A customer screening and advertisement recommendation method and system based on data analysis

Abstract

The present application relates to a customer screening and advertisement recommendation method and system based on data analysis. The method includes: in response to the acquired real-time image, parsing the real-time image to obtain the person objects included in the real-time image; grouping the person objects included in the real-time image to obtain person object groups, and the person objects in each person object group have an association relationship; acquiring the movement trajectories of the person object groups and generating analysis data according to the movement trajectories; using the analysis data to generate an advertisement recommendation scheme associated with the person object groups; acquiring the new movement trajectories of the person object groups and implementing the advertisement recommendation scheme on the display terminals associated with the new movement trajectories. The customer screening and advertisement recommendation method and system disclosed in the present application can achieve screening and recommendation through targeted analysis of customer groups. This method does not need to rely on historical data and can better balance privacy and targeted content recommendation.

Description

Technical Field

[0001] This application relates to the technical field of data processing, and in particular, to a method and system for customer screening and advertisement recommendation based on data analysis. Background Art

[0002] Customer screening and advertisement recommendation are two core links in precision marketing and recommendation systems. The former is used to identify high-value or potential target customers, and the latter pushes advertisements or products that meet their needs to these customers through algorithm models. For example, in content browsing, targeted advertisement recommendations are made based on the customer's viewing content and consumption habits, etc. These advertisements are naturally related to the customer's preferences and have better conversion rates.

[0003] In the shopping mall consumption scenario, most current advertisement placement methods still rely on bidding. One advantage of this method is to ensure the profit of the shopping mall, but there are natural defects in conversion and customer experience because the bidding method ignores key factors such as quality and customer matching.

[0004] A better way is to make targeted recommendations based on customers, but due to practical influencing factors such as customer privacy and the complexity of customer sources, it is difficult to achieve objective and real data records. Summary of the Invention

[0005] This application provides a method and system for customer screening and advertisement recommendation based on data analysis, which can achieve screening and recommendation through targeted analysis of customer groups. This method does not need to rely on historical data and can better balance privacy and targeted content recommendation.

[0006] The above object of this application is achieved through the following technical solutions:

[0007] In a first aspect, this application provides a method for customer screening and advertisement recommendation based on data analysis, including:

[0008] In response to the acquired real-time image, parse the real-time image to obtain the person objects included in the real-time image;

[0009] Group the person objects included in the real-time image to obtain person object groups, and the person objects in each person object group have an association relationship;

[0010] Obtain the movement trajectories of the person object groups and generate analysis data according to the movement trajectories, and each person object group corresponds to one analysis data;

[0011] Use the analysis data to generate an advertisement recommendation plan associated with the person object group;

[0012] Obtain the new movement trajectories of the group of human objects and implement an advertising recommendation scheme on the display terminals associated with the new movement trajectories.

[0013] In a possible implementation manner of the first aspect, grouping the human objects included in the real-time image includes:

[0014] Using the human objects included in the real-time image to establish an analysis reference network, and each grid in the analysis reference network has the same number of sides;

[0015] Obtain the size change degree of each grid in the analysis reference network in the time series;

[0016] When the size change degree of the grid exceeds the allowable range, perform extinction processing on the grid until the number of grids in the analysis reference network is less than or equal to the set number;

[0017] Calculate the sparse area of the analysis reference network and perform preliminary partitioning on the analysis reference network.

[0018] In a possible implementation manner of the first aspect, calculating the sparse area of the analysis reference network and performing preliminary partitioning on the analysis reference network includes:

[0019] Calculate the line segment length of each line segment in the analysis reference network, the average line segment length, the grid area of each grid, and the average grid area;

[0020] Mark the first range in the analysis reference network according to the line segment length and the average line segment length;

[0021] Mark the second range in the analysis reference network according to the grid area and the average grid area;

[0022] Determine the overlapping area of the first range and the second range, denoted as the overlapping range;

[0023] Use the overlapping range, the remainder of the first range, and the remainder of the second range to perform preliminary partitioning on the analysis reference network.

[0024] In a possible implementation manner of the first aspect, obtaining the movement trajectories of the group of human objects and generating analysis data according to the movement trajectories includes:

[0025] Create movement trajectories based on the movement of the group of human objects. The movement trajectories include line segments connected in sequence and points located on the line segments. The line segments have length and direction, and the points have area. The area of the points is negatively correlated with the movement speed;

[0026] Determine the content type associated with the points;

[0027] Generate analysis data according to the content type associated with the points. The analysis data includes content type, content type association time, and content type ratio.

[0028] In a possible implementation of the first aspect, it further includes determining the weight of each person object in the group of person objects and adjusting the advertisement recommendation scheme according to the weight of the person object.

[0029] In a possible implementation of the first aspect, determining the weight of each person object in the group of person objects includes:

[0030] Calculating the degree of change of each grid in the analysis reference network, where the degree of change includes the amount of size change and the change speed;

[0031] Sorting the person objects in the analysis reference network according to the degree of change, and the weights of the person objects on the sequential sequence decrease in turn.

[0032] In a possible implementation of the first aspect, it further includes correcting the sorting of person objects using the movement trajectory. Correcting the sorting of person objects using the movement trajectory includes:

[0033] Obtaining the area and attributes of all points on the movement trajectory;

[0034] Grouping all points according to the attributes of the points and calculating the total area of the points in each group to obtain the grouped values;

[0035] Using the grouped values to correct the sorting of person objects on the sequential sequence.

[0036] In the second aspect, the present application provides a customer screening and advertisement recommendation device based on data analysis, including:

[0037] An image analysis unit, configured to analyze the real-time image in response to the acquired real-time image to obtain the person objects included in the real-time image;

[0038] A grouping processing unit, configured to group the person objects included in the real-time image to obtain a group of person objects, and the person objects in each group of person objects have an association relationship;

[0039] A data generation unit, configured to obtain the movement trajectory of the group of person objects and generate analysis data according to the movement trajectory, and each group of person objects corresponds to one analysis data;

[0040] A scheme generation unit, configured to generate an advertisement recommendation scheme associated with the group of person objects using the analysis data;

[0041] A scheme implementation unit, configured to obtain the new movement trajectory of the group of person objects and implement the advertisement recommendation scheme on the display terminal associated with the new movement trajectory.

[0042] In a third aspect, the present application provides a customer screening and advertisement recommendation system based on data analysis, and the system includes:

[0043] One or more memories for storing instructions; and one or more processors for calling and running the instructions from the memory to execute the method described in the first aspect and any possible implementation manner of the first aspect.

[0044] In a fourth aspect, the present application provides a computer-readable storage medium, and the computer-readable storage medium includes:

[0045] A program, when the program is run by a processor, the method described in the first aspect and any possible implementation manner of the first aspect is executed.

[0046] In a fifth aspect, the present application provides a computer program product, including program instructions, when the program instructions are run by a computing device, the method described in the first aspect and any possible implementation manner of the first aspect is executed.

[0047] In a sixth aspect, the present application provides a chip system, and the chip system includes a processor for implementing the functions involved in the above aspects, for example, generating, receiving, sending, or processing the data and / or information involved in the above method.

[0048] The chip system may be composed of chips or may include chips and other discrete devices.

[0049] In a possible design, the chip system further includes a memory for storing necessary program instructions and data. The processor and the memory may be decoupled and disposed on different devices and connected by a wired or wireless manner, or the processor and the memory may also be coupled on the same device. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 is a schematic flow chart of the steps of a method for customer screening and advertisement recommendation based on data analysis provided by the present application.

[0051] Figure 2 is a schematic diagram of grouping a person object for grouping provided by the present application.

[0052] Figure 3 is a schematic diagram of implementing an advertisement recommendation scheme on a newly added movement trajectory provided by the present application.

[0053] Figure 4 is a schematic diagram of the process of obtaining an analysis reference network provided by the present application.

[0054] Figure 5It is a schematic diagram of the process of decomposing an analysis reference network provided by this application.

[0055] Figure 6 It is a schematic diagram of a movement trajectory provided by this application. Specific embodiments

[0056] The following further elaborates on the technical solutions in this application in conjunction with the accompanying drawings.

[0057] This application discloses a method for customer screening and advertisement recommendation based on data analysis. Please refer to Figure 1 , in some examples, the method for customer screening and advertisement recommendation based on data analysis disclosed in this application includes the following steps:

[0058] S101, in response to the acquired real-time image, parse the real-time image to obtain the human objects included in the real-time image;

[0059] S102, group the human objects included in the real-time image to obtain groups of human objects, and the human objects in each group of human objects have an association relationship;

[0060] S103, obtain the movement trajectory of the group of human objects and generate analysis data according to the movement trajectory, and each group of human objects corresponds to one analysis data;

[0061] S104, use the analysis data to generate an advertisement recommendation plan associated with the group of human objects;

[0062] S105, obtain the new movement trajectory of the group of human objects and implement the advertisement recommendation plan on the display terminal associated with the new movement trajectory.

[0063] In step S101, according to the received real-time image, parse the real-time image to obtain the human objects included in the real-time image. Here, the real-time image refers to the monitoring image in the shopping mall, and the monitoring image covers most or all areas of the shopping mall.

[0064] The human objects included in the real-time image refer to the people who enter the shopping mall. The purpose of parsing is to determine the people who enter the shopping mall and their numbers.

[0065] Then, in step S102, group the human objects included in the real-time image. As Figure 2 shown, obtain groups of human objects, and the human objects in each group of human objects have an association relationship. Here, the association relationship means that when the human objects enter the area corresponding to the real-time image, there is a potential possibility of walking together. The corresponding relationship of walking together may be various, such as friends, family members, colleagues, etc.

[0066] In step S103, the movement trajectories of the group of person objects are obtained and analysis data is generated based on the movement trajectories. When the number of groups of person objects is multiple, the number of analysis data is also multiple, and each group of person objects corresponds to one analysis data.

[0067] Then, in step S104, the analysis data is used to generate an advertisement recommendation scheme associated with the group of person objects. Finally, in step S105, the new movement trajectories of the group of person objects are obtained and the advertisement recommendation scheme is implemented on the display terminal associated with the new movement trajectories, as Figure 3 shown.

[0068] The technical solution provided by this application analyzes by grouping the person objects entering the corresponding area of the real-time image, and then implements the advertisement recommendation scheme according to the grouping. During the implementation of the advertisement recommendation scheme, the new movement trajectories are also referred to. The new movement trajectories refer to the new movement trajectories generated during the movement of the group of person objects.

[0069] That is, during the movement of the group of person objects, advertisement recommendation is carried out based on the movement trajectories of the group of person objects. This recommendation method changes with the change of the movement trajectories of the group of person objects, rather than simply relying on tagging the person objects or the group of person objects for single-type recommendation, but through the movement of the group of person objects for recommendation.

[0070] Or it can be described as that the advertisement recommendation method provided by this application is based on the current movement of the group of person objects. Because for the group of person objects, their current hobbies may be unpredictable or in a changing state, and the recommendation method by tagging will make the advertisement recommendation solidified.

[0071] In some examples, the specific method for grouping the person objects included in the real-time image is as follows:

[0072] S201, establishing an analysis reference network using the person objects included in the real-time image, and each grid in the analysis reference network has the same number of sides;

[0073] S202, obtaining the size change degree of each grid in the analysis reference network in the time series;

[0074] S203, when the size change degree of the grid exceeds the allowable range, performing extinction processing on the grid until the number of grids in the analysis reference network is less than or equal to the set number;

[0075] S204, calculating the sparse area of the analysis reference network and pre-partitioning the analysis reference network.

[0076] In steps S201 to S203, an analysis reference network will be first established using the person objects included in the real-time image. As Figure 4 shown, the grid obtained here is a basic analysis reference network, and then the processing of the analysis reference network begins.

[0077] The specific processing process is to obtain the dimensional change degree of each grid in the analysis reference network in the time series, and then when the dimensional change degree of the grid exceeds the allowable range, the grid is subjected to extinction processing until the number of grids in the analysis reference network is less than or equal to the set number.

[0078] Subjecting the grid to extinction processing means that the analysis reference network is decomposed into multiple parts. The corresponding scenario here is that when first entering the mall, the person objects are relatively concentrated and the grouping processing described above cannot be directly carried out. However, in order to ensure the continuity of the movement trajectory and the accumulation of previous data, it is necessary to start the analysis from when the person objects first enter the mall.

[0079] The dimensional change degree of the grid refers to the side length and / or area of the grid. When one of the side lengths of the grid or the area of the grid exceeds the set allowable value, it is determined that the grid is to be extinguished (deleted), corresponding to the grouping processing of the person objects described above.

[0080] In this step, a small analysis reference network may be obtained. At this time, the number of personnel corresponding to the analysis reference network includes one person (point), two people (line segment), and multiple people (grid), as Figure 5 shown.

[0081] For the case of one person (point) and two people (line segment), the subsequent processes of generating an advertisement recommendation plan and implementing the advertisement recommendation plan on the display terminal associated with the newly added movement trajectory can be directly carried out. However, for the case of multiple people (grid), further processing is still required.

[0082] The specific processing process is carried out in step S204, and the process is to calculate the sparse area of the analysis reference network and pre-divide the analysis reference network.

[0083] The specific method for calculating the sparse area of the analysis reference network and pre-dividing the analysis reference network is as follows:

[0084] S301, calculate the line segment length of each line segment in the analysis reference network, the average line segment length, the grid area of each grid, and the average grid area;

[0085] S302, mark the first range in the analysis reference network according to the line segment length and the average line segment length;

[0086] S303. Mark the second range in the analysis reference network according to the grid area and the average area of the grids.

[0087] S304. Determine the overlapping area of the first range and the second range, and denote it as the overlapping range.

[0088] S305. Use the overlapping range, the remaining part of the first range, and the remaining part of the second range to pre-divide the analysis reference network.

[0089] The content in steps S301 to S305 pre-divides the analysis reference network according to the average length and the average area, that is, divides the analysis reference network into three parts: the area where both the average area and the average length meet, the area where either the average area or the average length meets, and the area where neither the average area nor the average length meets.

[0090] This method can highlight the key areas and non-key areas of the analysis reference network. During the subsequent investigation process, the connections of these three parts will be focused on, rather than conducting a comprehensive analysis of the entire analysis reference network. The advantage of this local analysis is that it can focus on the key points.

[0091] In some examples, the specific method for obtaining the movement trajectories of the group of person objects and generating analysis data based on the movement trajectories is as follows:

[0092] Create movement trajectories based on the movement of the group of person objects. Figure 6 As shown, the movement trajectories include sequentially connected line segments and points located on the line segments. The line segments have length and direction, and the points have area. The area of the points is negatively correlated with the movement speed.

[0093] Determine the content type associated with the points.

[0094] Generate analysis data according to the content type associated with the points. The analysis data includes the content type, the content type association time, and the content type ratio.

[0095] In this method, it is mainly considered what content the group of person objects pays attention to during the movement process. The movement process is represented by line segments, and the staying time is represented by points. The longer the staying time, the more interested the group of person objects is in the content at the position of the point, or it can be described as willing to spend time on the content at the position of the point.

[0096] Finally, generate analysis data according to the content type associated with the points. The analysis data includes the content type, the content type association time, and the content type ratio.

[0097] In some examples, it is also added to determine the weight of each person object in the group of person objects and adjust the advertisement recommendation scheme according to the weight of the person object. The method for determining the weight of each person object in the group of person objects is as follows:

[0098] S401. Calculate the degree of change of each grid in the analysis reference network. The degree of change includes the amount of size change and the speed of change.

[0099] S402. Sort the person objects in the analysis reference network according to the degree of change. The weights of the person objects in the sequence decrease in turn.

[0100] In steps S401 to S402, the degree of change of each grid in the reference network is mainly examined. The scenario corresponding to the degree of change of the grid is the activity of the person object. The higher the activity of the person object, the higher its weight.

[0101] That is to say, in this part of the content, the needs or interests of the person objects in the active state will be mainly considered. For the situation of one person (point), this part of the content is not involved; for the situation of two people (line segment), the degrees of change of the two endpoints at both ends are directly examined. At this time, the degree of change includes two parameters: the moving speed and the moving amount.

[0102] Furthermore, it is also necessary to consider using the moving trajectory to correct the sorting of the person objects. The specific method of using the moving trajectory to correct the sorting of the person objects is as follows:

[0103] S501. Obtain the area and the attributes of all the points on the moving trajectory.

[0104] S502. Group all the points according to the attributes of the points and calculate the total area of the points in each group to obtain the grouped value.

[0105] S503. Use the grouped value to correct the sorting of the person objects in the sequence.

[0106] The purpose of using the moving trajectory to correct the sorting of the person objects is to reduce the weight of the grid change degree, because the grid change degree is only one aspect. At the same time, it is also necessary to consider the actual moving route and the time spent on the moving route.

[0107] In the specific process of correction, the obtained grouped value is a value less than 1 (the ratio of the total area of the points in the group to the total area of all the points). Multiply this value by the sorting of the person objects (1, 2, 3, and so on) to obtain a value, and arrange them in descending order to obtain a new sequence.

[0108] The above process is carried out dynamically. For example, reprocessing is performed every five minutes or every ten minutes. The main purpose of reprocessing is to consider the weight change of the person objects and the dynamic change of the moving route. In a complete process, the interests of the person objects in the active state in the person object group and the entire person object group will change.

[0109] The present application also provides a customer screening and advertisement recommendation device based on data analysis, including:

[0110] An image parsing unit, configured to parse a real-time image in response to the acquired real-time image, and obtain a human object included in the real-time image;

[0111] A grouping processing unit, configured to perform grouping processing on the human objects included in the real-time image to obtain a human object group, and the human objects in each human object group have an association relationship;

[0112] A data generation unit, configured to obtain the movement trajectory of the human object group and generate analysis data according to the movement trajectory, and each human object group corresponds to one analysis data;

[0113] A solution generation unit, configured to generate an advertisement recommendation solution associated with the human object group by using the analysis data;

[0114] A solution implementation unit, configured to obtain the new movement trajectory of the human object group and implement the advertisement recommendation solution on a display terminal associated with the new movement trajectory.

[0115] Further, performing grouping processing on the human objects included in the real-time image includes:

[0116] Using the human objects included in the real-time image to establish an analysis reference network, and each grid in the analysis reference network has the same number of sides;

[0117] Obtaining the size change degree of each grid in the analysis reference network in the time series;

[0118] When the size change degree of the grid exceeds the allowable range, performing extinction processing on the grid until the number of grids in the analysis reference network is less than or equal to the set number;

[0119] Calculating the sparse area of the analysis reference network and pre-partitioning the analysis reference network.

[0120] Further, calculating the sparse area of the analysis reference network and pre-partitioning the analysis reference network includes:

[0121] Calculating the line segment length of each line segment in the analysis reference network, the average line segment length, the grid area of each grid, and the average grid area;

[0122] Marking a first range in the analysis reference network according to the line segment length and the average line segment length;

[0123] Marking a second range in the analysis reference network according to the grid area and the average grid area;

[0124] Determining the overlapping area of the first range and the second range, denoted as the overlapping range;

[0125] Pre-divide the analysis reference network using the coincidence range, the remainder of the first range, and the remainder of the second range.

[0126] Furthermore, obtaining the movement trajectories of the group of person objects and generating analysis data based on the movement trajectories includes:

[0127] Create movement trajectories based on the movement of the group of person objects. The movement trajectories include line segments connected in sequence and points located on the line segments. The line segments have length and direction, and the points have area. The area of the points is negatively correlated with the movement speed;

[0128] Determine the content type associated with the points;

[0129] Generate analysis data according to the content type associated with the points. The analysis data includes content type, content type association time, and content type ratio.

[0130] Furthermore, it also includes determining the weight of each person object in the group of person objects and adjusting the advertisement recommendation plan according to the weight of the person objects.

[0131] Furthermore, determining the weight of each person object in the group of person objects includes:

[0132] Calculate the change degree of each grid in the analysis reference network. The change degree includes the size change amount and the change speed;

[0133] Sort the person objects in the analysis reference network according to the change degree. The weights of the person objects on the sequential sequence decrease in turn.

[0134] Furthermore, it also includes correcting the sorting of person objects using the movement trajectories. Correcting the sorting of person objects using the movement trajectories includes:

[0135] Obtain the area and attributes of all the points on the movement trajectories;

[0136] Group all the points according to the attributes of the points and calculate the total area of the points in each group to obtain the grouped values;

[0137] Use the grouped values to correct the sorting of the person objects on the sequential sequence.

[0138] In one example, the units in any of the above devices may be one or more integrated circuits configured to implement the above methods. For example: one or more application specific integrated circuits (ASICs), or one or more digital signal processors (DSPs), or one or more field programmable gate arrays (FPGAs), or a combination of at least two of these integrated circuit forms.

[0139] For another example, when the units in the device can be implemented in the form of a processing element scheduler, the processing element may be a general-purpose processor, such as a central processing unit (CPU) or other processors that can call programs. For another example, these units can be integrated together and implemented in the form of a system-on-a-chip (SOC).

[0140] In the present application, various objects such as various messages / information / devices / network elements / systems / devices / actions / operations / processes / concepts, etc. may be named. It can be understood that these specific names do not constitute a limitation on the relevant objects, and the assigned names may change with factors such as the scenario, context, or usage habits. The understanding of the technical meaning of the technical terms in the present application should be mainly determined from the functions and technical effects embodied / executed in the technical solution.

[0141] Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the systems, devices, and units described above can refer to the corresponding processes in the foregoing method embodiments, and will not be described in detail here.

[0142] In several embodiments provided in the present application, it should be understood that the disclosed systems, devices, and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the units is only a logical function division, and there may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed couplings or direct couplings or communication connections between each other can be through some interfaces, and the indirect couplings or communication connections of the devices or units can be in electrical, mechanical, or other forms.

[0143] The unit described as a separation component may or may not be physically separated. The component shown as a unit may or may not be a physical unit, that is, it may be located in one place or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0144] Those of ordinary skill in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of this application.

[0145] It should also be understood that in various embodiments of the present application, the first, second, etc. are only used to indicate that multiple objects are different. For example, the first time window and the second time window are only used to indicate different time windows. And it should not have any impact on the time window itself. The above first, second, etc. should not impose any restrictions on the embodiments of the present application.

[0146] It should also be understood that in various embodiments of the present application, if there is no special description and logical conflict, the terms and / or descriptions between different embodiments are consistent and can be mutually referred to. The technical features in different embodiments can be combined to form new embodiments according to their internal logical relationships.

[0147] If the described function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a computer-readable storage medium and includes several instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of this application. And the aforementioned computer-readable storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical discs that can store program codes.

[0148] This application also provides a customer screening and advertisement recommendation system based on data analysis. The system includes:

[0149] One or more memories for storing instructions; and one or more processors for invoking and running the instructions from the memory to execute the method described above.

[0150] This application also provides a computer program product, which includes instructions that, when executed, cause the terminal device and the network device to perform the operations of the terminal device and the network device corresponding to the above method.

[0151] This application also provides a chip system, which includes a processor for implementing the functions involved above. For example, generating, receiving, sending, or processing the data and / or information involved in the above method.

[0152] The chip system can be composed of chips or can include chips and other discrete devices.

[0153] The processor mentioned anywhere above can be a CPU, a microprocessor, an ASIC, or one or more integrated circuits for controlling the execution of the program of the method for transmitting the above feedback information.

[0154] In a possible design, the chip system further includes a memory for storing necessary program instructions and data. The processor and the memory can be decoupled and separately arranged on different devices and connected by wired or wireless means to support the chip system in implementing various functions in the above embodiments. Alternatively, the processor and the memory can also be coupled on the same device.

[0155] Optionally, the computer instructions are stored in the memory.

[0156] Optionally, the memory is a storage unit within the chip, such as a register, a cache, etc. The memory can also be a storage unit outside the chip within the terminal, such as a ROM or other types of static storage devices that can store static information and instructions, a RAM, etc.

[0157] It can be understood that the memory in this application can be a volatile memory or a non-volatile memory, or can include both volatile and non-volatile memories.

[0158] The non-volatile memory can be a ROM, a programmable ROM (PROM), an erasable PROM (EPROM), an electrically erasable PROM (EEPROM), or a flash memory.

[0159] The volatile memory can be a RAM, which is used as an external cache. There are various different types of RAM, such as static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synch link DRAM (SLDRAM), and direct rambus random access memory.

[0160] The embodiments of this specific implementation manner are all preferred embodiments of this application, and do not limit the protection scope of this application accordingly. Therefore, all equivalent changes made according to the structure, shape, and principle of this application shall be covered within the protection scope of this application.

Claims

1. A customer screening and advertisement recommendation method based on data analysis, characterized in that Including: In response to the acquired real-time image, parsing the real-time image to obtain the person objects included in the real-time image; Grouping the person objects included in the real-time image to obtain person object groups, and the person objects in each person object group have an associated relationship; Obtaining the movement trajectories of the person object groups and generating analysis data according to the movement trajectories, with each person object group corresponding to one analysis data; Using the analysis data to generate an advertising recommendation scheme associated with the person object group; Obtaining the new movement trajectories of the person object groups and implementing the advertising recommendation scheme on the display terminals associated with the new movement trajectories; Grouping the person objects included in the real-time image includes: Using the person objects included in the real-time image to establish an analysis reference network, and each grid in the analysis reference network has the same number of sides; Obtaining the size change degree of each grid in the analysis reference network in the time series; When the size change degree of the grid exceeds the allowable range, performing extinction processing on the grid until the number of grids in the analysis reference network is less than or equal to the set number. At this time, one person in the analysis reference network corresponds to a point, two people correspond to a line segment, and multiple people correspond to a grid; For the case of multiple people, calculating the sparse area of the analysis reference network and pre-partitioning the analysis reference network; Calculating the sparse area of the analysis reference network and pre-partitioning the analysis reference network includes: Calculating the line segment length of each line segment in the analysis reference network, the average line segment length, the grid area of each grid, and the average grid area; Marking the first range in the analysis reference network according to the line segment length and the average line segment length; Marking the second range in the analysis reference network according to the grid area and the average grid area; Determining the overlapping area of the first range and the second range, denoted as the overlapping range; Using the overlapping range, the remaining of the first range, and the remaining of the second range to pre-partition the analysis reference network.

2. The method for customer screening and advertisement recommendation based on data analysis according to claim 1, wherein Obtaining the movement trajectories of the person object groups and generating analysis data according to the movement trajectories includes: Creating movement trajectories based on the movement of the person object groups. The movement trajectories include sequentially connected line segments and points located on the line segments. The line segments have length and direction, and the points have area, and the area of the points is negatively correlated with the movement speed; Determining the content type associated with the points; Generating analysis data according to the content type associated with the points. The analysis data includes content type, content type association time, and content type ratio.

3. The method for customer screening and advertisement recommendation based on data analysis according to claim 1, characterized in that It also includes determining the weight of each person object in the person object group and adjusting the advertising recommendation scheme according to the weight of the person object.

4. The method for customer screening and advertisement recommendation based on data analysis according to claim 3, wherein Determining the weight of each person object in the person object group includes: Calculating the change degree of each grid in the analysis reference network, and the change degree includes size change amount and change speed; Sorting the person objects in the analysis reference network according to the change degree, and the weights of the person objects on the sequential sequence decrease in turn.

5. The method for customer screening and advertisement recommendation based on data analysis according to claim 4, wherein It also includes using the movement trajectories to correct the sorting of the person objects. Using the movement trajectories to correct the sorting of the person objects includes: Obtaining the area and attributes of all points on the movement trajectories; Grouping all points according to the attributes of the points and calculating the total area of the points in each group to obtain grouping values; Use grouping values to modify the order of person objects in a sequential sequence.

6. A customer screening and advertisement recommendation device based on data analysis, characterized in that, include: An image analysis unit, configured to analyze the real-time image in response to the acquired real-time image, and obtain a person object included in the real-time image; A grouping processing unit, used for grouping the human objects included in the real-time image to obtain human object groups, wherein the human objects in each human object group have an associated relationship; A data generating unit, used for acquiring the movement trajectory of the person object group and generating analysis data according to the movement trajectory, wherein each person object group corresponds to one piece of analysis data; A scheme generating unit, configured to generate an advertisement recommendation scheme associated with the character object group using the analysis data; A scheme implementation unit, used for acquiring a newly added movement track of the character object group and implementing the advertisement recommendation scheme on a display terminal associated with the newly added movement track; The grouping process of the human objects included in the real-time image includes: Using the human object included in the real-time image, an analysis reference network is established, wherein each grid in the analysis reference network has the same number of edges; Obtain the size variation of each grid in the analysis reference network in time series; When the size variation of a grid exceeds the allowable range, the grid is eliminated until the number of grids in the analysis reference network is less than or equal to the set number. At this time, one person in the analysis reference network corresponds to a point, two people correspond to a line segment, and multiple people correspond to grids. For the case of multiple persons, the sparse regions of the analysis reference network are calculated and the analysis reference network is pre-partitioned; Calculating the sparse area of the analysis reference network and pre-dividing the analysis reference network includes: Calculate and analyze the segment length of each line segment in the reference network, the average length of the line segment, the grid area of each grid and the average area of the grid; Annotate a first range in the analysis reference network according to the length of the line segment and the mean length of the line segment; Annotate the second range in the analysis reference network based on the grid area and the mean grid area; Determine an overlapping area between the first range and the second range, and record it as an overlapping range; The analysis reference network is pre-partitioned using the coincident range, first range remainder, and second range remainder.

7. A customer screening and advertisement recommendation system based on data analysis, characterized in that, The system comprises: one or more memories for storing instructions; and One or more processors, configured to call and execute the instructions from the memory to perform the method according to any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that, The computer readable storage medium comprises: The program, when the program is executed by a processor, the method according to any one of claims 1 to 5 is executed.

Images