Method for detecting number of queuing people in unattended ATM (Automatic Teller Machine) area and related device

Abstract

The invention discloses a method for detecting the number of queuing people in an unattended ATM area and a related device, and relates to the field of artificial intelligence or finance, and the method comprises the steps: obtaining a first position area of a customer in a video image through a personnel recognition model; and inputting the video image and the first position areas corresponding to the plurality of clients into a queuing curve identification model to obtain a second position area of the queuing team. And acquiring a first image located in a second position area in the video image. Acquiring an actual distance between every two adjacent clients in the first image; and if the actual distance between every two adjacent target customers is large, obtaining a second image between every two adjacent target customers from the first image. And obtaining the article type of the article in the second image. Obtaining a target article type belonging to a preset type, so as to obtain a first number of people between every two target customers; and determining the sum of the first number of people and the second number of people in the first image as the queuing number of people in the ATM area. The purpose of detecting the number of queuing people in the ATM area is achieved.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence technology, and in particular to a method and related device for detecting the number of people queuing in an unmanned ATM area. Background Art

[0002] The area where the ATM (Automated Teller Machine) is located (hereinafter referred to as the ATM area) is usually the area corresponding to the ATM in the 24-hour automatic withdrawal area. The ATM area is where customers deposit and withdraw funds, and its security is directly related to the safety of customer funds. ATMs are usually located in urban areas or busy commercial areas, where there are usually a large number of customers coming and going. The ATM area pays attention to privacy. If there are a large number of people queuing in the ATM area, unsafe incidents may occur.

[0003] Therefore, there is an urgent need for a detection method for the number of people queuing in unattended ATM areas. Summary of the invention

[0004] In view of the above problems, the present application provides a method and related device for detecting the number of people queuing in an unattended ATM area, so as to achieve the purpose of detecting the number of people queuing in an unattended ATM area. The specific scheme is as follows:

[0005] The first aspect of the present application provides a method for detecting the number of people queuing in an unattended ATM area, comprising:

[0006] Obtain real-time video images of the ATM area;

[0007] Inputting the video image into a pre-built personnel recognition model, and obtaining first location areas respectively corresponding to a plurality of customers located in the video image through the personnel recognition model;

[0008] Inputting the video image and the first position areas respectively corresponding to the plurality of customers into a pre-built queue curve recognition model, and obtaining the second position area of ​​the queue line through the queue curve recognition model;

[0009] Acquire a first image located in the second position area in the video image;

[0010] Based on the pixel distance between two adjacent customers in the first image and the shooting parameters of the camera shooting the video image, obtaining the actual distance between two adjacent customers in the first image;

[0011] Determine the corresponding pairwise adjacent target customers whose actual distances are greater than or equal to a first preset multiple*average distance, where the average distance is the average value of the actual distances between the pairwise adjacent customers in the first image, and the first preset multiple is greater than 1;

[0012] Acquire a second image between the two adjacent target customers from the first image;

[0013] Inputting the second image into a pre-built object recognition model, and obtaining the object type of the object in the second image through the object recognition model;

[0014] From the item categories, obtaining a target item category belonging to a preset category;

[0015] Determining a first number of people between each pair of target customers based on the number of items belonging to the target item type;

[0016] The sum of the first number of people and the second number of people in the first image is determined to be the number of people queuing in the ATM area, and the second number of people is the number of the first location area in the first image.

[0017] In a possible implementation, the method further includes:

[0018] Obtain multiple historical time ranges in which the number of queues corresponding to the ATM area is greater than or equal to a preset number of queues;

[0019] Dividing the historical time ranges having intersections into the same time range set;

[0020] For each of the time range sets, if the number of the historical time ranges contained in the time range set is greater than or equal to a first preset number, the union of the historical time ranges in the time range set is determined as the personnel gathering time range.

[0021] In a possible implementation, the method further includes:

[0022] If the current time reaches the personnel gathering time range, obtain a target ATM that is within a preset distance range from the ATM area and is in an idle state;

[0023] Play a prompt message to remind the customer to go to the target ATM to handle business.

[0024] In a possible implementation, each ATM corresponds to one ATM area, and the method for detecting the number of people queuing in the unattended ATM area further includes:

[0025] Obtain the personnel gathering time ranges corresponding to multiple ATMs with the same target function located in the same 24-hour automatic cash withdrawal area;

[0026] If the personnel gathering time ranges corresponding to multiple ATMs with the same target function are the same, an ATM with the target function is added to the 24-hour automatic cash withdrawal area;

[0027] If multiple ATMs with the same target function respectively correspond to different personnel gathering time ranges, the staff is prompted to set function prompt information on the ATMs with the same target function.

[0028] In a possible implementation, each ATM corresponds to one ATM area, and the method for detecting the number of people queuing in the unattended ATM area further includes:

[0029] splicing the video images of the ATM area respectively corresponding to a plurality of ATMs located in the same 24-hour automatic teller withdrawal area to obtain a spliced ​​image;

[0030] Inputting the stitched image into the person recognition model, and obtaining a third person in the stitched image through the person recognition model;

[0031] If the third number of persons is greater than or equal to the second preset number, a warning operation is performed.

[0032] The second aspect of the present application provides a device for detecting the number of people in a queue in an unattended ATM area, comprising:

[0033] The first acquisition module is used to acquire video images collected in real time for the ATM area;

[0034] A second acquisition module is used to input the video image into a pre-built personnel recognition model, and obtain first location areas corresponding to a plurality of customers located in the video image respectively through the personnel recognition model;

[0035] A third acquisition module is used to input the video image and the first position areas respectively corresponding to the plurality of customers into a pre-built queue curve recognition model, and acquire the second position area of ​​the queue line through the queue curve recognition model;

[0036] A fourth acquisition module, used for acquiring a first image located in the second position area in the video image;

[0037] a fifth acquisition module, configured to acquire the actual distance between two adjacent customers in the first image based on the pixel distance between two adjacent customers in the first image and the shooting parameters of the camera that shoots the video image;

[0038] A first determination module is used to determine the corresponding two adjacent target customers whose actual distance is greater than or equal to a first preset multiple * average distance, where the average distance is the average value of the actual distances between two adjacent customers in the first image, and the first preset multiple is greater than 1;

[0039] A sixth acquisition module, configured to acquire a second image between the two adjacent target customers from the first image;

[0040] a seventh acquisition module, configured to input the second image into a pre-built object recognition model, and obtain the object type of the object in the second image through the object recognition model;

[0041] An eighth acquisition module, used to acquire a target item category belonging to a preset category from the item categories;

[0042] A second determination module, configured to determine a first number of persons between the two target customers based on the number of items belonging to the target item type;

[0043] The third determining module is used to determine that the sum of the first number of people and the second number of people in the first image is the number of people queuing in the ATM area, and the second number of people is the number of the first position area in the first image.

[0044] In a possible implementation, the method further includes:

[0045] A ninth acquisition module, used to acquire multiple historical time ranges in which the number of queues corresponding to the ATM area is greater than or equal to a preset number of queues;

[0046] A division module, used for dividing the historical time ranges with intersections into the same time range set;

[0047] The fourth determination module is used to determine, for each of the time range sets, if the number of the historical time ranges contained in the time range set is greater than or equal to a first preset number, that the union of the historical time ranges in the time range set is a personnel gathering time range.

[0048] The third aspect of the present application provides a computer program product, including computer-readable instructions. When the computer-readable instructions are executed on an edge computing box, the edge computing box implements the method for detecting the number of people queuing in an unattended ATM area according to the first aspect or any implementation of the first aspect.

[0049] A fourth aspect of the present application provides an edge computing box, comprising at least one processor and a memory connected to the processor, wherein:

[0050] The memory is used to store computer programs;

[0051] The processor is used to execute the computer program so that the edge computing box can implement the method for detecting the number of people queuing in an unattended ATM area according to the first aspect or any implementation method of the first aspect.

[0052] A fifth aspect of the present application provides a computer storage medium, which carries one or more computer programs. When the one or more computer programs are executed by an edge computing box, the edge computing box can implement the method for detecting the number of people queuing in an unmanned ATM area according to the first aspect or any implementation of the first aspect.

[0053] By means of the above technical solution, the present application provides a method for detecting the number of people in a queue in an unattended ATM area, obtaining a video image collected in real time for the ATM area; inputting the video image into a pre-built personnel recognition model, and obtaining the first position areas corresponding to the multiple customers in the video image through the personnel recognition model. It can be understood that the first position area of ​​the customer in the video image can be obtained through the personnel recognition model, thereby obtaining the number of customers, that is, the number of the first position areas. However, some customers in the ATM area do not queue, that is, the customers are not in the queue. If these customers who are not in the queue are also counted as the number of people in the queue, the number of people in the queue is inaccurate. Based on this, the video image and the first position areas corresponding to the multiple customers are input into the pre-built queue curve recognition model, and the second position area of ​​the queue is obtained through the queue curve recognition model. A first image located in the second position area in the video image is obtained. The first image does not include customers who are not in the queue. Based on the pixel distance between two adjacent customers in the first image and the shooting parameters of the camera that shoots the video image, the actual distance between two adjacent customers in the first image is obtained; if the actual distance between two adjacent target customers among the two adjacent customers is greater than or equal to the first preset multiple * average distance, it means that there may be an object that is not recognized by the personnel recognition model between the two adjacent target customers, and it is necessary to further determine whether the object corresponds to a customer. A second image between two adjacent target customers is obtained from the first image. The second image is input into a pre-built item recognition model, and the item type of the item in the second image is obtained through the item recognition model. From the item types, the target item type belonging to the preset type is obtained. The items belonging to the preset type correspond to customers. Based on the number of items belonging to the target item type, the first number of people between two target customers is determined. The sum of the first number of people and the second number of people in the first image is determined to be the number of people queuing in the ATM area. Thereby, the purpose of detecting the number of people queuing in the ATM area is achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] The above and other features, advantages and aspects of the embodiments of the present disclosure will become more apparent with reference to the following detailed description in conjunction with the accompanying drawings. Throughout the accompanying drawings, the same or similar reference numerals represent the same or similar elements. It should be understood that the drawings are schematic and the originals and elements are not necessarily drawn to scale.

[0055] Figure 1A schematic diagram of a system architecture provided for this application;

[0056] Figure 2 A schematic diagram of the structure of a server 200 provided in this application;

[0057] Figure 3 A flow chart of a method for detecting the number of people in a queue in an unattended ATM area provided in an embodiment of the present application;

[0058] Figure 4 A schematic diagram of a queue provided in an embodiment of the present application;

[0059] Figure 5 A schematic diagram of the structure of a device for detecting the number of people in a queue in an unattended ATM area provided in an embodiment of the present application;

[0060] Figure 6 A schematic diagram of the structure of an edge computing box provided in an embodiment of the present application. DETAILED DESCRIPTION

[0061] The following describes the embodiments of the present application in conjunction with the drawings in the embodiments of the present application. The terms used in the implementation method section of the present application are only used to explain the specific embodiments of the present application, and are not intended to limit the present application.

[0062] The embodiments of the present application are described below in conjunction with the accompanying drawings. Those skilled in the art will appreciate that, with the development of technology and the emergence of new scenarios, the technical solutions provided in the embodiments of the present application are also applicable to similar technical problems.

[0063] The terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and need not be used to describe a specific order or sequential order. It should be understood that the terms used in this way can be interchangeable under appropriate circumstances, which is only to describe the distinction mode adopted by the objects of the same attributes when describing in the embodiments of the present application. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, so that the process, method, system, product or equipment comprising a series of units need not be limited to those units, but may include other units that are not clearly listed or inherent to these processes, methods, products or equipment.

[0064] See also Figure 1 , Figure 1 A schematic diagram of a system architecture is shown. The system may include an edge computing box 100, a server 200, and multiple cameras 300.

[0065] The server 200 may include one or more servers ( Figure 1A server is included as an example for explanation).

[0066] The camera 300 is used to capture still images or videos. The subject (object or organism) generates an optical image through the lens and projects it onto the photosensitive element. The photosensitive element can be a charge coupled device (CCD) or a complementary metal-oxide-semiconductor (CMOS) phototransistor. The photosensitive element converts the optical signal into an electrical signal, which is then transmitted to the ISP for conversion into a digital image signal.

[0067] Exemplary, multiple cameras 300 are deployed in a 24-hour automatic cash withdrawal area. It is understandable that the number of ATMs located in the 24-hour automatic cash withdrawal area may be one or more. If there are multiple ATMs, different cameras 300 collect video images in real time for different ATM areas located in the 24-hour automatic cash withdrawal area.

[0068] Exemplarily, the camera 300 may be connected to an NVR (Network Video Recorder), which receives, stores and manages the video images captured by the camera.

[0069] For example, the edge computing box 100 can obtain corresponding real-time video images from the video streams transmitted to the NVR by multiple cameras 300. Thus, the method for detecting the number of people in the queue of the unattended ATM area provided in the embodiment of the present application can be executed to determine the number of people in the queue of the ATM area, and if an early warning is required, the early warning information can be sent to the server 200.

[0070] Exemplarily, the edge computing box 100 can be deployed in a 24-hour ATM area.

[0071] Exemplarily, the edge computing box 100 may include a storage unit, a power distribution unit, a computing unit, and the like.

[0072] Exemplarily, the edge computing box 100 can send the warning results and the video images containing the abandoned items to the server 200.

[0073] Next describe Figure 1 The product form of the server 200;

[0074] Figure 2 A structural diagram of a server 200 is provided, such as Figure 2As shown, the server 200 includes a bus 201, a processor 202, a communication interface 203, and a memory 204. The processor 202, the memory 204, and the communication interface 203 communicate with each other via the bus 201.

[0075] The bus 201 may be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus. The bus may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 2 Only one thick line is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0076] The processor 202 may be any one or more of a central processing unit (CPU), a graphics processing unit (GPU), a microprocessor (MP), or a digital signal processor (DSP).

[0077] The memory 204 may include a volatile memory, such as a random access memory (RAM). The memory 204 may also include a non-volatile memory, such as a read-only memory (ROM), a flash memory, a hard disk drive (HDD), or a solid state drive (SSD).

[0078] It should be understood that the above-mentioned edge computing box 100 and server 200 can be centralized or distributed devices, and the processors in the above-mentioned edge computing box 100 and server 200 (such as processor 170 and processor 202) can be hardware circuits (such as application specific integrated circuit (ASIC), field-programmable gate array (FPGA), general-purpose processor, digital signal processor (DSP), microprocessor or microcontroller, etc.), or a combination of these hardware circuits. For example, the processor can be a hardware system with the function of executing instructions, such as CPU, DSP, etc., or a hardware system without the function of executing instructions, such as ASIC, FPGA, etc., or a combination of the above-mentioned hardware systems without the function of executing instructions and hardware systems with the function of executing instructions.

[0079] Reference Figure 3 , Figure 3 A schematic diagram of a process for detecting the number of people in a queue in an unattended ATM area provided in an embodiment of the present application is shown in FIG. Figure 3 As shown, a method for detecting the number of people queuing in an unattended ATM area provided in an embodiment of the present application may include steps S301 to S311, and these steps are described in detail below.

[0080] Step S301: Acquire video images collected in real time for the ATM area.

[0081] Each ATM corresponds to an ATM area. The ATM area corresponding to each ATM can be determined by the staff based on actual conditions.

[0082] Exemplarily, the ATM area may be a regular polygon, an irregular polygon, or a figure composed of curves.

[0083] Illustratively, the ATM area is virtual and invisible to the customer.

[0084] Exemplary, the ATM area actually exists and can be seen by customers. For example, the queuing area marked with colors on the ground is the ATM area.

[0085] Step S302: Input the video image into a pre-built personnel recognition model, and obtain The first location areas respectively correspond to a plurality of customers located in the video image.

[0086] It is understood that the person located in the ATM area is the customer.

[0087] Exemplarily, the person recognition model is trained by taking a sample video image as input and taking first position areas respectively corresponding to a plurality of customers marked in the sample video image as training targets.

[0088] Exemplarily, the first location area of ​​each customer is the area where the customer's full body outline is located, that is, the person recognition model can recognize the customer's head, shoulders, torso and limbs, and can determine which head, shoulders, torso and limbs belong to the same customer.

[0089] Exemplarily, the first location area of ​​each customer is the area where the customer's head is located, that is, the person recognition model can recognize the customer's head.

[0090] Step S303: input the video image and the first position areas respectively corresponding to the multiple customers into a pre-built queuing curve recognition model, and obtain the second position area of ​​the queuing team through the queuing curve recognition model.

[0091] It is understandable that the queue formed by multiple customers may be a straight line or a curve. The line of the queue formed by the customers can be determined by the queue curve recognition model, and the area where the line is located is the second position area of ​​the queue.

[0092] In order for those skilled in the art to better understand the second position area of ​​the queue provided in the embodiment of the present application, the following example is given for illustration. Figure 4 As shown, it is a schematic diagram of a queue provided in an embodiment of the present application.

[0093] Figure 4 The line formed by the queue in the middle is a curve. Figure 4 After the video image on the left side shown in FIG. is input into the queue curve recognition model, the following equation can be obtained: Figure 4 The image shown on the right. Figure 4 As shown, the second position area is the second position area 41 .

[0094] from Figure 4 It can be seen that there are also some customers outside the second location area 41, but these customers are not in the queue. In the related art, if all customers in the ATM area are regarded as queuing customers, the number of queuing people obtained will be inaccurate.

[0095] Exemplarily, the queuing curve recognition model is obtained by training by taking a sample video image as input and taking the annotated second position area of ​​the sample video image as a training target.

[0096] Step S304: Acquire a first image located in the second position area in the video image.

[0097] It can be understood that if the second position area can be a regular polygonal area, it can also be a curved area, such as Figure 4 As shown, the area formed by two thick solid lines is the second position area.

[0098] Step S305: based on the pixel distance between two adjacent customers in the first image and the shooting parameters of the camera that shoots the video image, the actual distance between two adjacent customers in the first image is obtained.

[0099] Exemplarily, the shooting parameters of the camera for shooting the video image include, but are not limited to: the focal length of the camera, the shooting angle of the camera, the actual distance between the camera and the set position in the video image, and the pixel position of the set position in the video image.

[0100] Step S306: determining the pairwise adjacent target customers whose corresponding actual distances are greater than or equal to the first preset multiple*average distance, where the average distance is the average value of the actual distances between the pairwise adjacent customers in the first image.

[0101] Exemplarily, the first preset multiple is greater than 1.

[0102] In the embodiment of the present application, “two adjacent customers whose actual distance is greater than or equal to the first preset multiple*average distance” are referred to as two adjacent target customers.

[0103] Exemplarily, the first preset multiple is determined based on actual conditions and is not limited here. For example, the first preset multiple may be 2.

[0104] For example, there are no heads, shoulders, torsos and limbs that can be identified by any person recognition model between any two adjacent target customers. This is because there may be objects such as umbrellas and sun hats between any two adjacent target customers that cover the heads, shoulders, torsos and limbs of the customers between the two target customers.

[0105] It is understandable that if the actual distance between two adjacent target customers is greater than the first preset multiple * average distance, and there is no head, shoulder, torso and limbs that can be identified by any person recognition model, then the two adjacent target customers may be customers holding umbrellas or customers wearing large sun hats. If only the person recognition model is used to identify the number of people queuing in the ATM area, customers holding umbrellas or customers wearing large sun hats may be missed, resulting in inaccurate number of people queuing.

[0106] Step S307: acquiring a second image between the two adjacent target customers from the first image.

[0107] Exemplarily, the number of “two-by-two adjacent target customers” may be one or more, and each “two-by-two adjacent target customer” corresponds to a second image.

[0108] Step S308: input the second image into a pre-built object recognition model, and obtain the object type of the object in the second image through the object recognition model.

[0109] Exemplarily, the object recognition model is obtained through training by taking sample images as input and taking the object types marked in the sample images as training results.

[0110] Step S309: acquiring target item categories belonging to preset categories from the item categories.

[0111] Exemplarily, the preset types may be items such as umbrellas and sun hats that customers may carry on their heads when conducting business.

[0112] For example, if the item type does not belong to the preset type, the item that does not belong to the preset type does not correspond to a customer. For example, after shopping in a supermarket, a customer puts multiple shopping bags in the queue. Then the actual distance between the two target customers in front and behind the shopping bags may be greater than or equal to the first preset multiple * average distance. However, the shopping bags do not correspond to any customer.

[0113] Step S310: determining a first number of people between each pair of target customers based on the number of items belonging to the target item type.

[0114] Exemplarily, if the number of customers with items belonging to the target item type is inserted between each pair of target customers, so that the actual distance between each pair of adjacent customers in the second image is smaller than the above average distance, it is determined that the first number of customers between each pair of target customers = the number of items belonging to the target item type.

[0115] Exemplarily, if a number of customers with items belonging to the target item type are inserted between each pair of target customers, resulting in a large difference between the actual distance and the average distance between each pair of adjacent customers in the second image, the first number of people can be determined based on the number of items belonging to the target item type and the actual distance between each pair of target customers.

[0116] Step S311: Determine that the sum of the first number of people and the second number of people in the first image is the number of people queuing in the ATM area, and the second number of people is the number of the first position area in the first image.

[0117] It can be understood that, since the first position areas respectively corresponding to the multiple customers in the video image have been obtained in step S302, the number of the first position areas in the first image can be determined.

[0118] The embodiment of the present application provides a method for detecting the number of people in a queue in an unattended ATM area, obtaining a video image collected in real time for the ATM area; inputting the video image into a pre-built personnel recognition model, and obtaining the first position areas corresponding to multiple customers in the video image through the personnel recognition model. It can be understood that the first position area of ​​the customer in the video image can be obtained through the personnel recognition model, thereby obtaining the number of customers, that is, the number of the first position areas. However, some customers in the ATM area do not queue, that is, the customers are not in the queue. If these customers who are not in the queue are also counted as the number of people in the queue, the number of people in the queue is inaccurate. Based on this, the video image and the first position areas corresponding to the multiple customers are input into a pre-built queue curve recognition model, and the second position area of ​​the queue is obtained through the queue curve recognition model. A first image located in the second position area in the video image is obtained. The first image does not include customers who are not in the queue. Based on the pixel distance between two adjacent customers in the first image and the shooting parameters of the camera that shoots the video image, the actual distance between two adjacent customers in the first image is obtained; if the actual distance between two adjacent target customers among the two adjacent customers is greater than or equal to the first preset multiple * average distance, it means that there may be an object that is not recognized by the personnel recognition model between the two adjacent target customers, and it is necessary to further determine whether the object corresponds to a customer. A second image between two adjacent target customers is obtained from the first image. The second image is input into a pre-built item recognition model, and the item type of the item in the second image is obtained through the item recognition model. From the item types, the target item type belonging to the preset type is obtained. The items belonging to the preset type correspond to customers. Based on the number of items belonging to the target item type, the first number of people between two target customers is determined. The sum of the first number of people and the second number of people in the first image is determined to be the number of people queuing in the ATM area. Thereby, the purpose of detecting the number of people queuing in the ATM area is achieved.

[0119] It is understandable that each ATM corresponds to an ATM area. The following is an explanation of obtaining the time range of personnel gathering at the ATM. The method includes but is not limited to the following steps A1 to A3.

[0120] Step A1: Acquire multiple historical time ranges in which the number of queues corresponding to the ATM area is greater than or equal to a preset number of queues.

[0121] For example, the preset number of people can be determined based on actual conditions and is not limited here.

[0122] Step A2: Divide the historical time ranges with intersections into the same time range set.

[0123] The following example illustrates this. Assume that the current time is November 4, 2024, and that the multiple historical time ranges of the ATM area of ​​the same ATM are: [10:00,11:00], [12:00,12:30] corresponding to October 30, 2024, [10:30,11:30] corresponding to October 31, 2024, [14:00,15:00] corresponding to November 2, 2024, and [12:00,12:30] corresponding to October 31, 2024. The time range set {[10:00,11:00], [10:30,11:30]} is obtained because the time ranges [10:00,11:00], [10:30,11:30] corresponding to October 30, 2024 and [10:30,11:30] corresponding to October 31, 2024 have the intersection [10:30,11:00]. The time range set {[10:00,11:00], [10:30,11:30]} is obtained because the time ranges [14:00,15:00] corresponding to November 2, 2024 and [14:00,16:00] corresponding to November 3, 2024 have the intersection [14:00,15:00].

[0124] Step A3: For each of the time range sets, if the number of the historical time ranges contained in the time range set is greater than or equal to a first preset number, determine the union of the historical time ranges in the time range set as the personnel gathering time range.

[0125] Exemplarily, the first preset number may be determined based on actual conditions and is not limited here.

[0126] Still taking the above example, assuming that the first preset number is 2, the personnel gathering time range corresponding to the time range set {[10:00, 11:00], [10:30, 11:30]} is [10:00, 11:30].

[0127] It is understandable that the personnel gathering time range of the ATM area corresponding to each ATM can be determined through steps A1 to A3. If the current time reaches the personnel gathering time range, then generally there will be a situation where people gather. Based on this, the application also provides the following method, which includes the following steps B1 to B2.

[0128] Step B1: If the current time reaches the personnel gathering time range, obtain a target ATM that is within a preset distance range from the ATM area and is in an idle state.

[0129] Step B2: Play a prompt message to remind the customer to go to the target ATM to handle business.

[0130] Through step B1 and step B2, customers queuing in the ATM area can be evacuated.

[0131] It is understandable that through steps A1 to A3, the time range of personnel gathering in the ATM area corresponding to each ATM can be determined. However, there may be the following two reasons for the gathering of personnel, and the operation methods for different reasons for the gathering of personnel are different. The following is a detailed description, and the process includes the following steps C1 to C3.

[0132] Step C1: Obtain the personnel gathering time ranges corresponding to multiple ATMs with the same target function located in the same 24-hour automatic teller withdrawal area.

[0133] Exemplarily, the target function may be one or more of withdrawing money, depositing money, and changing a mobile phone number.

[0134] Step C2: If a plurality of ATMs having the same target function respectively correspond to the same personnel gathering time range, an ATM having the target function is added to the 24-hour automatic cash withdrawal area.

[0135] Step C3: If the personnel gathering time ranges corresponding to multiple ATMs with the same target function are different, the staff is prompted to set function prompt information on the ATMs with the same target function.

[0136] Exemplarily, the function prompt information includes a description of the target function of the ATM.

[0137] It is understandable that the appearances of multiple ATMs with the same target function may be different. This may cause customers to misunderstand that the ATM has a target function. The following example is used to illustrate. Assuming that the target functions are withdrawal and deposit, the number of multiple ATMs with the same target function located in the same 24-hour automatic withdrawal area is 2, and they are the first ATM and the second ATM. Due to the different appearances of the first ATM and the second ATM, customers believe that the first ATM can only withdraw money but not deposit money, and the second ATM can deposit money and withdraw money. If there are many customers who handle deposit business, customers will gather in the ATM area corresponding to the second ATM, but not in the ATM area corresponding to the first ATM, resulting in different time ranges for gathering people corresponding to the first ATM and the second ATM. In summary, the reason why the time ranges for gathering people corresponding to the ATM areas of multiple ATMs with the same target function are different is that customers are not clear about the functions of each ATM, so step C3 needs to be executed. If the time ranges for gathering people corresponding to the ATM areas of multiple ATMs with the same target function are the same, there is no problem that customers are not clear about the functions of each ATM, so step C2 is executed.

[0138] It is understandable that if there are many people in the 24-hour automatic withdrawal area, even if there are few people in the ATM area, an early warning is required. Based on this, the embodiment of the present application provides the following method, which includes the following steps D1 to D3 when implemented.

[0139] Step D1: stitching the video images of the ATM area respectively corresponding to a plurality of ATMs located in the same 24-hour automatic teller withdrawal area to obtain a stitched image.

[0140] Step D2: inputting the stitched image into the person recognition model, and obtaining a third person in the stitched image through the person recognition model.

[0141] Step D3: If the third number of persons is greater than or equal to the second preset number, a warning operation is performed.

[0142] Exemplarily, the second preset number may be determined based on actual conditions and is not limited here.

[0143] Exemplarily, the warning operation includes, but is not limited to: sending warning information to a terminal held by a security guard, and sending warning information to a terminal held by a bank staff.

[0144] The above describes a method for detecting the number of people queuing in an unattended ATM area provided by an embodiment of the present application. The following describes a device for executing the above method for detecting the number of people queuing in an unattended ATM area.

[0145] See also Figure 5 , Figure 5 A schematic diagram of the structure of a device for detecting the number of people in a queue in an unattended ATM area provided in an embodiment of the present application. Figure 5 As shown, the queuing number detection device in the unattended ATM area includes:

[0146] The first acquisition module 501 is used to acquire video images collected in real time from the ATM area;

[0147] A second acquisition module 502 is used to input the video image into a pre-built personnel recognition model, and obtain first location areas corresponding to a plurality of customers located in the video image respectively through the personnel recognition model;

[0148] A third acquisition module 503 is used to input the video image and the first position areas respectively corresponding to the plurality of customers into a pre-built queue curve recognition model, and acquire the second position area of ​​the queue line through the queue curve recognition model;

[0149] A fourth acquisition module 504 is used to acquire a first image located in the second position area in the video image;

[0150] A fifth acquisition module 505, configured to acquire the actual distance between two adjacent customers in the first image based on the pixel distance between two adjacent customers in the first image and the shooting parameters of the camera shooting the video image;

[0151] A first determination module 506 is used to determine the corresponding two adjacent target customers whose actual distance is greater than or equal to a first preset multiple * average distance, where the average distance is the average value of the actual distances between two adjacent customers in the first image, and the first preset multiple is greater than 1;

[0152] A sixth acquisition module 507, configured to acquire a second image between the two adjacent target customers from the first image;

[0153] A seventh acquisition module 508, configured to input the second image into a pre-built object recognition model, and obtain the object type of the object in the second image through the object recognition model;

[0154] An eighth acquisition module 509, configured to acquire a target item category belonging to a preset category from the item categories;

[0155] A second determining module 510 is used to determine a first number of people between the two target customers based on the number of items belonging to the target item type;

[0156] The third determining module 511 is used to determine that the sum of the first number of people and the second number of people in the first image is the number of people queuing in the ATM area, and the second number of people is the number of the first position area in the first image.

[0157] In an optional implementation, the method further includes:

[0158] A ninth acquisition module, used to acquire multiple historical time ranges in which the number of queues corresponding to the ATM area is greater than or equal to a preset number of queues;

[0159] A division module, used for dividing the historical time ranges with intersections into the same time range set;

[0160] The fourth determination module is used to determine, for each of the time range sets, if the number of the historical time ranges contained in the time range set is greater than or equal to a first preset number, that the union of the historical time ranges in the time range set is a personnel gathering time range.

[0161] In an optional implementation, the method further includes:

[0162] A tenth acquisition module, configured to acquire a target ATM that is within a preset distance range from the ATM area and is in an idle state if the current time reaches the personnel gathering time range;

[0163] The playing module is used to play the prompt information that prompts the customer to go to the target ATM to handle the business.

[0164] In an optional implementation, each ATM corresponds to one of the ATM areas, further comprising:

[0165] An eleventh acquisition module is used to acquire the personnel gathering time ranges corresponding to multiple ATMs with the same target function located in the same 24-hour automatic cash withdrawal area;

[0166] A first processing module is used to add an ATM with the target function in the 24-hour automatic cash withdrawal area if the personnel gathering time ranges corresponding to multiple ATMs with the same target function are the same;

[0167] The second processing module is used to prompt the staff to set function prompt information on the ATMs with the same target function if the personnel gathering time ranges corresponding to multiple ATMs with the same target function are different.

[0168] In an optional implementation, each ATM corresponds to one of the ATM areas, further comprising:

[0169] A twelfth acquisition module, used for splicing the video images of the ATM area respectively corresponding to multiple ATMs located in the same 24-hour automatic teller withdrawal area to obtain a spliced ​​image;

[0170] A thirteenth acquisition module, used for inputting the stitched image into the personnel recognition model, and acquiring a third person in the stitched image through the personnel recognition model;

[0171] The execution module is used to execute an early warning operation if the third number of people is greater than or equal to a second preset number.

[0172] The present application also provides an edge computing box. Figure 6 As shown, it shows a schematic diagram of the structure of an edge computing box suitable for implementing the embodiment of the present application. The edge computing box in the embodiment of the present application may include but is not limited to fixed terminals such as mobile phones, laptops, PDAs (personal digital assistants), PADs (tablet computers), desktop computers, etc. Figure 6 The edge computing box shown is merely an example and should not bring any limitations to the functionality and scope of use of the embodiments of the present application.

[0173] like Figure 6As shown, the edge computing box may include a processing device (e.g., a central processing unit, a graphics processing unit, etc.) 601, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 602 or a program loaded from a storage device 608 to a random access memory (RAM) 603. When the edge computing box is powered on, various programs and data required for the operation of the edge computing box are also stored in the RAM 603. The processing device 601, the ROM 602, and the RAM 603 are connected to each other via a bus 604. An input / output (I / O) interface 605 is also connected to the bus 604.

[0174] Typically, the following devices may be connected to the I / O interface 605: input devices 606 including, for example, a touch screen, a touchpad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; output devices 607 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; storage devices 608 including, for example, a memory card, a hard disk, etc.; and communication devices 609. The communication device 609 may allow the edge computing box to communicate with other devices wirelessly or by wire to exchange data. Although Figure 6 An edge computing box with various devices is shown, but it should be understood that it is not required to implement or have all the devices shown. More or fewer devices may be implemented or have instead.

[0175] Also provided in an embodiment of the present application is a computer program product including computer-readable instructions. When the computer-readable instructions are executed on an edge computing box, the edge computing box implements any method for detecting the number of people in queue in an unattended ATM area provided in an embodiment of the present application.

[0176] A computer-readable storage medium is also provided in an embodiment of the present application. The storage medium carries one or more computer programs. When one or more computer programs are executed by an edge computing box, the edge computing box can implement any of the queue number detection methods for unmanned ATM areas provided in the embodiments of the present application.

[0177] It should be noted that the method and related devices for detecting the number of people in a queue in an unattended ATM area provided by the present invention can be used in the field of artificial intelligence or finance. The above is only an example and does not limit the application field of the method and related devices for detecting the number of people in a queue in an unattended ATM area provided by the present invention.

[0178] It should also be noted that the device embodiments described above are merely schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed over multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. In addition, in the drawings of the device embodiments provided by the present application, the connection relationship between the modules indicates that there is a communication connection between them, which may be specifically implemented as one or more communication buses or signal lines.

[0179] Through the description of the above implementation mode, the technicians in the field can clearly understand that the present application can be implemented by means of software plus necessary general hardware, and of course, it can also be implemented by special hardware including special integrated circuits, special CPUs, special memories, special components, etc. In general, all functions completed by computer programs can be easily implemented by corresponding hardware, and the specific hardware structure used to implement the same function can also be various, such as analog circuits, digital circuits or special circuits. However, for the present application, software program implementation is a better implementation mode in more cases. Based on such an understanding, the technical solution of the present application is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which is stored in a readable storage medium, such as a computer floppy disk, a U disk, a mobile hard disk, a ROM, a RAM, a disk or an optical disk, etc., including a number of instructions to enable a computer device (which can be a personal computer, a training device, or a network device, etc.) to execute the methods described in each embodiment of the present application.

[0180] In the above embodiments, all or part of the embodiments may be implemented by software, hardware, firmware or any combination thereof. When implemented by software, all or part of the embodiments may be implemented in the form of a computer program product.

[0181] The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions may be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions may be transmitted from a website site, a computer, a training device, or a data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode to another website site, computer, training device, or data center. The computer-readable storage medium may be any available medium that a computer can store or a data storage device such as a training device, a data center, etc. that includes one or more available media integrations. The available medium may be a magnetic medium, (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid-state drive (SSD)), etc.

Claims

1. A method for detecting the number of people queuing in an unattended ATM area, characterized in that: include: Obtain real-time video images of the ATM area; Inputting the video image into a pre-built personnel recognition model, and obtaining first location areas respectively corresponding to a plurality of customers located in the video image through the personnel recognition model; Inputting the video image and the first position areas respectively corresponding to the plurality of customers into a pre-built queue curve recognition model, and obtaining the second position area of ​​the queue line through the queue curve recognition model; Acquire a first image located in the second position area in the video image; Based on the pixel distance between two adjacent customers in the first image and the shooting parameters of the camera shooting the video image, obtaining the actual distance between two adjacent customers in the first image; Determine the corresponding pairwise adjacent target customers whose actual distances are greater than or equal to a first preset multiple*average distance, where the average distance is the average value of the actual distances between the pairwise adjacent customers in the first image, and the first preset multiple is greater than 1; Acquire a second image between the two adjacent target customers from the first image; Inputting the second image into a pre-built object recognition model, and obtaining the object type of the object in the second image through the object recognition model; From the item categories, obtaining a target item category belonging to a preset category; Determining a first number of people between each pair of target customers based on the number of items belonging to the target item type; The sum of the first number of people and the second number of people in the first image is determined to be the number of people queuing in the ATM area, and the second number of people is the number of the first location area in the first image.

2. The method for detecting the number of people queuing in an unattended ATM area according to claim 1, characterized in that: Also includes: Obtain multiple historical time ranges in which the number of queues corresponding to the ATM area is greater than or equal to a preset number of queues; Dividing the historical time ranges having intersections into the same time range set; For each of the time range sets, if the number of the historical time ranges contained in the time range set is greater than or equal to a first preset number, the union of the historical time ranges in the time range set is determined as the personnel gathering time range.

3. The method for detecting the number of people queuing in an unattended ATM area according to claim 2, characterized in that: Also includes: If the current time reaches the personnel gathering time range, obtain a target ATM that is within a preset distance range from the ATM area and is in an idle state; Play a prompt message to remind the customer to go to the target ATM to handle business.

4. The method for detecting the number of people queuing in an unattended ATM area according to claim 2, characterized in that: Each ATM corresponds to one of the ATM areas, and the method for detecting the number of people queuing in the unattended ATM area further includes: Obtain the personnel gathering time ranges corresponding to multiple ATMs with the same target function located in the same 24-hour automatic cash withdrawal area; If the personnel gathering time ranges corresponding to multiple ATMs with the same target function are the same, an ATM with the target function is added to the 24-hour automatic cash withdrawal area; If multiple ATMs with the same target function respectively correspond to different personnel gathering time ranges, the staff is prompted to set function prompt information on the ATMs with the same target function.

5. The method for detecting the number of people queuing in an unattended ATM area according to claim 1, characterized in that: Each ATM corresponds to one of the ATM areas, and the method for detecting the number of people queuing in the unattended ATM area further includes: splicing the video images of the ATM area respectively corresponding to a plurality of ATMs located in the same 24-hour automatic teller withdrawal area to obtain a spliced ​​image; Inputting the stitched image into the person recognition model, and obtaining a third person in the stitched image through the person recognition model; If the third number of persons is greater than or equal to the second preset number, a warning operation is performed.

6. A device for detecting the number of people in a queue in an unattended ATM area, characterized in that: include: The first acquisition module is used to acquire video images collected in real time for the ATM area; A second acquisition module is used to input the video image into a pre-built personnel recognition model, and obtain first location areas corresponding to a plurality of customers located in the video image respectively through the personnel recognition model; A third acquisition module is used to input the video image and the first position areas respectively corresponding to the plurality of customers into a pre-built queue curve recognition model, and acquire the second position area of ​​the queue line through the queue curve recognition model; A fourth acquisition module, used for acquiring a first image located in the second position area in the video image; a fifth acquisition module, configured to acquire the actual distance between two adjacent customers in the first image based on the pixel distance between two adjacent customers in the first image and the shooting parameters of the camera that shoots the video image; A first determination module is used to determine the corresponding two adjacent target customers whose actual distance is greater than or equal to a first preset multiple * average distance, where the average distance is the average value of the actual distances between two adjacent customers in the first image, and the first preset multiple is greater than 1; A sixth acquisition module, configured to acquire a second image between the two adjacent target customers from the first image; a seventh acquisition module, configured to input the second image into a pre-built object recognition model, and obtain the object type of the object in the second image through the object recognition model; An eighth acquisition module, used to acquire a target item category belonging to a preset category from the item categories; A second determination module, configured to determine a first number of persons between the two target customers based on the number of items belonging to the target item type; The third determining module is used to determine that the sum of the first number of people and the second number of people in the first image is the number of people queuing in the ATM area, and the second number of people is the number of the first position area in the first image.

7. The queuing number detection device in the unattended ATM area according to claim 6, characterized in that: Also includes: A ninth acquisition module, used to acquire multiple historical time ranges in which the number of queues corresponding to the ATM area is greater than or equal to a preset number of queues; A division module, used for dividing the historical time ranges with intersections into the same time range set; The fourth determination module is used to determine, for each of the time range sets, if the number of the historical time ranges contained in the time range set is greater than or equal to a first preset number, that the union of the historical time ranges in the time range set is a personnel gathering time range.

8. A computer program product, characterized in that It includes computer-readable instructions. When the computer-readable instructions are executed on the edge computing box, the edge computing box implements the method for detecting the number of people in queue in an unattended ATM area as described in any one of claims 1 to 5.

9. An edge computing box, characterized in that: The method comprises at least one processor and a memory connected to the processor, wherein: The memory is used to store computer programs; The processor is used to execute the computer program so that the edge computing box can implement the method for detecting the number of people in queue in an unattended ATM area as described in any one of claims 1 to 5.

10. A computer storage medium, characterized in that: The storage medium carries one or more computer programs. When the one or more computer programs are executed by the edge computing box, the edge computing box can implement the method for detecting the number of people in queue in an unmanned ATM area as described in any one of claims 1 to 5.

Images