file stores a set of the first information of the first object obtained within the predetermined time period. The acquisition module is configured to, when the target platform includes the target file, store the first information into the target file to obtain an updated file, wherein the updated file includes a second information set of the first object, and the second information set includes the first information set and the first information; The second determining module is used to determine whether the first object is an object of a specified type based on the updated file; The second determining module includes: a first determining submodule, configured to determine a first duration of the first object's appearance in the target area based on the updated archive, wherein the first information set is information about the first object obtained by object detection of a set of image frames, the set of image frames being image frames captured by a target camera device within the target area before the current image frame; and a second determining submodule, configured to determine whether the first object is an object of a specified type if the first duration is greater than or equal to a predetermined duration threshold. The second determining submodule includes: a first acquiring unit, used to acquire a group of monitored objects associated with the first object, wherein the group of monitored objects are the group of image frames and the objects detected in the current image frame that have a predetermined contact method with the first object; a second acquiring unit, used to acquire family member information of each of the monitored objects in the group of monitored objects, to obtain a group of family member information; and a third determining unit, used to determine whether the first object is an object of the specified type based on the group of family member information.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, wherein the computer program, when executed by a processor, implements the steps of the method described in any one of claims 1 to 6.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method described in any one of claims 1 to 6.

Images