Data processing method and apparatus, analysis system, electronic device, and storage medium

Abstract

A data processing method and apparatus, an analysis system, an electronic device, and a storage medium. The data processing method comprises: on the basis of an association relationship between a plurality of image acquisition devices and a plurality of recognition algorithms, using each recognition algorithm to recognize an image acquired by an associated image acquisition device, so as to obtain a plurality of groups of recognition data respectively corresponding to the plurality of recognition algorithms, wherein configuration information is provided for the association relationship between the plurality of image acquisition devices and the plurality of recognition algorithms; and on the basis of a plurality of statistical rules, performing statistical analysis on the plurality of groups of recognition data to obtain a plurality of pieces of statistical data respectively corresponding to the plurality of statistical rules.

Description

Data processing method and device, analysis system, electronic device, and storage medium TECHNICAL FIELD

[0001] Embodiments of the present disclosure relate to a data processing method, a data processing device, an analysis system, an electronic device, and a computer-readable storage medium. BACKGROUND

[0002] A factory park, a school, or a public place usually installs image collection devices such as cameras, and performs passenger flow analysis based on collected images, for example, counts passenger flow, which can improve the safety management level of the place and provide a decision basis for the operation of the place.

[0003] SUMMARY

[0004] At least one embodiment of the present disclosure provides a data processing method, comprising: based on an association relationship between a plurality of image collection devices and a plurality of recognition algorithms, using each of the recognition algorithms to recognize and process images collected by the associated image collection device, to obtain a plurality of sets of recognition data corresponding to the plurality of recognition algorithms respectively, wherein configuration information is set for the association relationship between the plurality of image collection devices and the plurality of recognition algorithms; based on a plurality of statistical rules, statistically analyzing the plurality of sets of recognition data to obtain a plurality of items of statistical data corresponding to the plurality of statistical rules respectively.

[0005] For example, in the data processing method provided by at least one of the embodiments of the present disclosure, the association relationship between each of the image collection devices and the plurality of recognition algorithms is configured based on algorithm configuration instructions.

[0006] For example, in the data processing method provided by at least one of the embodiments of the present disclosure, the plurality of image collection devices include a first image collection device, and the algorithm configuration instructions include algorithm addition instructions, algorithm deletion instructions, algorithm deactivation instructions, and algorithm activation instructions. Based on the algorithm configuration instructions, the association relationship between each of the image collection devices and the plurality of recognition algorithms is configured, including at least one of the following: in response to the algorithm addition instructions for the first image collection device, associating the first image collection device with a first recognition algorithm; in response to the algorithm deletion instructions for the first image collection device, disassociating the first image collection device from the first recognition algorithm; in response to the algorithm deactivation instructions for the first image collection device, suspending the association relationship between the first image collection device and the first recognition algorithm; and in response to the algorithm activation instructions for the first image collection device, resuming the association relationship between the first image collection device and the first recognition algorithm.

[0007] For example, in the data processing method provided in at least one of the embodiments of the present disclosure, the plurality of image acquisition devices includes a first image acquisition device. Based on the association relationship, the image acquired by the image acquisition device associated with each of the recognition algorithms is identified and processed, including: based on the association relationship, determining at least one target recognition algorithm associated with the first image acquisition device from the plurality of recognition algorithms; and taking the image acquired by the first image acquisition device as the input data of each of the target recognition algorithms.

[0008] For example, in the data processing method provided in at least one of the embodiments of the present disclosure, based on a plurality of statistical rules, the plurality of sets of recognition data are statistically analyzed to obtain a plurality of items of statistical data corresponding to the plurality of statistical rules, including: for each of the statistical rules, selecting the recognition data corresponding to the statistical rule from the plurality of sets of recognition data, and statistically analyzing the selected recognition data to obtain an item of statistical data corresponding to the statistical rule.

[0009] For example, in the data processing method provided in at least one of the embodiments of the present disclosure, each piece of data in the plurality of sets of recognition data includes a device identifier, and the device identifier is used to indicate the image acquisition device from which each piece of data comes; the plurality of statistical rules includes a plurality of sets of statistical rules corresponding to the plurality of recognition algorithms respectively, each set of statistical rules includes at least one statistical rule, and any two statistical rules in the same set of statistical rules are to be statistically analyzed from different two sets of image acquisition devices; the plurality of recognition algorithms includes a first recognition algorithm, the plurality of sets of recognition data includes a first set of recognition data corresponding to the first recognition algorithm, the plurality of statistical rules includes a first statistical rule corresponding to the first recognition algorithm, and the data to be statistically analyzed by the first statistical rule comes from a first set of image acquisition devices. Based on a plurality of statistical rules, the plurality of sets of recognition data are statistically analyzed to obtain a plurality of items of statistical data corresponding to the plurality of statistical rules, including: for the first statistical rule, selecting the recognition data from the first set of image acquisition devices from the first set of recognition data, and statistically analyzing the selected recognition data to obtain an item of statistical data corresponding to the first statistical rule.

[0010] For example, in the data processing method provided in at least one of the embodiments of the present disclosure, each set of statistical rules includes a preset statistical rule, and the data to be statistically analyzed by the preset statistical rule comes from all image acquisition devices associated with the corresponding algorithm; at least one set of statistical rules includes a custom statistical rule, and the data to be statistically analyzed by the custom statistical rule comes from part of the image acquisition devices associated with the corresponding algorithm.

[0011] For example, in the data processing method provided in at least one of the embodiments of the present disclosure, the method further includes: configuring at least one custom statistical rule based on a rule configuration instruction. The rule configuration instruction includes a rule addition instruction, a rule deletion instruction, and a rule replication instruction. Configuring at least one custom statistical rule based on the rule configuration instruction includes: in response to the rule addition instruction, creating a custom statistical rule based on an identification algorithm and a data source corresponding to the rule addition instruction; in response to the rule deletion instruction, deleting a corresponding custom statistical rule; and in response to the rule replication instruction, replicating a corresponding statistical rule and modifying the replicated statistical rule according to modification information.

[0012] For example, in the data processing method provided in at least one of the embodiments of the present disclosure, the method further includes: storing the plurality of sets of identification data in a plurality of message queues respectively; extracting data from the plurality of message queues and storing the data in a plurality of columnar data tables respectively. The statistical analysis of the plurality of sets of identification data based on the plurality of statistical rules to obtain a plurality of items of statistical data corresponding to the plurality of statistical rules respectively includes: extracting data from the plurality of columnar data tables; for each statistical rule, selecting identification data corresponding to the statistical rule from the extracted data, and performing statistical analysis on the selected identification data to obtain an item of statistical data corresponding to the statistical rule.

[0013] For example, in the data processing method provided in at least one of the embodiments of the present disclosure, the method further includes: determining a start time of a plurality of statistical periods based on a configured statistical start time and a period length, and generating a plurality of statistical tasks corresponding to the plurality of statistical rules respectively in each of the periods. The statistical analysis of the plurality of sets of identification data based on the plurality of statistical rules to obtain a plurality of items of statistical data corresponding to the plurality of statistical rules respectively includes: in each of the periods, extracting data corresponding to the period from the plurality of sets of identification data, and performing statistical analysis on the data corresponding to the period based on the plurality of statistical rules to obtain a plurality of items of statistical data corresponding to the period.

[0014] For example, in the data processing method provided in at least one of the embodiments of the present disclosure, the method further includes: forming a plurality of charts based on the plurality of items of statistical data, and displaying the plurality of charts.

[0015] For example, in the data processing method provided in at least one of the embodiments of the present disclosure, the method further includes: in response to a device display instruction, displaying the plurality of image acquisition devices in a tree structure based on a hierarchical relationship of the plurality of image acquisition devices.

[0016] For example, in the data processing method provided in at least one of the embodiments of the present disclosure, the plurality of recognition algorithms comprises at least two of a single-person attribute recognition algorithm, a multi-person attribute recognition algorithm, a residence duration algorithm, a flow calculation algorithm, and a direction algorithm.

[0017] For example, in the data processing method provided in at least one of the embodiments of the present disclosure, the plurality of graphs comprises at least one of a proportion graph, a column graph, a heat map, and a direction graph; the proportion graph is obtained based on statistical analysis of recognition data of the single-person attribute recognition algorithm and / or the multi-person attribute recognition algorithm, and is used to represent a proportion of at least one attribute in all statistical objects within a statistical range; the heat map is obtained based on statistical analysis of recognition data of the residence duration algorithm, and is used to represent residence durations of statistical objects within a plurality of regions within a predetermined time period; the column graph is obtained based on statistical analysis of recognition data of the flow calculation algorithm, and is used to represent changes in flow in and out within a plurality of time periods; and the direction graph is obtained based on statistical analysis of recognition data of the direction algorithm, and is used to represent action directions of statistical objects.

[0018] For example, in the data processing method provided in at least one of the embodiments of the present disclosure, the multi-person attribute recognition algorithm is used to recognize relationships between a plurality of companions, and the recognition result of the multi-person attribute recognition algorithm comprises ages and genders of the plurality of companions; the data processing method further comprises: obtaining the ages and genders of the plurality of companions in each group; in the case of two companions, if the genders of the two companions are different and the ages belong to the same age group, the two companions are recognized as a couple relationship; if the age of one of the plurality of companions belongs to a children age group and the age of another one belongs to another age group, the two companions are recognized as a family relationship; and in the case that the plurality of companions do not belong to the couple relationship or the family relationship, the plurality of companions are recognized as a friend relationship.

[0019] For example, in the data processing method provided in at least one of the embodiments of the present disclosure, the residence duration algorithm is used to determine residence durations of statistical objects within a predetermined time period; the residence duration is determined based on at least one of the following: for statistical objects that enter and exit within the predetermined time period, a leaving time minus an entering time is taken as the residence duration; for statistical objects that enter within the predetermined time period, an end time of the predetermined time period minus the entering time is taken as the residence duration; for statistical objects that exit within the predetermined time period, a leaving time minus a start time of the predetermined time period is taken as the residence duration; and for statistical objects that enter before the predetermined time period and do not exit by the end of the predetermined time period, a duration of the predetermined time period is taken as the residence duration.

[0020] For example, in the data processing method provided in at least one of the embodiments of the present disclosure, the single-person attribute recognition algorithm is used to recognize gender and age; and the flow statistics algorithm is used to determine the number of statistical objects entering and the number of statistical objects leaving in a predetermined time period.

[0021] The data processing apparatus provided in at least one of the embodiments of the present disclosure comprises an identification module and a statistics module. The identification module is configured to perform identification processing on images collected by each image collection device associated with each recognition algorithm based on an association relationship between the image collection devices and the recognition algorithms, to obtain a plurality of sets of identification data corresponding to the plurality of recognition algorithms respectively. Configuration information is set for the association relationship between the image collection devices and the recognition algorithms. The statistics module is configured to perform statistical analysis on the plurality of sets of identification data based on a plurality of statistical rules, to obtain a plurality of items of statistical data corresponding to the plurality of statistical rules respectively.

[0022] For example, in the data processing apparatus provided in at least one of the embodiments of the present disclosure, the apparatus further comprises a chart module configured to form a plurality of charts based on the plurality of items of statistical data.

[0023] For example, in the data processing apparatus provided in at least one of the embodiments of the present disclosure, the apparatus further comprises a data flow module, a data extraction module, and a storage module. The data flow module is configured to store the plurality of sets of identification data in a plurality of message queues respectively. The data extraction module is configured to extract the plurality of sets of identification data from the plurality of message queues. The storage module is configured to store the plurality of sets of identification data extracted by the data extraction module in a plurality of columnar data tables respectively.

[0024] For example, in the data processing apparatus provided in at least one of the embodiments of the present disclosure, the apparatus further comprises a scheduling module configured to determine the start time of a plurality of statistical periods based on a configured statistical start time and a period length, and to generate a plurality of statistical tasks corresponding to the plurality of statistical rules respectively in each of the periods. The statistics module is further configured to extract data corresponding to a current period from the plurality of sets of identification data in each of the periods, and to perform statistical analysis on the data corresponding to the current period based on the plurality of statistical rules, to obtain a plurality of items of statistical data corresponding to the current period.

[0025] The analysis system provided in at least one of the embodiments of the present disclosure comprises a plurality of image collection devices and a data processing apparatus as described in at least one of the embodiments of the present disclosure.

[0026] The electronic device provided in at least one embodiment of the present disclosure comprises a processor, a memory storing one or more computer program modules, wherein the one or more computer program modules are configured to be executed by the processor to implement the data processing method provided in any embodiment of the present disclosure.

[0027] The computer readable storage medium provided in at least one embodiment of the present disclosure stores non-transitory computer readable instructions, which, when executed by a computer, can implement the data processing method provided in any embodiment of the present disclosure. BRIEF DESCRIPTION OF DRAWINGS

[0028] In order to more clearly illustrate the technical solutions of the embodiments of the present disclosure, the drawings of the embodiments will be briefly introduced below. Obviously, the drawings described below only relate to some embodiments of the present disclosure, but not limit the present disclosure.

[0029] FIG. 1 shows a flowchart of a data processing method provided in at least one embodiment of the present disclosure;

[0030] FIG. 2 shows a schematic diagram of an algorithm configuration interface provided in at least one embodiment of the present disclosure;

[0031] FIG. 3 shows a schematic diagram of another algorithm configuration interface provided in at least one embodiment of the present disclosure;

[0032] FIG. 4 shows a schematic diagram of a processing flow provided in at least one embodiment of the present disclosure;

[0033] FIG. 5 shows a schematic diagram of a statistical rule configuration interface provided in at least one embodiment of the present disclosure;

[0034] FIGS. 6 and 7 show schematic diagrams of statistical charts provided in at least one embodiment of the present disclosure;

[0035] FIG. 8 shows a schematic diagram of a tree structure display of a plurality of image acquisition devices provided in at least one embodiment of the present disclosure;

[0036] FIG. 9 shows a schematic diagram of a processing system provided in at least one embodiment of the present disclosure;

[0037] FIG. 10 shows a schematic diagram of another processing flow provided in at least one embodiment of the present disclosure;

[0038] FIG. 11 shows a schematic block diagram of a data processing apparatus provided in at least one embodiment of the present disclosure

[0039] FIG. 12 shows a schematic block diagram of an electronic device provided in at least one embodiment of the present disclosure;

[0040] FIG. 13 shows a schematic block diagram of another electronic device according to at least one embodiment of the present disclosure; and

[0041] FIG. 14 shows a schematic diagram of a computer-readable storage medium according to at least one embodiment of the present disclosure. DETAILED DESCRIPTION

[0042] For the purpose of making the objects, technical solutions and advantages of the embodiments of the present disclosure clearer, the technical solutions of the embodiments of the present disclosure will be described clearly and completely below with reference to the drawings of the embodiments of the present disclosure. Obviously, the described embodiments are only some of the embodiments of the present disclosure, rather than all the embodiments of the present disclosure. Based on the described embodiments of the present disclosure, all other embodiments obtained by those of ordinary skill in the art without creative effort belong to the scope of the present disclosure.

[0043] Unless otherwise defined, technical terms or scientific terms used in the present disclosure should be understood as having the same meaning as commonly understood by one of ordinary skill in the art to which the present disclosure belongs. The terms "first", "second", and similar terms used in the present disclosure do not denote any order, quantity, or importance, but are used to distinguish different components. Similarly, the terms "one", "a", or "the" and similar terms do not denote a quantity restriction, but mean that there is at least one. The terms "include", "comprise", and similar terms mean that the elements or objects before the terms encompass the elements or objects listed after the terms and their equivalents, and do not exclude other elements or objects. The terms "connect" or "connected" and similar terms are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. The terms "upper", "lower", "left", "right", and the like are used only to represent relative positional relationships, and when the absolute positions of the described objects are changed, the relative positional relationships can also be changed accordingly.

[0044] The flow of the existing passenger flow analysis method is single and fixed, for example, a uniform algorithm and analysis are performed on images collected by each camera, which cannot meet different analysis requirements.

[0045] The data processing method, data processing apparatus, analysis system, electronic device, and computer-readable storage medium provided by at least one embodiment of the present disclosure. The data processing method comprises: based on the association relationship between a plurality of image collection devices and a plurality of identification algorithms, using each identification algorithm to identify the image collected by the associated image collection device, to obtain a plurality of sets of identification data corresponding to the plurality of identification algorithms respectively, wherein configuration information is set for the association relationship between the plurality of image collection devices and the plurality of identification algorithms; based on a plurality of statistical rules, statistically analyzing a plurality of sets of identification data to obtain a plurality of statistical data corresponding to a plurality of statistical rules respectively.

[0046] The data processing method can configure the association relationship between the image acquisition device and the recognition algorithm, can make the images collected by each image acquisition device participate in various types of statistical analysis, that is, by configuring the algorithm for the image acquisition device, it can be determined which type of statistics the image of each image acquisition device participates in, so that corresponding configuration can be made for different statistical requirements to obtain the desired statistical result and meet different statistical requirements.

[0047] FIG. 1 shows a flowchart of a data processing method provided by at least one embodiment of the present disclosure.

[0048] As shown in FIG. 1, the method can include steps S110-S120.

[0049] Step S110: Based on the association relationship between the plurality of image acquisition devices and the plurality of recognition algorithms, each recognition algorithm is used to recognize the images collected by the associated image acquisition device to obtain a plurality of sets of recognition data corresponding to the plurality of recognition algorithms respectively. The configuration information is set for the association relationship between the plurality of image acquisition devices and the plurality of recognition algorithms.

[0050] Step S120: Based on a plurality of statistical rules, a plurality of sets of recognition data are statistically analyzed to obtain a plurality of statistical data corresponding to the plurality of statistical rules respectively.

[0051] For example, the image acquisition device can be a camera or other device capable of collecting images, and a plurality of image acquisition devices can be arranged at different positions. The analysis range can be a park, a factory, a school or other place, and the analysis range includes a plurality of areas, such as a plurality of buildings or other activity sites, wherein at least part of the areas can be provided with one or more image acquisition devices.

[0052] For example, in some embodiments, the plurality of recognition algorithms can include at least two of a single-person attribute recognition algorithm, a multi-person attribute recognition algorithm, a residence time algorithm, and a flow calculation algorithm. For example, the single-person attribute recognition algorithm is used to recognize the gender and age of a single statistical object and other single-person attributes, the multi-person attribute recognition algorithm is used to recognize the relationship between a plurality of companions, the residence time algorithm is used to determine the residence time of a statistical object within a predetermined time period, and the flow calculation algorithm is used to determine the number of entering statistical objects and the number of leaving statistical objects. The statistical object may, for example, be a person. Each recognition algorithm can be implemented by means of a neural network or the like.

[0053] For example, in the following embodiments, the single-person attribute recognition algorithm, the multi-person attribute recognition algorithm, the resident time length algorithm, and the traffic calculation algorithm are taken as examples for illustration. However, the present disclosure is not limited thereto, and in other embodiments, the new and old customer recognition algorithm, the customer group trend algorithm, and other algorithms can also be included. The new and old customer recognition algorithm can be used to identify whether each visitor is a new or old customer, and the customer group trend algorithm can be used to identify the action track or destination of each visitor.

[0054] For example, the image acquisition, face recognition, attribute analysis and statistics, and other operations involved in the embodiments of the present disclosure are performed in the case of meeting the relevant legal provisions.

[0055] For example, the association relationship between the plurality of image acquisition devices and the plurality of recognition algorithms can be configured by the user according to the needs, and the images acquired by each image acquisition device can be sent to each recognition algorithm associated with the image acquisition device.

[0056] For example, the data processing method of the embodiments of the present disclosure can further include: based on the algorithm configuration instruction, configuring the association relationship between each image acquisition device and the plurality of recognition algorithms.

[0057] For example, the plurality of image acquisition devices include a first image acquisition device, and the algorithm configuration instruction includes an algorithm addition instruction, an algorithm deletion instruction, an algorithm deactivation instruction, and an algorithm activation instruction. Based on the algorithm configuration instruction, the association relationship between each image acquisition device and the plurality of recognition algorithms is configured, including at least one of the following: in response to the algorithm addition instruction for the first image acquisition device, associating the first image acquisition device with a first recognition algorithm; in response to the algorithm deletion instruction for the first image acquisition device, disassociating the first image acquisition device from the first recognition algorithm; in response to the algorithm deactivation instruction for the first image acquisition device, suspending the association relationship between the first image acquisition device and the first recognition algorithm; and in response to the algorithm activation instruction for the first image acquisition device, resuming the association relationship between the first image acquisition device and the first recognition algorithm. The first image acquisition device can be any image acquisition device in the plurality of image acquisition devices, and the first recognition algorithm can be any recognition algorithm in the plurality of recognition algorithms.

[0058] FIG. 2 shows a schematic diagram of an algorithm configuration interface provided by at least one embodiment of the present disclosure.

[0059] As shown in FIG. 2, the interface displays the device ID corresponding to each camera, the region to which the camera belongs, and the algorithm and other information. For each camera, an algorithm can be added to the camera through a first control (for example, the "add algorithm" control shown in FIG. 2). After the "add algorithm" control is clicked, an algorithm adding interface can be displayed, in which the algorithm to be added can be selected. After the algorithm is added successfully, the interface shown in FIG. 2 is returned, and the newly added algorithm can be displayed in the information bar of the corresponding camera. In addition, when the algorithm is added, the region in which the algorithm acts can also be framed in the picture taken by the camera. After the algorithm takes effect, the acting region can be intercepted from the image taken by the camera, and the corresponding algorithm can be executed on the acting region.

[0060] For example, the algorithm configuration instruction described above can be an instruction input for setting and updating the algorithm in the page corresponding to the "algorithm configuration" shown in FIG. 2 by the user.

[0061] FIG. 3 shows a schematic diagram of another algorithm configuration interface provided by at least one embodiment of the present disclosure.

[0062] As shown in FIGS. 2 and 3, for each camera, the added algorithm can be edited through a second control (for example, the "details" control shown in FIG. 2). For example, after the "details" control is clicked, a right-side drawer pop-up window can be popped up. In the pop-up window, a switch button is provided for each algorithm. Clicking the switch button of any algorithm can enable or disable the algorithm. After the algorithm is disabled, the association between the camera and the algorithm can be temporarily disconnected. The algorithm no longer runs on the camera, and the statistical configuration including the algorithm of the camera suspends counting on the data source. After the algorithm is re-enabled, the association between the camera and the algorithm can be restored. If the start button is clicked and the camera still fails to start after a period of time, or the algorithm is in a fault state, the "restart algorithm" control (not shown in the figure) corresponding to the algorithm can be clicked to reset and start the algorithm. In addition, the algorithm can also be deleted in the pop-up window. The association between the deleted algorithm and the camera is released. If you want to re-associate, you need to add the algorithm to the camera again.

[0063] For example, before step S110, the already configured association relationship can be obtained, including the added algorithms of each camera and the enabled or disabled state of each camera and the added algorithm. In the case where the camera adds an algorithm and the state of the algorithm is enabled, it is considered that the camera is associated with the algorithm.

[0064] For example, taking the first image collection device as an example, step S110 can include: determining at least one target recognition algorithm associated with the first image collection device from a plurality of recognition algorithms based on the association relationship; and taking the image collected by the first image collection device as the input data of each target recognition algorithm.

[0065] Figure 4 shows a schematic diagram of a processing flow provided by at least one embodiment of the present disclosure.

[0066] As shown in Figure 4, multiple cameras, including cameras 001 to 008, each capture images at a predetermined acquisition frequency and input one or more captured images into associated algorithms for recognition processing. For example, camera 001 is associated with a single-person attribute recognition algorithm and a multi-person attribute recognition algorithm. The image captured by camera 001 is copied twice and input into the single-person attribute recognition algorithm and the multi-person attribute recognition algorithm respectively, obtaining single-person attribute data and multi-person attribute data corresponding to the image captured by camera 001. As another example, if camera 008 is associated with a dwell time algorithm and a traffic statistics algorithm, then the image captured by camera 008 can be copied twice and input into the dwell time algorithm and the traffic statistics algorithm respectively, obtaining dwell time data and traffic statistics data corresponding to the image captured by camera 008.

[0067] For example, any recognition algorithm can obtain one recognition data point by processing one image. Each recognition data point can include the data source (such as camera ID), recognition result, time, etc. Each recognition algorithm can obtain a corresponding set of recognition data after processing multiple images.

[0068] For example, step S120 may include: for each statistical rule, selecting the identification data corresponding to the statistical rule from multiple sets of identification data, and performing statistics on the selected identification data to obtain a statistical data item corresponding to the statistical rule.

[0069] For example, multiple statistical rules can include multiple sets of statistical rules corresponding to multiple recognition algorithms, with each set of statistical rules including at least one statistical rule. For example, a single-person attribute recognition algorithm can correspond to a set of single-person statistical rules (such as single-person statistical rules 1-4), and a multi-person attribute recognition algorithm can correspond to a set of multi-person statistical rules (such as multi-person statistical rules 1-2). The statistical rules corresponding to each recognition algorithm can be used to perform statistical analysis on a set of recognition data obtained by the recognition algorithm. For example, single-person statistical rules can be used to analyze and statistically analyze a set of recognition data obtained by a single-person attribute recognition algorithm to obtain a single-person attribute statistical table, and multi-person statistical rules can be used to analyze and statistically analyze a set of recognition data obtained by a multi-person attribute recognition algorithm to obtain a multi-person attribute statistical table.

[0070] For example, the algorithm adding instruction described above can be an instruction input of adding an algorithm by a user in a page corresponding to the "add algorithm" shown in FIG. 2. The algorithm adding instruction can indicate which recognition algorithm to add for the selected camera. The algorithm adding instruction for the camera can trigger the association between the camera and the newly added algorithm. For example, as shown in FIG. 4, the camera 001 has been associated with the single attribute recognition algorithm and the multi-attribute recognition algorithm. The user inputs the algorithm adding instruction for the camera 001, which indicates to add the stay duration algorithm for the camera 001. Then the camera 001 can be associated with the stay duration algorithm. The images captured by the camera 001 can be input to the single attribute recognition algorithm, the multi-attribute recognition algorithm and the stay duration algorithm at the same time. Thus, the images captured by the camera 001 can participate in the single attribute statistical analysis process, the multi-attribute statistical analysis process and the stay duration statistical analysis process.

[0071] For example, the algorithm deleting instruction described above can be an instruction input of deleting an algorithm by a user in the device details pop-up window shown in FIG. 3. The algorithm deleting instruction can indicate which algorithm to delete from the algorithms associated with the selected camera. The algorithm deleting instruction for the camera can trigger the disassociation between the camera and the associated algorithm. For example, as shown in FIG. 4, the camera 001 has been associated with the single attribute recognition algorithm and the multi-attribute recognition algorithm. The user inputs the algorithm deleting instruction for the camera 001, which indicates to delete the multi-attribute recognition algorithm associated with the camera 001. Then the camera 001 can be disassociated with the multi-attribute recognition algorithm. The images captured by the camera 001 are no longer input to the multi-attribute recognition algorithm, so as not to participate in the multi-attribute statistical analysis process.

[0072] For example, the above-mentioned algorithm disable / enable instruction is input by the user in the device details pop-up window shown in FIG. 3. The algorithm disable instruction can indicate which algorithm associated with the selected camera is temporarily disabled, and the algorithm enable instruction can indicate which disabled algorithm is re-enabled. For example, clicking the control 301 shown in FIG. 3 can switch between disabling and enabling the algorithm. Temporary disable means temporarily stopping inputting the image of the camera into the disabled algorithm, and after the algorithm is re-enabled, inputting the image of the camera into the enabled algorithm can be resumed. For example, camera 001 is associated with a single-person attribute recognition algorithm and a multi-person attribute recognition algorithm. The user inputs an algorithm disable instruction for camera 001, which indicates that the multi-person attribute recognition algorithm associated with camera 001 is disabled. Then, inputting the image of camera 001 into the multi-person attribute recognition algorithm can be temporarily stopped, and the multi-person attribute recognition algorithm is temporarily excluded from the multi-person attribute statistical analysis process. When the multi-person attribute recognition algorithm is re-enabled, inputting the image of camera 001 into the multi-person attribute recognition algorithm can be resumed. Compared with algorithm deletion and addition, algorithm disable and enable can more conveniently control the association state between the camera and the algorithm.

[0073] According to the data processing method of the embodiments of the present disclosure, the association relationship between the image acquisition device and the recognition algorithm can be configured, and the images acquired by each image acquisition device can be involved in various types of statistical analysis, that is, by configuring the algorithm for the image acquisition device, it can be determined which type of statistics the image of each image acquisition device participates in. Therefore, corresponding configuration can be performed according to different statistical requirements to obtain the desired statistical result and meet different statistical requirements.

[0074] For example, in some embodiments, the user can also configure statistical rules. As described above, the plurality of statistical rules includes a plurality of groups of statistical rules corresponding to a plurality of recognition algorithms respectively, each group of statistical rules includes at least one statistical rule, and any two statistical rules in the same group of statistical rules are to be counted from different two groups of image acquisition devices, wherein the different two groups of image acquisition devices can mean that the two groups of image acquisition devices are not completely the same. For example, as shown in FIG. 4, the single-person statistical rule group includes single-person statistical rule 1 and single-person statistical rule 2. The data to be counted by single-person statistical rule 1 is derived from, for example, all cameras (such as cameras 001-004) associated with the single-person attribute recognition algorithm, and the data to be counted by single-person statistical rule 2 is derived from, for example, part of the cameras (such as cameras 001 and 002) associated with the single-person attribute recognition algorithm. The user can add statistical rules as needed and set the algorithms and data sources corresponding to the added statistical rules.

[0075] For example, each group of statistical rules can include a preset statistical rule, and the preset statistical rule is configured to count data from all image acquisition devices associated with the corresponding algorithm. At least one group of statistical rules includes a custom statistical rule, and the custom statistical rule is configured to count data from part of the image acquisition devices associated with the corresponding algorithm.

[0076] For example, taking the single-person statistical rule as an example, the single-person statistical rule 1 shown in FIG. 4 can be a preset single-person statistical rule, which is configured to count data from all cameras (e.g., cameras 001-004) associated with the single-person attribute recognition algorithm, and obtain a global statistical result of the single-person attribute. The single-person statistical rules 2-4 can be three custom single-person statistical rules, that is, the three single-person statistical rules can be statistical rules customized by a user according to a range to be counted. For example, all cameras associated with the single-person attribute recognition algorithm are cameras 001-004, cameras 001 and 002 are located on the first floor of building A1, and cameras 001 and 002 are located on the second floor of building A1. When the user wants to obtain a single-person statistical result of the first floor of building A1, a custom single-person statistical rule can be added, and in the custom single-person statistical rule, the statistical range can be set to camera 001 and camera 002. In this way, flexible configuration of the statistical rule can be achieved, and various analysis requirements can be further met.

[0077] The data processing method of the embodiments of the present disclosure can further include: configuring at least one custom statistical rule based on a rule configuration instruction. The rule configuration instruction includes a rule adding instruction, a rule deleting instruction, and a rule copying instruction. Based on the rule configuration instruction, the at least one custom statistical rule is configured, including: in response to the rule adding instruction, creating a custom statistical rule based on the recognition algorithm and the data source corresponding to the rule adding instruction; in response to the rule deleting instruction, deleting the corresponding custom statistical rule; and in response to the rule copying instruction, copying the corresponding statistical rule and modifying the copied statistical rule according to modification information.

[0078] FIG. 5 shows a schematic diagram of a statistical rule configuration interface provided by at least one embodiment of the present disclosure.

[0079] As shown in FIG. 5, the interface can display the name, algorithm, chart type, update time, and the like of each statistical rule. The statistical rule adding page can be opened through the "add statistical rule" control in the interface. In the statistical rule adding page, the name of the newly added rule can be input, the identification algorithm corresponding to the newly added rule can be selected, the data source to be counted, i.e., the camera carrying the corresponding algorithm, can be selected, if the action area of the image collected by the camera needs to be drawn, multiple area boxes to be included in the statistics can be selected, after the data source is selected, the display method, the corresponding data display chart, and the statistical dimension can be selected. After the addition is successful, the newly added statistical rule can be displayed in the interface shown in FIG. 5.

[0080] For example, by clicking the "details" control corresponding to a statistical rule, the details page is entered, and by clicking the "copy this statistical rule" control in the details page, the new statistical rule new page can be jumped to, the configuration properties are the same as those of this statistical rule by default, and the new statistical rule can be submitted after manual modification. By clicking the "delete" control corresponding to a statistical rule, the statistical rule can be deleted. The preset statistical rule can be configured to be non-deletable.

[0081] For example, each piece of data in the multiple sets of identification data includes a device identifier (such as a camera ID), and the device identifier is used to indicate that each piece of data comes from an image collection device. The multiple identification algorithms include a first identification algorithm, the first identification algorithm can be any one of the multiple identification algorithms, the multiple sets of identification data include a first set of identification data corresponding to the first identification algorithm, and the multiple statistical rules include a first statistical rule corresponding to the first identification algorithm. The data to be counted by the first statistical rule comes from the first set of image collection devices. Step S120 can include: for the first statistical rule, selecting identification data from the first set of image collection devices from the first set of identification data, and counting the selected identification data to obtain one item of statistical data corresponding to the first statistical rule.

[0082] For example, taking a single-person attribute identification algorithm as an example, the data to be counted by the first statistical rule comes from cameras 001 and 002, and the source of each piece of data is marked in a set of identification data output by the single-person attribute identification algorithm. Based on the first statistical rule, multiple pieces of data corresponding to the cameras 001 and 002 can be selected from the set of identification data, and the selected multiple pieces of data can be statistically analyzed to obtain statistical data corresponding to the first statistical rule.

[0083] For example, as described above, any identification algorithm processing an image can obtain an identification data, and each identification data can include data source (e.g. camera ID), identification result, time, etc. For a single-person attribute identification algorithm, the identification result can include gender and age range of each person, and when the single-person attribute is counted, the number or proportion of each gender or each age range can be counted.

[0084] For example, the multi-person attribute identification algorithm is used to identify the relationship between multiple companions, and for the multi-person attribute identification algorithm, the identification result can include the age and gender of each group of multiple companions. For example, in the case of two people accompanying each other, the multi-person attribute identification algorithm can obtain the gender and age range of each of the two people. The data processing method can further include: obtaining the age and gender of each of the multiple companions for each group of multiple companions; in the case of two companions, if the gender of the two companions is different and the age belongs to the same age range, the two companions are identified as a couple relationship; if the age of one of the multiple companions belongs to the child age range and the age of the other companion belongs to other age range, the two companions are identified as a family relationship; in the case of multiple companions not belonging to a couple relationship and a family relationship, they are identified as a friend relationship. For example, when the multi-person attribute is counted, the proportion of each relationship can be counted.

[0085] For example, the residence duration algorithm is used to determine the residence duration of the statistical object within a predetermined time period. For the statistical object entering and leaving within the predetermined time period, the residence duration is the leaving time minus the entering time; for the statistical object entering within the predetermined time period, the residence duration is the end time of the time period minus the entering time; for the statistical object leaving within the predetermined time period, the residence duration is the leaving time minus the start time of the time period; for the statistical object entering before the predetermined time period and not leaving by the end of the predetermined time period, the residence duration is the duration of the predetermined time period. For example, when the residence duration is counted, the average residence duration of people in some positions (or areas) within a predetermined duration (e.g. 30 minutes) can be counted.

[0086] For example, the flow calculation algorithm is used to determine the number of entering statistical objects and the number of leaving statistical objects. When the flow is counted, the number of entering people and the number of leaving people in each time period can be counted.

[0087] For example, in some embodiments, the data processing method can further include: based on multiple statistical data, forming multiple charts, and displaying the multiple charts.

[0088] For example, the multiple charts include at least one of a proportion chart, a column chart, a heat map, and a trend chart. The proportion chart is obtained based on statistical analysis of the recognition data of the single-person attribute recognition algorithm and / or the multi-person attribute recognition algorithm, and is used to represent the proportion of at least one attribute in all statistical objects in a statistical range. The heat map is obtained based on statistical analysis of the recognition data of the residence time algorithm, and is used to represent the residence time of the statistical objects in multiple regions within a predetermined time period. The column chart is obtained based on statistical analysis of the recognition data of the traffic calculation algorithm, and is used to represent the change of the in-out traffic in multiple time periods. The trend chart is obtained based on statistical analysis of the recognition data of the trend algorithm, and is used to represent the action direction of the statistical objects.

[0089] FIGS. 6 and 7 show schematic diagrams of statistical charts according to at least one embodiment of the present disclosure.

[0090] As shown in FIGS. 6 and 7, the park panoramic crowd heat map can be generated based on the statistical result of the residence time, and the residence time at each location or region can be presented in the form of hot spots in the heat map. The park single-person visitor portrait can be generated based on the statistical result of the single-person attribute, and the proportion of the gender and the age range of the visitors can be represented in the form of a pie chart. The park multi-person visitor portrait can be generated based on the statistical result of the multi-person attribute, and the proportion of the relationship of the visitors can be represented in the form of a pie chart. The statistical result based on the traffic can be presented in the form of a column chart to show the in-out traffic in different time periods.

[0091] For example, in some embodiments, the data processing method can further include: in response to the device display instruction, displaying the multiple image acquisition devices in a tree structure based on a hierarchical relationship of the multiple image acquisition devices.

[0092] FIG. 8 shows a schematic diagram of multiple image acquisition devices displayed in a tree structure according to at least one embodiment of the present disclosure. As shown in FIG. 8, the cameras can be grouped according to the hierarchical relationship of the regions to which the multiple image acquisition devices belong. For example, the park A includes the building A1 and the apartment A2, the apartment A2 includes multiple regions (for example, the first underground floor, the first floor, the second floor, etc.), the park A can belong to the first level, the building A1 and the apartment A2 both belong to the second level, and each region (for example, the first underground floor, the first floor, the second floor, etc.) of the apartment A2 can be taken as the third level. One or more cameras can be arranged in each region, and the camera(s) can be displayed in the corresponding third level. In this way, the cameras can be managed, and when the cameras are selected, the cameras can be selected in batches and step by step according to the tree structure. The tree structure shown in FIG. 6 is only an example, and other forms of tree structure can also be used.

[0093] For example, in the process of algorithm configuration for the camera, when selecting the camera to be configured, multiple cameras can be displayed in a tree structure so as to quickly select the camera according to the area. In the process of configuring the statistical rule, when selecting the data source for the newly added statistical rule, multiple cameras can also be displayed in a tree structure so as to quickly select the camera to be counted according to the area.

[0094] For example, when selecting the camera, sometimes the selection needs to be combined with the business of the park and the geographical location of the area where the camera is located. If the camera is managed using a common list structure, the geographical information needs to be strictly defined when naming, and the position information needs to be searched when referring. When the number of cameras is large and the number of areas involved is large, it is inconvenient. By using a tree structure with geographical information to manage the camera, the level is selected when naming, and the geographical information does not need to be added. When referring to the camera, the camera can be referred to according to the tree structure level by level and in batches.

[0095] For example, in some embodiments, the data processing method can further include: determining the start time of multiple statistical periods based on the configured statistical start time and period length, and generating multiple statistical tasks corresponding to the multiple statistical rules in each period. Step S120 can include: in each period, extracting data corresponding to the current period from multiple groups of identification data, and based on the multiple statistical rules, statistically analyzing the data corresponding to the current period to obtain multiple statistical data of the current period.

[0096] For example, the statistical start time is from 0 o'clock to 12 o'clock, and the period length is 1 hour. Every 1 hour can be regarded as a statistical period. In each statistical period, the images collected by each camera in the current period can be identified and statistically analyzed to obtain the statistical result of the current period. Taking the statistical period from 0 o'clock to 1 o'clock as an example, in the time period from 0 o'clock to 1 o'clock, multiple images are collected by each camera according to the predetermined collection frequency. The images collected by each camera are input into the associated algorithm to obtain a group of identification data corresponding to each identification algorithm. Based on the configured statistical rule, the multiple groups of identification data are statistically analyzed to obtain the statistical result of the current statistical period. The above-mentioned statistical start time and period length are only examples, which can be set as needed in actual application.

[0097] For example, in some embodiments, the data processing method can further include: storing multiple groups of identification data in multiple message queues respectively; extracting data from the multiple message queues and storing the data in multiple columnar data tables respectively. Step S120 can include: extracting data from the multiple columnar data tables; for each statistical rule, selecting identification data corresponding to the statistical rule from the extracted data, and statistically analyzing the selected identification data to obtain one statistical data corresponding to the statistical rule.

[0098] FIG. 9 shows a schematic diagram of a processing system according to at least one embodiment of the present disclosure.

[0099] As shown in FIG. 9, the recognition result of each algorithm can be reported to the corresponding message queue (for example, kafka), and the big data platform can extract data from each message queue and perform statistical analysis on the extracted data. The big data platform can include a data integration module, a storage module (such as a clickhouse module), a data processing module, and an offline task scheduling module. The data integration module can extract data from each message queue and place it in the storage module, and the storage module stores the data in the form of a columnar data table (clickhouse). For example, the storage module can include a plurality of first configuration tables corresponding to a plurality of recognition algorithms respectively, each configuration table stores all recognition data obtained by the corresponding recognition algorithm, and each piece of recognition data can include information such as data source (such as camera ID), time, etc. When performing statistical analysis based on a certain preset statistical rule, all data of the current period can be extracted from the corresponding first configuration table and the data can be statistically analyzed. The storage module can also be provided with a second configuration table corresponding to each custom statistical rule, and the second configuration table includes information such as the identification (ID) of the custom statistical rule, the statistical range (such as the identification of the camera to be counted or the identification of the area to be counted), etc. When performing statistical analysis based on a certain custom statistical rule, the second configuration table corresponding to the custom statistical rule can be associated with the first configuration table of the corresponding algorithm to find the data to be counted. For example, the range to be counted can be found from the second configuration table corresponding to the custom statistical rule, and then the data corresponding to the statistical range in the current period can be extracted from the first configuration table of the corresponding algorithm.

[0100] FIG. 10 shows a schematic diagram of another processing flow according to at least one embodiment of the present disclosure.

[0101] As shown in FIG. 9 and FIG. 10, the cameras 001-008 are distributed at various locations in the park, and multiple algorithms can be configured for each camera. Taking camera 001 as an example, single-person customer group attribute algorithm and multi-person customer group attribute algorithm are configured. The image captured by the camera is sent to the corresponding kafka queue after being processed by the configured algorithm to generate result data, and the data in kafka is stored in the corresponding data table of clickhouse (data warehouse) through the data integration task of the big data platform. In order to analyze the data in the table, statistical rules in the system can be calculated, and the statistical rules can be divided into preset statistical rules and custom statistical rules. The preset statistical rules can be rules preset by the system, and all data generated by the corresponding algorithm participates in the analysis and calculation of the preset statistical rules. Custom statistical rules can be user-defined statistical rules, and data generated by certain cameras in certain areas can be specified to participate in the analysis and calculation. The results of the analysis and calculation based on the statistical rules are written into the business database through the data processing task of the big data platform, which can be used for subsequent data dashboard display. For example, the above process can be universal and extensible for various recognition algorithms, and a corresponding processing line can be extended based on the above process when other recognition algorithms are needed.

[0102] FIG. 11 shows a schematic block diagram of a data processing apparatus 1100 according to at least one embodiment of the present disclosure.

[0103] For example, as shown in FIG. 11, the data processing apparatus 1100 includes a recognition module 1110 and a statistical module 1120. These components are interconnected by a bus system and / or other forms of connection mechanisms (not shown). For example, these modules can be implemented by hardware (such as circuit) modules, software modules, or any combination of the two, and the following embodiments are the same, which will not be repeated. For example, these units can be implemented by a central processing unit (CPU), a graphics processing unit (GPU), a tensor processing unit (TPU), a field programmable gate array (FPGA), or other forms of processing units with data processing capabilities and / or instruction execution capabilities, and corresponding computer instructions. It should be noted that the components and structures of the data processing apparatus 1100 shown in FIG. 7 are only exemplary and not limiting, and the data processing apparatus 1100 can also have other components and structures as needed.

[0104] The recognition module 1110 is configured to utilize each of the recognition algorithms to perform recognition processing on images collected by the image collection devices associated with the recognition algorithms based on an association relationship between the image collection devices and the recognition algorithms, to obtain multiple sets of recognition data corresponding to the recognition algorithms respectively. Configuration information is set for the association relationship between the image collection devices and the recognition algorithms. The recognition module 1110 can perform step S110 described in FIG. 1, for example.

[0105] The statistical module 1120 is configured to perform statistical analysis on the multiple sets of recognition data based on multiple statistical rules, to obtain multiple items of statistical data corresponding to the multiple statistical rules respectively. The statistical module 1120 may, for example, perform step S120 described in FIG. 1.

[0106] For example, the recognition module 1110 and the statistical module 1120 can be hardware, software, firmware, and any feasible combination thereof. For example, the recognition module 1110 and the statistical module 1120 can be a special-purpose or general-purpose circuit, chip, or device, or a combination of a processor and a memory. The embodiments of the present disclosure do not limit the specific implementation forms of the above-mentioned units.

[0107] For example, the recognition module 1110 and the statistical module 1120 can include codes and programs stored in a memory; and a processor can execute the codes and programs to implement some or all functions of the image recognition module 1110 and the statistical module 1120 as described above. For example, the recognition module 1110 and the statistical module 1120 can be a special-purpose hardware device to implement some or all functions of the recognition module 1110 and the statistical module 1120 as described above. For example, the recognition module 1110 and the statistical module 1120 can be one circuit board or a combination of multiple circuit boards to implement the functions as described above. In the embodiments of the present disclosure, the one circuit board or the combination of multiple circuit boards can include: (1) one or more processors; (2) one or more non-transitory memories connected to the processors; and (3) firmware stored in the memory and executable by the processor.

[0108] It should be noted that in the embodiments of the present disclosure, the units of the data processing apparatus 1100 correspond to the steps of the data processing method described above, and the specific functions of the data processing apparatus 1100 can be referred to the description of the data processing method, which will not be described herein. The components and structures of the data processing apparatus 1100 shown in FIG. 11 are only exemplary and are not limiting, and the data processing apparatus 1100 can also include other components and structures as needed. The data processing apparatus 1100 can include more or fewer circuits or units, and the connection relationship between the circuits or units is not limited, and can be determined according to actual needs. The specific implementation manner of each circuit or unit is not limited, which can be constituted by an analog device according to the circuit principle, or can be constituted by a digital chip, or constituted in other applicable manners.

[0109] For example, the recognition module 1110 can correspond to the “algorithm” module shown in FIG. 9, and the statistical module 1120 can correspond to the “data processing” module shown in FIG. 9.

[0110] For example, in some embodiments, the data processing apparatus can further include a chart module configured to form a plurality of charts based on the plurality of statistical data. For example, the chart module can correspond to the "data board" module shown in FIG. 9.

[0111] For example, in some embodiments, the data processing apparatus can further include a data flow module, a data extraction module, and a storage module. The data flow module is configured to store the plurality of sets of identification data in a plurality of message queues, respectively. For example, the data flow module can correspond to the "message queue" module shown in FIG. 9. The data extraction module is configured to extract the plurality of sets of identification data from the plurality of message queues. For example, the data extraction module can correspond to the "data integration" module shown in FIG. 9. The storage module is configured to store the plurality of sets of identification data extracted by the data extraction module in a plurality of columnar data tables, respectively. For example, the storage module can correspond to the "clickhouse" module shown in FIG. 9.

[0112] For example, in some embodiments, the data processing apparatus can further include a scheduling module configured to determine a start time of each of a plurality of statistical periods based on a configured statistical start time and a period duration, and generate a plurality of statistical tasks corresponding to the plurality of statistical rules, respectively, in each of the periods. For example, the scheduling module can correspond to the "offline scheduling task" module shown in FIG. 9. The statistical module is further configured to extract data corresponding to a current period from the plurality of sets of identification data in each of the periods, and perform statistical analysis on the data corresponding to the current period based on the plurality of statistical rules, to obtain a plurality of items of statistical data of the current period.

[0113] At least one embodiment of the present disclosure further provides an analysis system including a plurality of image acquisition devices and a data processing apparatus as described in at least one embodiment of the present disclosure.

[0114] At least one embodiment of the present disclosure further provides an electronic device including a processor and a memory. The memory stores one or more computer program modules. The one or more computer program modules are configured to be executed by the processor to implement the data processing method described above.

[0115] FIG. 12 is a schematic block diagram of an electronic device according to some embodiments of the present disclosure. As shown in FIG. 12, the electronic device 1200 includes a processor 1210 and a memory 1220. The memory 1220 stores non-transitory computer-readable instructions (e.g., one or more computer program modules). The processor 1210 is configured to execute the non-transitory computer-readable instructions, which, when executed by the processor 1210, perform one or more steps of the data processing method described above. The memory 1220 and the processor 1210 can be interconnected by a bus system and / or other forms of connection mechanisms (not shown). For the specific implementation of each step of the data processing method and related explanations, reference can be made to the embodiments of the data processing method described above, and the repeated parts will not be described herein.

[0116] It should be noted that the components of the electronic device 1200 shown in FIG. 12 are merely exemplary and are not intended to be limiting. The electronic device 1200 can also have other components depending on the actual application needs.

[0117] For example, the processor 1210 and the memory 1220 can communicate with each other directly or indirectly.

[0118] For example, the processor 1210 and the memory 1220 can communicate through a network. The network can include a wireless network, a wired network, and / or any combination of a wireless network and a wired network. The processor 1210 and the memory 1220 can also communicate with each other through a system bus, and the present disclosure does not limit the communication between the processor 1210 and the memory 1220.

[0119] For example, the processor 1210 and the memory 1220 can be disposed on a server side (or a cloud side).

[0120] For example, the processor 1210 can control other components in the electronic device 1200 to perform desired functions. For example, the processor 1210 can be a central processing unit (CPU), a graphics processing unit (GPU), or other forms of processing units having data processing capabilities and / or program execution capabilities. For example, the central processing unit (CPU) can be of X86 or ARM architecture, etc. The processor 1210 can be a general-purpose processor or a special-purpose processor, and can control other components in the electronic device 1200 to perform desired functions.

[0121] For example, the memory 1220 can include any combination of one or more computer program products. The computer program product can include various forms of computer-readable storage media, for example, volatile storage and / or non-volatile storage. The volatile storage can include, for example, random access memory (RAM), cache memory, and / or the like. The non-volatile storage can include, for example, read only memory (ROM), hard disk drives, erasable programmable read only memories (EPROM), compact disk read only memories (CD-ROM), USB memories, flash memories, and / or the like. One or more computer program modules can be stored on the computer-readable storage media. The processor 1210 can execute the one or more computer program modules to implement various functions of the electronic device 1200. Various application programs and various data used and / or generated by the application programs can also be stored in the computer-readable storage media.

[0122] For example, the electronic device 1200 can include a display panel, which can be used to display a configuration interface and a statistical chart. The electronic device 1200 can include a mouse, a keyboard, a touch screen, and / or the like input device to receive user input and configure according to the user input.

[0123] It should be noted that, in the embodiments of the present disclosure, the specific functions and technical effects of the electronic device 1200 can refer to the description of the data processing method in the foregoing description, which will not be described herein.

[0124] FIG. 13 is a schematic block diagram of another electronic device according to some embodiments of the present disclosure. The electronic device 1300 is suitable for implementing the data processing method according to some embodiments of the present disclosure. The electronic device 1300 can be a terminal device, and / or the like. It should be noted that the electronic device 1300 shown in FIG. 13 is merely an example, and should not be construed as limiting the functions and usage range of the embodiments of the present disclosure.

[0125] As shown in FIG. 13, the electronic device 1300 can include a processing device (for example, a central processing unit, a graphics processing unit, and / or the like) 1310, which can perform various appropriate actions and processes according to programs stored in a read only memory (ROM) 1320 or loaded from a storage device 1380 to a random access memory (RAM) 1330. Various programs and data required for the operation of the electronic device 1300 are also stored in the RAM 1330. The processing device 1310, the ROM 1320, and the RAM 1330 are connected to each other through a bus 1340. An input / output (I / O) interface 1350 is also connected to the bus 1340.

[0126] In general, the following devices can be connected to the I / O interface 1350: input devices 1360 including, for example, a touch screen, a touch pad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, and the like; output devices 1370 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, and the like; storage devices 1380 including, for example, a tape, a hard disk, and the like; and communication devices 1390. The communication devices 1390 can allow the electronic device 1300 to communicate wirelessly or wiredly with other electronic devices to exchange data. While FIG. 13 illustrates the electronic device 1300 having various devices, it is understood that all of the illustrated devices are not required to be implemented or possessed, and the electronic device 1300 can instead implement or possess more or fewer devices.

[0127] For example, according to embodiments of the present disclosure, the above-described data processing method can be implemented as a computer software program. For example, embodiments of the present disclosure include a computer program product comprising a computer program carried on a non-transitory computer readable medium, the computer program comprising program code for executing the above-described data processing method. In such embodiments, the computer program can be downloaded and installed from a network through the communication devices 1390, or installed from the storage devices 1380, or installed from the ROM 1320. When the computer program is executed by the processing devices 1310, the functions defined in the data processing method provided by embodiments of the present disclosure can be implemented.

[0128] At least one embodiment of the present disclosure also provides a computer readable storage medium storing non-transitory computer readable instructions which, when executed by a computer, can implement the above-described data processing method.

[0129] FIG. 14 is a schematic diagram of a storage medium according to some embodiments of the present disclosure. As shown in FIG. 14, the storage medium 1400 stores non-transitory computer readable instructions 1410. For example, when the non-transitory computer readable instructions 1410 are executed by a computer, one or more steps of the data processing method according to the above description are performed.

[0130] For example, the storage medium 1400 can be applied in the above-described electronic device 1200. For example, the storage medium 1400 can be the memory 1220 in the electronic device 1200 shown in FIG. 12. For example, the relevant description of the storage medium 1400 can refer to the corresponding description of the memory 1220 in the electronic device 1200 shown in FIG. 12, which will not be described here again.

[0131] The above description merely illustrates the preferred embodiments of the present disclosure and the principles of the technology applied. It should be understood by those skilled in the art that the disclosed scope of the present disclosure is not limited to the technical solutions formed by the specific combinations of the above technical features, and should also cover other technical solutions formed by the combinations of the above technical features or their equivalent features without departing from the above disclosed concept. For example, the technical solutions formed by replacing the above features with the technical features disclosed in the present disclosure (but not limited to) having similar functions.

[0132] In addition, although each operation is described in a particular order, this should not be understood as requiring the operations to be performed in the specific order shown or in a sequential order. In certain circumstances, multitasking and parallel processing can be advantageous. Similarly, although several implementation details are included in the above discussion, these should not be interpreted as limiting the scope of the present disclosure. Certain features described in the context of separate embodiments can also be combined in a single embodiment. Conversely, various features described in the context of a single embodiment can also be separated and implemented in multiple embodiments.

[0133] Although the subject matter has been described in language specific to structural features and / or methodological acts, it is to be understood that the subject defined in the appended claims is not necessarily limited to the specific features or acts described above. Rather, the specific features and acts described above are merely illustrative of example forms of implementing the claims.

[0134] For the present disclosure, the following points need to be explained:

[0135] (1) The drawings of the embodiments of the present disclosure only relate to the structures involved in the embodiments of the present disclosure, and other structures can refer to the general design.

[0136] (2) In the case of no conflict, the embodiments of the present disclosure and the features in the embodiments can be combined to obtain new embodiments.

[0137] The above is merely a specific implementation of the present disclosure, but the protection scope of the present disclosure is not limited thereto, and the protection scope of the present disclosure should be subject to the protection scope of the claims.

Claims

1. A data processing method, comprising: Based on the association between multiple image acquisition devices and multiple recognition algorithms, each recognition algorithm is used to perform recognition processing on the images acquired by the associated image acquisition devices to obtain multiple sets of recognition data corresponding to the multiple recognition algorithms respectively. Configuration information is set for the association between the multiple image acquisition devices and the multiple recognition algorithms. Based on multiple statistical rules, statistical analysis is performed on the multiple sets of identification data to obtain multiple statistical data corresponding to the multiple statistical rules respectively.

2. The data processing method according to claim 1 further includes: Based on the algorithm configuration instructions, the association between each image acquisition device and the plurality of recognition algorithms is configured.

3. The data processing method according to claim 2, wherein, The plurality of image acquisition devices includes a first image acquisition device, and the algorithm configuration instructions include an algorithm add instruction, an algorithm delete instruction, an algorithm disable instruction, and an algorithm enable instruction; Based on algorithm configuration instructions, configure the association between each of the image acquisition devices and the plurality of recognition algorithms, including at least one of the following: In response to an algorithm addition instruction for the first image acquisition device, the first image acquisition device is associated with the first recognition algorithm; In response to the algorithm deletion command for the first image acquisition device, the association between the first image acquisition device and the first recognition algorithm is terminated; In response to the algorithm deactivation command for the first image acquisition device, the association between the first image acquisition device and the first recognition algorithm is suspended; In response to the algorithm enable command for the first image acquisition device, the association between the first image acquisition device and the first recognition algorithm is restored.

4. The data processing method according to claim 2 or 3, wherein, The plurality of image acquisition devices includes a first image acquisition device. Based on the aforementioned association, each of the aforementioned recognition algorithms is used to perform recognition processing on the images acquired by the associated image acquisition devices, including: Based on the association, at least one target recognition algorithm associated with the first image acquisition device is determined from the plurality of recognition algorithms; The images acquired by the first image acquisition device are used as input data for each of the target recognition algorithms.

5. The data processing method according to any one of claims 1-4, wherein, Based on multiple statistical rules, statistical analysis is performed on the multiple sets of identification data to obtain multiple statistical data corresponding to each of the multiple statistical rules, including: For each statistical rule, identification data corresponding to the statistical rule is selected from the multiple sets of identification data, and statistics are performed on the selected identification data to obtain a statistical data item corresponding to the statistical rule.

6. The data processing method according to claim 5, wherein, Each piece of data in the multiple sets of identification data includes a device identifier, which is used to indicate the image acquisition device from which each piece of data comes; The plurality of statistical rules include multiple sets of statistical rules corresponding to the plurality of recognition algorithms respectively. Each set of statistical rules includes at least one statistical rule. The data to be statistically analyzed by any two statistical rules in the same set of statistical rules comes from two different sets of image acquisition devices. The plurality of recognition algorithms include a first recognition algorithm, the plurality of recognition data includes a first set of recognition data corresponding to the first recognition algorithm, the plurality of statistical rules include a first statistical rule corresponding to the first recognition algorithm, and the data to be statistically analyzed by the first statistical rule comes from a first set of image acquisition devices; Based on multiple statistical rules, statistical analysis is performed on the multiple sets of recognition data to obtain multiple statistical data corresponding to the multiple statistical rules, including: for the first statistical rule, selecting recognition data from the first set of image acquisition devices from the first set of recognition data, and performing statistics on the selected recognition data to obtain a statistical data corresponding to the first statistical rule.

7. The data processing method according to claim 6, wherein, Each set of statistical rules includes a preset statistical rule, and the data to be statistically analyzed by the preset statistical rule comes from all image acquisition devices associated with the corresponding algorithm; At least one set of statistical rules includes custom statistical rules, wherein the data to be statistically analyzed by the custom statistical rules comes from a portion of the image acquisition devices associated with the corresponding algorithm.

8. The data processing method according to claim 7 further includes: Configure at least one custom statistical rule based on the rule configuration instructions; The rule configuration instructions include rule addition instructions, rule deletion instructions, and rule copy instructions; Based on the rule configuration instructions, configure at least one custom statistics rule, including: In response to the rule addition instruction, a custom statistical rule is created based on the recognition algorithm and data source corresponding to the rule addition instruction; In response to the rule deletion command, the corresponding custom statistics rule is deleted; In response to the rule copy instruction, the corresponding statistical rule is copied, and the copied statistical rule is modified according to the modification information.

9. The data processing method according to any one of claims 1-8, further comprising: The multiple sets of identification data are stored in multiple message queues respectively; Data is extracted from the multiple message queues and stored in multiple columnar data tables respectively; Based on multiple statistical rules, statistical analysis is performed on the multiple sets of identification data to obtain multiple statistical data corresponding to each of the multiple statistical rules, including: Extract data from the multiple columnar data tables; For each statistical rule, identification data corresponding to the statistical rule is selected from the extracted data, and the selected identification data is statistically analyzed to obtain a statistical data item corresponding to the statistical rule.

10. The data processing method according to any one of claims 1-9, further comprising: Based on the configured statistical start time and period duration, the start time of multiple statistical periods is determined, and multiple statistical tasks corresponding to the multiple statistical rules in each period are generated. Specifically, based on multiple statistical rules, statistical analysis is performed on the multiple sets of identification data to obtain multiple statistical data corresponding to each of the multiple statistical rules, including: In each cycle, data corresponding to the current cycle is extracted from the multiple sets of identification data, and statistical analysis is performed on the data corresponding to the current cycle based on the multiple statistical rules to obtain multiple statistical data for the current cycle.

11. The data processing method according to any one of claims 1-10, further comprising: Based on the aforementioned statistical data, multiple charts are generated and displayed.

12. The data processing method according to any one of claims 1-11, further comprising: In response to a device display command, the multiple image acquisition devices are displayed in a tree structure based on their hierarchical relationship.

13. The data processing method according to claim 11 or 12, wherein, The multiple identification algorithms include at least two of the following: single-person attribute identification algorithm, multi-person attribute identification algorithm, dwell time algorithm, traffic statistics algorithm, and trend algorithm.

14. The data processing method according to claim 13, wherein, The plurality of charts includes at least one of percentage charts, bar charts, heatmaps, and trend charts; The proportion chart is obtained by statistical analysis of the identification data of single-person attribute recognition algorithm and / or multi-person attribute recognition algorithm. The proportion chart is used to represent the proportion of at least one attribute in all statistical objects within the statistical range. The heatmap is obtained by statistical analysis of the identification data of the dwell time algorithm, and the heatmap is used to characterize the dwell time of the statistical object in multiple areas within a predetermined time period; The bar chart is obtained by statistical analysis of the identification data of the traffic statistics algorithm, and the bar chart is used to represent the changes in inflow and outflow over multiple time periods. The movement graph is obtained by statistical analysis of the identification data of the movement algorithm, and the movement graph is used to characterize the movement direction of the statistical object.

15. The data processing method according to claim 13 or 14, wherein, The multi-person attribute recognition algorithm is used to identify the relationship between multiple companions, and the recognition result of the multi-person attribute recognition algorithm includes the age and gender of each of the multiple companions; The data processing method further includes: For each group of multiple partners, obtain the age and gender of each of the multiple partners; If there are two companions, and the two companions are of different genders and belong to the same age group, then the two companions are identified as partners. If one of the multiple companions is a child and the other is in another age group, then the two companions are identified as family members. In cases where multiple companions do not belong to the stated partnership or family relationship, they are identified as friends.

16. The data processing method according to any one of claims 13-15, wherein, The dwell time algorithm is used to determine the dwell time of the statistical objects within a predetermined time period; The duration of stay is determined based on at least one of the following: For the statistical objects that enter and exit within the predetermined time period, the departure time minus the entry time is used as the dwell time; For statistical objects that enter within the predetermined time period, the dwell time is calculated by subtracting the entry time from the end of the time period. For statistical subjects who leave within the predetermined time period, the departure time minus the start time of the time period is taken as the length of stay. For statistical objects that enter before the predetermined time period and do not leave by the end of the predetermined time period, the duration of the predetermined time period is taken as the dwell time.

17. The data processing method according to any one of claims 13-16, wherein, The single-person attribute recognition algorithm is used to identify gender and age; The traffic statistics algorithm is used to determine the number of statistical objects entering and leaving within a predetermined time period.

18. A data processing apparatus, comprising: The recognition module is configured to perform recognition processing on images acquired by the associated image acquisition devices based on the association relationship between multiple image acquisition devices and multiple recognition algorithms, thereby obtaining multiple sets of recognition data corresponding to the multiple recognition algorithms respectively. Configuration information is set for the association relationship between the multiple image acquisition devices and the multiple recognition algorithms. The statistics module is configured to perform statistical analysis on the multiple sets of identified data based on multiple statistical rules, and obtain multiple statistical data corresponding to the multiple statistical rules respectively.

19. The data processing apparatus according to claim 18, further comprising: The chart module is configured to generate multiple charts based on the aforementioned statistical data.

20. The data processing apparatus according to claim 18 or 19, further comprising: The data stream module is configured to store the multiple sets of identification data in multiple message queues (the disclosure document Kafka); The data extraction module is configured to extract the multiple sets of identification data from the multiple message queues; The storage module is configured to store the multiple sets of identification data extracted by the data extraction module into multiple columnar data tables (clickhouse).

21. The data processing apparatus according to claim 20, further comprising: The scheduling module is configured to determine the start time of multiple statistical periods based on the configured statistical start time and period duration, and generate multiple statistical tasks in each period that correspond to the multiple statistical rules respectively. The statistics module is further configured to extract data corresponding to the current period from the multiple sets of identification data in each period, and perform statistical analysis on the data corresponding to the current period based on the multiple statistical rules to obtain multiple statistical data for the current period.

22. An analysis system, comprising: Multiple image acquisition devices; The data processing apparatus as described in any one of claims 18-21.

23. An electronic device, comprising: processor; Memory, which stores one or more computer program modules; The one or more computer program modules are configured to be executed by the processor to implement the data processing method according to any one of claims 1-17.

24. A computer-readable storage medium storing non-transitory computer-readable instructions that, when executed by a computer, can implement the data processing method according to any one of claims 1-17.

Images