Method, system, computer device and storage medium for identifying private cashier

Abstract

The present application relates to a method for identifying private cashiering, the method comprising: acquiring a first image of a first target environment area; wherein the first image includes at least a first detection area, the first detection area including a cash register; identifying cash targets and cash box targets in the first detection area, determining whether the position of the cash target is located within the cash box target based on the position information of the cash target and the cash box target, and if the position of the cash target is not located within the cash box target, determining that a first transaction event has occurred; determining whether a receipt appears in a time period associated with the first transaction event, and if not, outputting an alarm instruction, wherein the associated time period is a first preset time period before the first transaction event occurs and / or a second preset time period after the first transaction event occurs. The use of this method can improve management efficiency and the accuracy of identifying private cashiering, and reduce the difficulty and cost of managing private cashiering by staff.

Description

Technical Field

[0001] The present application relates to the field of computer vision technology, and in particular to a method, system, computer device and storage medium for identifying private cashiers. Background Art

[0002] In chain stores or restaurants, cash registers are usually installed to process cash. However, in the actual operation of the store, some staff members collect cash privately. For example, staff members receive cash from customers but do not put the cash into the cash box of the store, etc. Instead, they collect the cash privately; or staff members show customers personal payment codes and other payment tools. In the above cases, staff members pocket the money that originally belonged to the store, and the store management system cannot detect these behaviors in time, thus damaging the interests of the store's operation. Summary of the invention

[0003] Based on this, it is necessary to provide a method for identifying private cashiers to address the above technical issues, which can automatically and effectively identify the behavior of staff members collecting cash privately and improve the cashier management of the store.

[0004] In the first aspect, the present application provides a method for identifying private cashiers, the method comprising: acquiring a first image of a first target environment area; wherein the first image includes at least a first detection area, and the first detection area includes a cash register; identifying cash targets and cash box targets in the first detection area, and determining whether the position of the cash target is located within the cash box target based on the position information of the cash target and the cash box target, and if the position of the cash target is not located within the cash box target, determining that a first transaction event has occurred; determining whether a receipt appears in a time period associated with the first transaction event, and if not, outputting an alarm management, wherein the associated time period is a first preset time period before the time when the first transaction event occurs and / or a second preset time period after the time when the first transaction event occurs.

[0005] In one embodiment, the first image also includes a second detection area, and the second detection area overlaps or intersects with the first detection area. The method also includes: identifying whether there is a staff member in the second detection area, and the positional relationship between the staff member and the cash target; if the cash target is within a preset range around the staff member, or there is an intersection between the staff member and the cash target, determining that a second transaction event has occurred; judging whether a receipt appears in a time period associated with the second transaction event, and if not, outputting an alarm instruction, wherein the associated time period is a first preset time period before the second transaction event occurs and / or a second preset time period after the second transaction event occurs.

[0006] In one embodiment, identifying whether there are staff members in the second detection area includes: if it is identified that a human target in the second detection area is wearing a uniform and / or carrying specific items, determining that there are staff members in the second detection area; and / or, predetermining a working area in the second detection area, and if a human target is identified in the working area, determining that there are staff members in the second detection area; and / or, when it is identified that there are human targets in the second detection area, determining whether the human target is facing the first camera, and determining the relative positions of the first camera, the human target, and the cash register; determining whether the human target is a staff member based on the orientation of the human target relative to the first camera and the relative positions of the human target and the cash register; wherein the first camera is used to capture a first image of the first target environment area.

[0007] In one embodiment, based on the orientation of the human target relative to the first camera and the relative position of the human target and the cash register, determining whether the human target is a staff member includes: if the human target is facing away from the first camera and the human target is closer to the first camera relative to the cash register, determining that the human target is a staff member; or, if the human target is facing the first camera and the human target is farther from the first camera relative to the cash register, determining that the human target is a staff member.

[0008] In one embodiment, the first image also includes a third detection area, and the third detection area includes an operation area; the method also includes: if no receipt appears within a third preset time period after the second transaction event occurs, and a staff member is identified in the third detection area, an alarm instruction is output.

[0009] In one embodiment, determining whether the cash target is within a preset range around the staff member includes: enlarging the identification frame of the cash target and / or the staff member by a preset proportion, and determining whether the cash target identification frame and the staff member identification frame after the enlargement by the preset proportion have an intersection, and if so, determining that the cash target is within the preset range around the staff member; and / or, if a receipt appears in the associated time period of the second transaction event, and the cash box target does not pop up within a fourth preset time period after the second transaction event occurs, outputting an alarm instruction.

[0010] In one embodiment, the first camera is configured to be able to identify cash, and the first camera is used to capture a first image of a first target environment area. The method also includes: when the first camera does not identify a cash target in the first detection area, the first camera captures the first image at a first time interval; when the first camera identifies a cash target in the first detection area, the first camera continuously captures the first image at a second time interval; wherein the second time interval is smaller than the first time interval.

[0011] In the second aspect, the present application also provides a method for identifying private cashiers, the method comprising: obtaining a second image of a second target environment area, the second image including at least a fourth detection area, the fourth detection area including a cash register; identifying whether there are at least two portable terminal devices in the fourth detection area, and whether there are at least two human targets; if at least two human targets are identified in the fourth detection area, and one of the portable terminal devices is located within a preset range around the other portable terminal device, determining that a third transaction event has occurred; determining whether a receipt appears in a time period associated with the third transaction event, and if not, outputting an alarm instruction, wherein the associated time period is a preset time period before and / or a preset time period after the occurrence of the third transaction event.

[0012] In one embodiment, the second image also includes a fifth detection area, and the fifth detection area overlaps or intersects with the fourth detection area. The method also includes: identifying whether there is a staff member in the fifth detection area, and the positional relationship between the staff member and the portable terminal device; if the portable terminal device is located within a preset range around the staff member, or there is an intersection between the staff member and the portable terminal device, determining that a fourth transaction event has occurred; judging whether a receipt appears in a time period associated with the fourth transaction event, and if not, outputting an alarm instruction, wherein the associated time period is the fifth preset time period before the occurrence of the fourth transaction event and / or the sixth preset time period after the occurrence of the fourth transaction event.

[0013] In one embodiment, one of the portable terminal devices is located within a preset range around another portable terminal device, including: enlarging the identification frame of at least one portable terminal device by a preset ratio, and determining whether there is an intersection between the identification frames of the two portable terminal devices after the enlargement by the preset ratio, and if so, determining that one of the portable terminal devices is located within the preset range around the other portable terminal device.

[0014] In one embodiment, if at least two human targets are identified in the fourth detection area, and one of the portable terminal devices is located within a preset range around the other portable terminal device, it is determined that a third transaction event has occurred, including: defining one of the two portable terminal devices that is closer to the second camera as a first device and one that is farther from the second camera as a second device, and enlarging the first device by a preset multiple; determining whether the first device, after being enlarged by the first preset multiple, overlaps with a staff member among the two human targets; if so, enlarging the second device by a second preset multiple, the second preset multiple being smaller than the first preset multiple; determining whether the second device, after being enlarged by the second preset multiple, overlaps with a customer among the two human targets, and if so, determining that a third transaction event has occurred.

[0015] In one embodiment, identifying whether there are staff members in the fifth detection area includes: if it is identified that a human target in the fifth detection area is wearing a uniform and / or carrying specific items, determining that there are staff members in the fifth detection area; and / or, predetermining a working area in the fifth detection area, and if a human target is identified in the working area, determining that there are staff members in the fifth detection area; and / or, when it is identified that there are human targets in the fifth detection area, determining whether the human target is facing the second camera, and determining the relative positions of the second camera, the human target, and the cash register; determining whether the human target is a staff member based on the orientation of the human target relative to the second camera and the relative positions of the human target and the cash register; wherein the second camera is used to capture a second image of the second target environment area.

[0016] In one embodiment, based on the orientation of the human target relative to the second camera and the relative position of the human target and the cash register, determining whether the human target is a staff member includes: if the human target is facing away from the second camera and the human target is closer to the second camera relative to the cash register, determining that the human target is a staff member; or, if the human target is facing the second camera and the human target is farther from the second camera relative to the cash register, determining that the human target is a staff member.

[0017] In one embodiment, the second image also includes a sixth detection area, and the sixth detection area includes an operation area; the method also includes: when a receipt appears in the time period associated with the second transaction event, if there is an overlap between the two portable terminal devices in the fourth detection area, and the staff drives away from the fourth detection area to the sixth detection area within a third preset time, an alarm instruction is output.

[0018] In one embodiment, the method also includes: when it is determined that a third transaction event has occurred, if the facial image of one of the human targets is identified to match a facial image in a preset database, or if the uniform of one of the human targets is identified to be a specific work uniform, no alarm instruction is output.

[0019] On the third aspect, the present application also provides a system for identifying private cashiers, the system comprising: a first image acquisition device, used to acquire a first image of a first target environment area; wherein the first image includes at least a first detection area, and the first detection area includes a cash register; a first server, used to identify cash targets and cash box targets in the first detection area, and determine whether the position of the cash target is located inside the cash box target based on the position information of the cash target and the cash box target, if the position of the cash target is not located inside the cash box target, determine that a first transaction event has occurred; and also used to determine whether a receipt appears in a time period associated with the first transaction event, if not, output an alarm instruction, wherein the associated time period is a first preset time period before the time when the first transaction event occurs and / or a second preset time period after that.

[0020] In the fourth aspect, the present application also provides another system for identifying private cashiers, the system comprising: a second image acquisition device, used to acquire a second image of a second target environment area, the second image including at least a fourth detection area, and the fourth detection area including a cash register; a second server, used to identify whether there are at least two portable terminal devices in the fourth detection area, and whether there are at least two human targets; if at least two human targets are identified in the fourth detection area, and one of the portable terminal devices is located within a preset range around the other portable terminal device, it is determined that a third transaction event has occurred; it is also used to determine whether a receipt appears in a time period associated with the third transaction event, and if not, output an alarm instruction, wherein the associated time period is a preset time period before and / or a preset time period after the second transaction event occurs.

[0021] In a fifth aspect, the present application further provides a computer device, comprising a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the steps of any one of the methods in the first aspect or the steps of any one of the methods in the second aspect are implemented.

[0022] In a sixth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of any one of the methods in the first aspect or the steps of any one of the methods in the second aspect.

[0023] The above-mentioned method for identifying private cash collection obtains the cash target and cash box target in the cash register area, and determines whether the cash target is located within the cash box target based on the position information of the cash target and the cash box target, so as to subsequently analyze whether there is any staff member's behavior of private cash collection through cash; if the cash target is not located within the cash box target, and no receipt appears in the associated time period of the first transaction event, it means that the staff member did not conduct cash transactions based on the receipt and did not place the cash in the cash box, that is, the staff member has engaged in private cash collection, so an alarm instruction is output to promptly remind relevant management personnel to handle it.

[0024] The above-mentioned method for identifying private cash collection also identifies whether there are portable terminal devices and human targets in the detection area, so as to subsequently analyze whether there is any staff member using the portable terminal device to collect cash privately; if at least two human targets are identified in the fourth detection area, and one of the portable terminal devices is located within a preset range around the other portable terminal device, and no receipt appears in the associated time period of the third transaction time, it means that the staff member did not use the portable terminal device to conduct the transaction based on the receipt, that is, the staff member has collected cash privately, then an alarm instruction is output to promptly remind relevant management personnel to handle it.

[0025] It can be seen that the technical solution of the present application can not only identify the behavior of staff members performing private cash collection through cash, but also identify the behavior of staff members performing private cash collection through portable terminal devices. It can automatically and effectively identify the behavior of staff members performing private cash collection, avoid or reduce the occurrence of private cash collection in stores, improve the efficiency of store cashier management, improve the accuracy of private cash collection identification, and reduce the difficulty and cost of managing staff members' private cash collection. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 A first flow chart of a method for identifying private cashiering in one embodiment;

[0027] Figure 2 A schematic diagram of a first application scenario of a method for identifying private cashiers in an embodiment;

[0028] Figure 3 A second flow chart of a method for identifying private cashiering in one embodiment;

[0029] Figure 4 A schematic diagram of a second application scenario of a method for identifying private cashiers in an embodiment;

[0030] Figure 5 A schematic diagram of a third application scenario of a method for identifying private cashiers in an embodiment;

[0031] Figure 6 A third flow chart of a method for identifying private cashiering in one embodiment;

[0032] Figure 7 A fourth flow chart of a method for identifying private cashiering in one embodiment;

[0033] Figure 8 A fifth flow chart of a method for identifying private cashiering in one embodiment;

[0034] Fig. 9 A first structural block diagram of a system for identifying private cashiers in one embodiment;

[0035] Fig.10 A second structural block diagram of a system for identifying private cashiers in one embodiment;

[0036] Fig.11 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0037] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0038] In many chain supermarkets, staff members may collect cash privately. This is usually avoided through corporate management methods, but it brings difficulties and costs to corporate management. This application provides a method for identifying private cashiers, which can automatically and effectively identify the behavior of staff members collecting cash privately, thereby improving the efficiency of store cashier management.

[0039] Please refer to Figure 1 The present application embodiment provides a method for identifying private cashiers, which specifically includes the following steps:

[0040] Step S101, obtaining a first image of a first target environment area; wherein the first image at least includes a first detection area, and the first detection area includes a cashier counter.

[0041] Exemplarily, the first target environmental area can be an environmental area within a chain store, restaurant, or the like, and the first target environmental area includes at least a first detection area. For example, the environmental area of ​​a restaurant store is used as the first target environmental area, and the cashier area within the store is used as the first detection area, and the first detection area includes the cashier counter.

[0042] Exemplarily, a first image corresponding to a first target environment area is collected and acquired through a camera installed near the cashier area, image recognition is performed on the first image, a first detection area in the first image is acquired, and target object recognition is performed on the first detection area, and a cashier target in the first detection area is acquired. Based on the cashier target, the first detection area is determined as an area that needs to be subsequently identified and analyzed, and identification and analysis are performed on target objects in the area to determine whether there is a situation where staff members collect cash privately.

[0043] Step S102, identifying the cash target and the cash box target in the first detection area, and determining whether the cash target is located in the cash box target according to the position information of the cash target and the cash box target; if the cash target is not located in the cash box target, determining that a first transaction event occurs.

[0044] In this embodiment, the cash object refers to the cash in the hands of customers or staff, and the cash box object refers to the box at the cash register for storing cash.

[0045] Exemplarily, image recognition is performed on the first detection area to identify the cash target and the cash box target in the first detection area, obtain the position coordinates of the cash target in the first detection area, and obtain the position coordinates of the cash box target in the first detection area. Based on the relative position relationship of the two position coordinates, determine whether there is an overlapping area between the cash target and the cash box target, that is, determine whether the cash target is located in the cash box target. If there is an overlapping area, it indicates that the cash is placed in the cash box. If there is no overlapping area, it indicates that the cash is not placed in the cash box. When it is determined that the position of the cash target is not located in the cash box target, it means that the cash is not placed in the cash box, and it is determined that the first transaction event occurs. It can be understood that the target object in the first detection area can be located and identified by the target detection network model to obtain the cash target and the cash box target, and then obtain the position information of the cash target and the cash box target. The target detection network model can be a neural network model, which is not limited here.

[0046] For example, Figure 2 As shown, in a restaurant, a first image is acquired based on a camera, and the objects in the first image include customers who buy food, staff in the store, cash registers, and other related items. The green frame line in the image represents the first detection area 201 including the cash register. It can be seen that the staff is receiving the cash paid by the customer, and the cash target 202 is in the red frame line. Under normal circumstances, after the staff receives the cash paid by the customer at the cash register, they will put the cash into the cash box at the cash register (the cash box is blocked by the staff in the figure and is not shown), or the staff will take out the cash from the cash box and give the change to the customer. Therefore, according to the location information of the cash and the cash box, it can be determined whether the cash is in the cash box. If the cash is in the cash box, it indicates that the staff may be sorting out the cash in the cash box or a normal cash transaction; if the cash is not in the cash box, it indicates that the staff may receive the cash paid by the customer but not put it in the cash box. Therefore, it is determined that the first transaction event occurs, that is, the potential private cash collection bank is, so as to prepare for further judgment of whether it is private cash collection in the future.

[0047] Step S103, determine whether a receipt appears in the time period associated with the first transaction event, and if not, output an alarm instruction, wherein the associated time period is a first preset time period before the first transaction event occurs and / or a second preset time period after the first transaction event occurs.

[0048] Exemplarily, the first preset time period and the second preset time period can be set to different durations, or can be set to the same duration, which is not limited here. The first preset time period and the second preset time period can be set according to human experience. For example, if the first preset time period and the second preset time period are both set to 1 minute, then if a receipt appears within 1 minute before and / or within 1 minute after the first transaction event occurs, it indicates that the staff member conducts a cash transaction based on the receipt.

[0049] Exemplarily, the cloud server can communicate with the cash register device on the cash register counter, for example, the communication connection can be a wireless connection. If the cloud server receives a receipt sent from the cash register device within the first preset time period and / or the second preset time period thereafter, it means that within the time period, the staff issued a receipt and received the cash target accordingly, that is, a cash transaction under normal circumstances. If the cloud server does not receive the receipt sent from the cash register device within the first preset time period and / or the second preset time period thereafter, it means that within the time period, the staff did not issue a receipt but the staff has received the cash target, and it is determined that the cash transaction was not completed through formal channels, and the staff has engaged in private cash collection, and an alarm instruction is output. This alarm instruction can be to send an alarm message to the management personnel, or to display a warning on the monitoring system to remind the relevant management personnel to take corresponding measures to control the private collection of cash.

[0050] This embodiment obtains the cash target and cash box target in the cash register area, and determines whether the cash target is located within the cash box target based on the position information of the cash target and the cash box target. If the cash target is not located within the cash box target and no receipt voucher appears in the associated time period of the first transaction event, it means that the staff has collected money privately from the bank. An alarm instruction is output to remind relevant management personnel, which can automatically and effectively identify the staff's behavior of collecting money privately, improve the accuracy of identifying private bank receipts, improve the efficiency of store cashier management, and reduce the difficulty and cost of managing staff's private collection of money.

[0051] In one embodiment, the first image also includes a second detection area, and the second detection area overlaps or intersects with the first detection area. Exemplarily, the area position coordinates of the first detection area in the first image are obtained, and the area position coordinates of the second detection area in the first image are obtained. Based on the two area position coordinates, it can be determined that the two detection areas overlap or intersect, so as to identify the target object in the second detection area, so as to more accurately identify the behavior of the staff member collecting money privately.

[0052] like Figure 3 As shown, in this embodiment, the method for identifying private cashiers also includes the following steps:

[0053] Step S301, identifying whether there is a staff member in the second detection area, and the positional relationship between the staff member and the cash target.

[0054] Among them, the second detection area can be represented as the area inside the first target environment area facing the cash register, that is, the activity area where the staff should be located under normal circumstances, so as to evaluate whether the staff is in a reasonable position and the positional relationship between the staff and the cash target.

[0055] Exemplarily, target object recognition is performed on the second detection area to determine that the target object in the area is a staff member, and the position coordinates of the staff member target in the second detection area are obtained to determine the position relationship between the position coordinates and the position coordinates of the cash target. Step S302: If the cash target is within a preset range around the staff member, or there is an intersection between the staff member and the cash target, it is determined that a second transaction event occurs.

[0056] Among them, a certain range can be set according to specific application scenarios and is not limited here.

[0057] Exemplarily, based on the position coordinates of the cash target and the position coordinates of the staff target, the distance between the cash target and the staff target is determined. If the distance is within a threshold range, it is determined that the second transaction event has occurred, or based on the intersection of the position coordinates of the cash target and the position coordinates of the staff target, it is determined that the second transaction event has occurred. If the cash target is within a preset range around the staff, or there is an intersection between the staff and the cash target, it means that the staff may have privately collected the cash, that is, privately collected cash, so it is determined as the second transaction event, so as to further identify whether there is a situation of privately collecting cash.

[0058] Step S303, determine whether a receipt appears in the time period associated with the second transaction event, and if not, output an alarm instruction, wherein the associated time period is a first preset time period before the second transaction event occurs and / or a second preset time period after the second transaction event occurs.

[0059] Exemplarily, the cloud server can communicate with the cash register device on the cash register counter, for example, the communication connection can be a wireless connection. When determining that the second transaction event has occurred, if the cloud server receives a receipt sent from the cash register device within the first preset time period and / or the second preset time period thereafter, it means that within the time period, the staff issued a receipt and received the cash target accordingly, that is, a cash transaction under normal circumstances. When determining that the second transaction event has occurred, if the cloud server does not receive a receipt sent from the cash register device within the first preset time period and / or the second preset time period thereafter, it means that within the time period, the staff did not issue a receipt but received the cash target, which may be that the staff privately sold the goods and took the cash for themselves, that is, privately received the bank's money, therefore, an alarm instruction is output to remind the relevant management personnel.

[0060] In this embodiment, by identifying the staff in the second detection area and the positional relationship between the staff and the cash target, if the cash target is within the preset range around the staff, or there is an intersection between the staff and the cash target, and no receipt appears in the associated time period of the second transaction event, it means that the staff is not conducting cash transactions based on the receipt, and it is confirmed that the staff has direct contact with the cash, that is, a private cash collection situation has occurred, then an alarm instruction is output to remind relevant management personnel, thereby further improving the accuracy of identifying private cash collection and improving management efficiency.

[0061] In one embodiment, identifying whether there are staff members in the second detection area includes: if it is identified that the human target in the second detection area is wearing a uniform and / or carrying a specific item, it is determined that there are staff members in the second detection area. Among them, in the first target environment area, most staff members wear company-specific work uniforms, hats, masks or item lists, etc. The target object recognition in the second detection area can determine whether there are human targets wearing uniforms and / or carrying specific items in the second detection area. If there are, they are determined to be staff members. It is understandable that the staff members generally wear uniforms specified by the store, or wear the store's work badge, etc. Based on this, the target object in the second detection area can be located and identified by the target detection network model to locate the human target, and further identify the human target to determine whether it is a staff member. The target detection network model can be a neural network model, which is not limited here. Therefore, by identifying the uniform and / or carrying specific items, the human target in the second detection area is further identified, and the human target is determined to be a staff member, so as to improve the accuracy of identifying the staff members' private collection of bank.

[0062] In another embodiment, identifying whether there is a staff member in the second detection area further includes: pre-determining a working area in the second detection area, and if a human target is identified in the working area, determining that there is a staff member in the second detection area. The working area refers to an area in the second detection area where staff members are active but customers cannot enter. The working area is pre-determined, and if a human target is identified in the working area, it indicates that the human target is a staff member, so as to improve the accuracy and efficiency of identifying whether there is a staff member in the second detection area.

[0063] In another embodiment, identifying whether there is a staff member in the second detection area further includes: when identifying the presence of a human target in the second detection area, determining whether the human target is facing the first camera, and determining the relative positions of the first camera, the human target and the cashier; determining whether the human target is a staff member based on the orientation of the human target relative to the first camera and the relative positions of the human target and the cashier; wherein the first camera is used to collect the first image of the first target environment area. The human targets in the second detection area may include staff members, customers, etc. Therefore, it is determined whether the human target is facing the camera, and based on the installation position of the first camera, the positional relationship between the first camera, the human target and the cashier is determined, and further determining whether the human target is a staff member. Exemplarily, when the first camera is installed in the first target environment area from the inside to the outside of the door, in the case of a cash transaction, the staff member faces away from the first camera, and the relative distance between the first camera and the staff member is less than the relative distance between the first camera and the cashier. When the first camera is installed in the first target environment area from the outside of the door to the inside, in the case of a cash transaction, the staff member faces the first camera, and the relative distance between the first camera and the staff member is greater than the relative distance between the first camera and the cashier.

[0064] Therefore, based on the relative positions of the first camera, the human target and the cashier counter, the orientation of the human target relative to the first camera, and the relative positions of the human target and the cashier counter, it is possible to determine whether the human target is a staff member, thereby improving the accuracy and efficiency of identifying whether there are staff members in the second detection area.

[0065] It should be noted that the above three implementation methods for identifying whether there are staff members in the second detection area can also be combined with two or more of the implementation methods to further improve the accuracy of identifying whether there are staff members in the second detection area.

[0066] In one embodiment, whether the human target is a staff member is determined based on the orientation of the human target relative to the first camera and the relative position of the human target and the cash register, including: if the human target is facing away from the first camera and the human target is closer to the first camera relative to the cash register, determining that the human target is a staff member.

[0067] In one embodiment, if the human target faces away from the first camera, and the human target is further away from the first camera than the cashier counter, the human target is determined to be a non-staff member.

[0068] For example, Figure 4 As shown, assuming that the first camera 401 is installed in the first target environment area from the inside to the outside of the door, the red arrow in the figure indicates the direction of the current human target. Here, the relative position of the human target and the cashier counter is determined by comparing the shortest distance between the camera and the human target and the shortest distance between the camera and the cashier counter. It is noted that the shortest distance between the camera and the human target can be the distance between the center of the human target (human target recognition frame) and the center of the camera; or it can be the point closest to the human target (human target recognition frame) from the center of the camera as the reference point, and the distance between the reference point and the center of the camera is the shortest distance; similarly, the shortest distance between the camera and the cashier counter can also be determined in the above manner. It can be seen that in the case of a cash transaction, the human target 403 is away from the first camera 401, and the human target 403 is closer to the first camera 401 relative to the cashier counter 402; the human target 404 is facing the first camera 401, and the human target 404 is farther away from the first camera 401 relative to the cashier counter 402, so the human target 403 is determined to be a staff member, and the human target 404 may be a non-staff member, or a customer.

[0069] In another embodiment, determining whether the human target is a staff member is based on the orientation of the human target relative to the first camera and the relative position of the human target and the cash register, and further includes: if the human target is facing the first camera and the human target is farther away from the first camera relative to the cash register, determining that the human target is a staff member.

[0070] In another embodiment, if the human target is facing the first camera and is closer to the first camera than the cashier counter, the human target is determined to be a non-staff member.

[0071] For example, Figure 5As shown, assuming that the first camera 501 is installed in the first target environment area from the outside to the inside of the door, the red arrow in the figure indicates the direction of the current human target. Here, the relative position of the human target and the cashier counter is determined by comparing the shortest distance between the camera and the human target and the shortest distance between the camera and the cashier counter. It can be seen that in the case of a cash transaction, the human target 504 is away from the first camera 501, and the human target 504 is closer to the first camera 501 relative to the cashier counter 502; the human target 503 is facing the first camera 501, and the human target 503 is farther away from the first camera 501 relative to the cashier counter 502. Therefore, it is determined that the human target 503 is a staff member, and the human target 504 is a non-staff member, possibly a customer.

[0072] It should be noted that, regarding the above two implementation methods of determining whether the human target is a staff member based on the orientation of the human target relative to the first camera and the relative position of the human target and the cashier counter, these two implementation methods can also be used at the same time to further improve the accuracy of identifying whether the human target is a staff member.

[0073] In one embodiment, the first image also includes a third detection area, and the third detection area includes an operation area. The method also includes: if no receipt appears within a third preset time period after the second transaction event occurs, and a staff member is identified in the third detection area, an alarm instruction is output.

[0074] The operation area refers to the area where the staff is active when performing work tasks, such as the food and beverage production area, the commodity sorting area, etc. The third preset time period can be set according to the experience value or the needs of the specific scene, and is not limited here. Specifically, if no receipt appears in the third preset time period after the second transaction event, it means that there is an abnormal cash transaction. At the same time, the staff is identified in the third detection area, indicating that there is a private bank collection, that is, no goods are sold, but the staff has collected money, then an alarm instruction is output to remind the relevant personnel.

[0075] By continuously identifying within the third preset time, it is possible to avoid misjudgment of private cashiering at a certain moment, thereby further improving the accuracy of private cashiering identification.

[0076] In one embodiment, determining whether the cash target is within a preset range around the staff member includes: enlarging the identification frame of the cash target and / or the staff member by a preset ratio, and determining whether the cash target identification frame and the staff member identification frame enlarged by the preset ratio have an intersection, and if so, determining that the cash target is within the preset range around the staff member.

[0077] The identification frame can be, but is not limited to, a rectangular frame, a circular frame, etc., which is used to identify the position of the cash target and the staff in the detection area. The preset ratio is a pre-set value used to define the size of a "certain range". For example, if the preset ratio is 1.2, then each side of the identification frame will be extended outward by 20% of its original length. Specifically, if there is an intersection between the two identification frames after the preset ratio is expanded, that is, there is a partial overlap between the boundaries or the internal areas of the two, then the two are close to each other in space, thereby determining that the cash target is within the preset range around the staff.

[0078] In this way, it is possible to effectively determine whether the cash target is within the preset range around the staff, thereby improving the sensitivity and accuracy of identifying private cash collection behavior, and helping to promptly detect and prevent improper private cash collection behavior.

[0079] In another embodiment, the method for identifying private cashiering further includes: if a receipt appears in the associated time period of the second transaction event, and the cash box target does not pop up within the fourth preset time period after the second transaction event occurs, then an alarm instruction is output. Among them, since the staff needs to put the cash into the cash box after the legal transaction is completed, it will cause the cash box to pop up or open. However, if a receipt appears, but the cash box does not pop up, it means that the transaction has occurred but the cash has not been put into the cash box. Therefore, it can be inferred that since the interval between the two transactions is short, the receipt at this time may be the receipt of the previous transaction, that is, there is no corresponding receipt for this transaction and the cash has not been put into the cash box, that is, private cashiering may have occurred, so an alarm instruction is output. In this way, possible private cashiering can be identified more accurately, avoiding the problem of missing or failing to identify private cashiering, thereby reducing the behavior of staff taking the opportunity to conduct private cashiering when there are many transactions.

[0080] It should be noted that the above two implementation methods for determining whether the cash target is within the preset range around the staff can also be combined with the two implementation methods to further improve the accuracy of the determination of whether the cash target is within the preset range around the staff.

[0081] In one embodiment, the first camera is configured to be able to identify cash, and the first camera is used to capture a first image of a first target environment area. The method also includes: when the first camera does not identify a cash target in the first detection area, the first camera captures the first image at a first time interval; when the first camera identifies a cash target in the first detection area, the first camera continuously captures the first image at a second time interval.

[0082] Among them, the second time interval is smaller than the first time interval. The first time interval and the second time interval can be set according to empirical values. Continuous capture means taking several consecutive images or video images lasting for several seconds or animated images lasting for several seconds. Specifically, when the first camera does not recognize the cash target in the first detection area, it means that there is no cash transaction event at present, and there is no need to identify whether there is private cash collection at this time. Therefore, the first camera captures the first image at a longer first time interval to reduce equipment consumption. When the first camera recognizes the cash target in the first detection area, it means that there is a cash transaction event at present, which belongs to the peak time period for private cash collection. Therefore, the first camera continuously captures the first image at a shorter second time interval to ensure that the complete cash transaction process can be accurately monitored to avoid the occurrence of private cash collection.

[0083] like Figure 6 As shown, the embodiment of the present application also provides another method for identifying private cashiers, which specifically includes the following steps:

[0084] Step S601: Acquire a second image of a second target environment area, wherein the second image at least includes a fourth detection area, and the fourth detection area includes a cashier counter.

[0085] Among them, the second target environment area refers to the first target environment area, the second image refers to the first image, and the fourth detection area refers to the first detection area, which will not be repeated here.

[0086] Step S602, identifying whether there are at least two portable terminal devices and at least two human targets in the fourth detection area; if it is identified that there are at least two human targets in the fourth detection area, and one of the portable terminal devices is located within a preset range around the other portable terminal device, it is determined that a third transaction event has occurred.

[0087] Among them, portable terminal devices include smart phones, tablet computers, smart watches, POS machines and other terminal devices with collection and payment functions.

[0088] Exemplarily, the target object is identified in the fourth detection area, and the target objects in the area are determined to be at least two human targets and at least two portable terminal devices. The position coordinates of the two portable terminal devices in the fourth detection area are obtained respectively. Based on the position coordinates of the two, the distance between the two is determined. If the distance is within the threshold range, it is determined that the third transaction time occurs. Based on this, the target object in the fourth detection area can be located and identified by the target detection network model to locate the human target and the portable terminal device. The target detection network model can be a neural network model, which is not limited here. If there are two or more human targets at the cash register in the fourth detection area, and the position of one of the portable terminal devices is within a certain range around another portable terminal device, it means that the staff and the customer may be conducting a transaction at the cash register. The transaction may be a normal transaction or it may be that the staff uses their position to collect cash privately. At this time, it is determined to be a third transaction event, that is, a potential private collection bank is, in preparation for further judgment of whether it is private collection.

[0089] Step S603: determine whether a receipt appears in the time period associated with the third transaction event. If not, output an alarm instruction, wherein the associated time period is a preset time period before and / or a preset time period after the occurrence of the third transaction event.

[0090] Exemplarily, the cloud server can communicate with one of the above-mentioned portable terminal devices, for example, the communication connection can be a wireless connection. It is understandable that in normal transactions in the store, the staff will also conduct transactions with customers through the portable terminal device, which is usually a public device in the store, rather than a personal device, so the cloud server can communicate data with the portable terminal device. When determining that the third transaction event has occurred, if the cloud server receives a receipt sent from the portable terminal device in the preset time period before and / or the preset time period after the occurrence of the third transaction event, it means that within the time period, the staff issued a receipt and received the cash target accordingly, which is determined to be a normal transaction. If the cloud server does not receive the receipt sent by the portable terminal device during the preset time period before and / or the preset time period after the occurrence of the third transaction event, it means that during the time period, the staff did not issue the receipt but collected the money, and it is determined that an abnormal transaction has occurred, that is, the transaction was not completed through formal channels. For example, the staff took advantage of their position to collect money privately or took advantage of preferential treatment to obtain interest rate spreads and other private bank collection behaviors. Therefore, an alarm instruction is output to remind relevant management personnel to take corresponding measures to control the private bank collection behaviors.

[0091] The method for identifying private cashiering in this embodiment identifies whether there are portable terminal devices and human targets in the detection area, so as to subsequently analyze whether there is any staff member using the portable terminal device to perform private cashiering; if at least two human targets are identified in the fourth detection area, and one of the portable terminal devices is located within a preset range around the other portable terminal device, and no receipt appears in the associated time period of the third transaction time, it means that the staff member did not use the portable terminal device to conduct transactions based on the receipt, that is, private cashiering has occurred, then an alarm instruction is output to remind relevant management personnel, thereby improving the accuracy of identifying private cashiering, improving management efficiency, and reducing the difficulty and cost of managing private cashiering by staff in the store.

[0092] In one embodiment, the second image also includes a fifth detection area, and the fifth detection area overlaps or intersects with the fourth detection area. Exemplarily, the area position coordinates of the fourth detection area in the second image are obtained, and the area position coordinates of the fifth detection area in the second image are obtained. Based on the two area position coordinates, it can be determined that the two detection areas overlap or intersect, so as to identify the target object in the fifth detection area, so as to more accurately identify the behavior of the staff member collecting money privately.

[0093] In this embodiment, if Figure 7 As shown, the method for identifying private cashiers also includes:

[0094] Step S701: Identify whether there is a staff member in the fifth detection area, and the position relationship between the staff member and the portable terminal device.

[0095] Exemplarily, target object recognition is performed on the fifth detection area to determine whether there is a staff member in the fifth detection area. The identification of the staff member object can refer to the above-mentioned implementation method, which is not repeated here. By identifying the position coordinates of the staff member and the portable terminal device in the fifth detection area, the positional relationship between the two is determined based on the position coordinates of the two.

[0096] Step S702: If the portable terminal device is located within a preset range around the staff member, or if there is an intersection between the staff member and the portable terminal device, it is determined that a fourth transaction event occurs.

[0097] The distance between the portable terminal device and the staff is determined through image recognition, and based on the distance between the two, it is determined that the portable terminal device is within a preset range around the staff, or that there is an intersection between the staff and the portable terminal device. If it is recognized that the portable terminal device is within a certain range around the staff (for example, a distance threshold is set), or the staff and the portable terminal device have an intersection in the image (that is, the positions of the two overlap in the image), it is determined that a fourth transaction event has occurred, that is, a private bank may have occurred, for example, the staff is using the portable terminal device to perform informal cashier activities.

[0098] Step S703, determine whether a receipt appears in the time period associated with the fourth transaction event, and if not, output an alarm instruction, wherein the associated time period is the fifth preset time period before the occurrence of the fourth transaction event and / or the sixth preset time period after the occurrence of the fourth transaction event.

[0099] The fifth preset time period and the sixth preset time period may be set with reference to the first preset time period and the second preset time period, which will not be described in detail herein.

[0100] Specifically, if a receipt, such as a receipt printed by a cash register or an electronic payment record, appears during the associated time period of the fourth transaction event, it indicates that the transaction is normal. If no receipt is detected during the associated time period of the fourth transaction event, it indicates that during this time period, the staff did not issue a receipt but did collect money, and it is determined that an abnormal transaction has occurred, that is, the transaction was not completed through formal channels, such as the staff taking advantage of their position to collect money privately or taking advantage of preferential treatment to arbitrage interest rate spreads, etc. Therefore, an alarm instruction is output. This instruction can be to send an alarm message to the management personnel, or to display a warning message on the monitoring screen to remind the relevant management personnel to take corresponding measures to control the private collection of money from the bank.

[0101] In this way, potential private bank collection activities can be identified more accurately, especially when staff have specific interactions with portable terminal devices. This not only improves the accuracy of monitoring, but also helps stores take quick action when private bank collection activities occur, thereby improving management efficiency and reducing management difficulty and costs.

[0102] In one embodiment, one of the portable terminal devices is located within a preset range around another portable terminal device, including: enlarging the identification frame of at least one portable terminal device by a preset ratio, and determining whether there is an intersection between the identification frames of the two portable terminal devices after the enlargement by the preset ratio, and if so, determining that one of the portable terminal devices is located within the preset range around the other portable terminal device.

[0103] The identification frame of the portable terminal device refers to the processing process of identifying the cash target and the staff, which will not be repeated here. Specifically, if the identification frames of the two portable terminal devices after being enlarged by a preset ratio have an intersection, that is, the boundaries or internal areas of the two overlap, and the two are close to each other in space, it is determined that one of the portable terminal devices is located within the preset range around the other portable terminal device.

[0104] In this way, it is possible to effectively determine whether one of the portable terminal devices is within a preset range around another portable terminal device, so as to identify potential private bank collection activities, especially in situations where two portable terminal devices need to interact. For example, staff may use one portable terminal device to record transactions while using another portable terminal device to illegally collect money, thereby improving the sensitivity and accuracy of identifying private bank collection activities and helping to promptly detect and prevent improper behavior.

[0105] In one embodiment, Figure 8 As shown, if at least two human targets are identified in the fourth detection area, and one of the portable terminal devices is located within a preset range around the other portable terminal device, it is determined that a third transaction event occurs, including:

[0106] Step S801: define one of the two portable terminal devices that is closer to the second camera as a first device, and define one that is farther from the second camera as a second device, and enlarge the first device by a preset multiple.

[0107] For example, according to the positions of the two portable terminal devices relative to the second camera, the portable terminal device closer to the second camera is defined as the first device, and the device farther from the camera is defined as the second device. The identification frame of the first device is enlarged by a preset multiple. The preset multiple is a pre-set value used to determine the relative position between the device and the human target.

[0108] Step S802, determining whether the first device after being enlarged by a first preset multiple overlaps with the staff members among the two human targets; if so, enlarging the second device by a second preset multiple, the second preset multiple being smaller than the first preset multiple.

[0109] Specifically, if it is determined that the identification frame of the first device after being enlarged by the first preset multiple overlaps with the identification frame of the staff member among the two human targets, it indicates that the staff member may be operating the portable terminal device, therefore, the second device is enlarged by the second preset multiple,

[0110] Step S803: determine whether the second device after being enlarged by a second preset multiple overlaps with the customer among the two human targets, and if so, determine that a third transaction event occurs.

[0111] Specifically, if it is determined that the second device after being enlarged by the second preset multiple overlaps with the customer target, then it is determined that a third transaction event has occurred, that is, a potential private collection bank, for example, a staff member is using the first device to record transactions, while the customer target is using the second device to make illegal payments.

[0112] In this way, potential private banking activities can be more accurately identified, especially when it involves two portable terminal devices and an employee's interaction with a customer target, thereby helping to improve the effectiveness and reliability of identifying private banking activities to protect the interests of the enterprise.

[0113] In one embodiment, identifying whether there are staff members in the fifth detection area includes: if it is identified that the human target in the fifth detection area is wearing a uniform and / or carrying specific items, determining that there are staff members in the fifth detection area. Among them, image recognition technology can be used to detect whether the human target is wearing an employee uniform. Uniforms usually have specific colors, patterns or logos, and these features help to distinguish between staff members and other customer targets. It is also possible to check whether the human target is carrying specific items, such as employee badges, work cards, portable cash registers or other work-related equipment. If it is identified that the human target is wearing a uniform and / or carrying specific items, the human target is determined to be a staff member. In this way, it is possible to effectively identify whether there are staff members in the fifth detection area, improve the accuracy of identification, and thus help to subsequently determine whether private bank collection has occurred.

[0114] In one embodiment, identifying whether there are staff members in the fifth detection area includes: predetermining a working area in the fifth detection area, and if a human target is identified in the working area, determining that there are staff members in the fifth detection area. The working area refers to the area where store staff usually occupy or move when performing commodity transactions, such as a service counter next to a cash register or a cargo sorting area. It can be to set a specific coordinate range or to clarify the boundaries of the working area by means of virtual lines, and mark the working area in the fifth detection area. Specifically, the fifth detection area is monitored in real time, and by identifying the human outline, movement or specific features in the marked working area, it is determined whether there is a human target entering the working area. If a human target is identified in the working area, it is determined that there are staff members in the fifth detection area.

[0115] In this way, it is possible to effectively identify whether there are staff members in the fifth detection area, which helps to monitor the staff members' behavioral patterns and potential private bank collection behavior. Moreover, this method simplifies the identification process, reduces misjudgments through preset work areas, and improves the efficiency and accuracy of identification.

[0116] In one embodiment, identifying whether there is a staff member in the fifth detection area includes: when identifying the presence of a human target in the fifth detection area, determining whether the human target is facing the second camera, and determining the relative positions of the second camera, the human target and the cash register; based on the orientation of the human target relative to the second camera and the relative positions of the human target and the cash register, determining whether the human target is a staff member; wherein the second camera is used to capture a second image of the second target environment area.

[0117] Specifically, the fifth detection area is monitored in real time by the second camera. The orientation of the detected human target is determined by identifying the orientation of the human body contour, facial features or specific body parts (such as shoulders and head). The angle between the human target and the second camera is calculated to determine whether the target is facing the camera directly or sideways. The position of the human target in the image is determined and compared with its known cashier counter position. By comparing the relative position between the human target and the cashier counter, the system can determine whether the human target is near the cashier counter or in its working area. Based on the orientation of the human target relative to the second camera and the relative position of the human target and the cashier counter, it is determined whether the human target is a staff member. If the human target is facing the second camera and is located in the working area near the cashier counter, the system will be more likely to determine that the human target is a staff member. If the human target is facing away from the second camera or is located far away from the cashier counter, the system may determine it as a non-staff member or require further verification.

[0118] It should be noted that the above three implementation methods for identifying whether there are staff members in the fifth detection area can also be combined with any two of the implementation methods or all three implementation methods to further improve the accuracy of identifying whether there are staff members in the fifth detection area.

[0119] In one embodiment, based on the orientation of the human target relative to the second camera and the relative position of the human target and the cashier counter, determining whether the human target is a staff member includes: if the human target faces away from the second camera and the human target is closer to the second camera than the cashier counter, determining that the human target is a staff member. Figure 4 The processing of the first camera will not be repeated here.

[0120] In another embodiment, based on the orientation of the human target relative to the second camera and the relative position of the human target and the cashier counter, determining whether the human target is a staff member includes: if the human target is facing the second camera and the human target is farther from the second camera than the cashier counter, determining that the human target is a staff member. Figure 5 The processing of the first camera will not be repeated here.

[0121] In one embodiment, the second image also includes a sixth detection area, and the sixth detection area includes an operation area. The method for identifying private cashiering also includes: in the case where a receipt such as a POS slip appears in the time period associated with the second transaction event, if there is an overlap between two portable terminal devices in the fourth detection area, and the staff moves away from the fourth detection area to the sixth detection area within the third preset time, then an alarm instruction is output. Among them, the operation area includes places such as commodity preparation and background operation area. The appearance of a POS slip indicates that a legal transaction has occurred. Usually, when a legal transaction occurs, the staff needs to take the item from the operation area to the cashier counter and hand it over to the customer. However, if the two portable terminal devices overlap in the cashier counter area, and the staff moves from the cashier counter to the operation area, this may be that the staff privately trades to obtain discounts, rather than trading with customers, that is, suspicious private bank collection occurs, so an alarm instruction is output. In this way, possible private bank collection can be identified more accurately, avoiding the problem of missing or failing to identify private bank collection, thereby reducing the private bank collection of staff private transactions to obtain discounts.

[0122] By monitoring the use of portable terminal devices and the behavior patterns of staff, if the overlap of portable terminal devices is detected and the staff quickly leaves the cashier area and goes to the operation area after the transaction occurs, this may indicate that the staff is trying to hide or process unrecorded transactions.

[0123] In one embodiment, the method for identifying private cashiers further includes: in the case of determining that the third transaction event occurs, if the facial image of one of the human targets is identified to match the facial image in the pre-set database, or the uniform of one of the human targets is identified to be a specific work uniform, then no alarm instruction is output. Specifically, a face database including authorized staff or management personnel can be pre-set, and the facial image of the human target in the fourth detection area is compared with the pre-set face database using face recognition technology. If the facial image is identified to match a face in the database, or the uniform of a human target is identified to be a specific work uniform, it means that the human target is an authorized staff member and the transaction is normal, not a private cashier, so no alarm instruction is output. In this way, the pre-set database and specific work uniforms are used to distinguish between customers, unauthorized staff and authorized staff, thereby reducing false alarms and ensuring that the alarm instruction is only for private cashiers with higher suspicion, thereby improving the accuracy and recognition efficiency of identifying private cashiers.

[0124] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0125] Based on the same inventive concept, the embodiment of the present application also provides a system for implementing the above-mentioned method for identifying private cashiers. The implementation scheme for solving the problem provided by the system is similar to the implementation scheme recorded in the above-mentioned method, so the specific limitations in the embodiment of the system for identifying private cashiers provided below can be referred to the limitations of the method for identifying private cashiers above, and will not be repeated here.

[0126] In one embodiment, Fig. 9 As shown, a system for identifying private cashiers is provided, the system comprising:

[0127] The first image acquisition device is used to acquire a first image of a first target environment area; wherein the first image at least includes a first detection area, and the first detection area includes a cash register.

[0128] The first server is used to identify the cash target and the cash box target in the first detection area, determine whether the cash target is located inside the cash box target based on the position information of the cash target and the cash box target, and if the cash target is not located inside the cash box target, determine that a first transaction event has occurred; and is also used to determine whether a receipt appears in a time period associated with the first transaction event, and if not, output an alarm instruction, wherein the associated time period is a first preset time period before the time when the first transaction event occurs and / or a second preset time period after the time when the first transaction event occurs.

[0129] In one embodiment, the first image also includes a second detection area, and the second detection area overlaps or intersects with the first detection area. The first server is also used to identify whether there is a staff member in the second detection area, and the positional relationship between the staff member and the cash target; if the cash target is within a preset range around the staff member, or there is an intersection between the staff member and the cash target, it is determined that a second transaction event has occurred; it is determined whether a receipt appears in a time period associated with the second transaction event, and if not, an alarm instruction is output, wherein the associated time period is a first preset time period before the second transaction event occurs and / or a second preset time period after the second transaction event occurs.

[0130] In one embodiment, the first server is further used to determine that there are staff members in the second detection area if it is identified that the human target in the second detection area is wearing a uniform and / or carrying specific items; and / or, predetermine the working area in the second detection area, and determine that there are staff members in the second detection area if a human target is identified in the working area; and / or, when it is identified that there is a human target in the second detection area, determine whether the human target is facing the first image acquisition device, and determine the relative positions of the first image acquisition device, the human target, and the cash register; determine whether the human target is a staff member based on the orientation of the human target relative to the first image acquisition device and the relative position of the human target and the cash register; wherein the first image acquisition device is used to capture a first image of the first target environment area.

[0131] In one embodiment, the first server is further used to determine that the human target is a staff member if the human target is away from the first image acquisition device and is closer to the first image acquisition device relative to the cash register; or, if the human target is facing the first image acquisition device and is farther away from the first image acquisition device relative to the cash register, determine that the human target is a staff member.

[0132] In one embodiment, the first image also includes a third detection area, and the third detection area includes an operation area. The first server is also used to output an alarm instruction if no receipt appears within a third preset time period after the second transaction event and a staff member is identified in the third detection area.

[0133] In one embodiment, the first server is also used to enlarge the identification box of the cash target and / or the staff member by a preset ratio, and determine whether the cash target identification box and the staff member identification box after the enlargement by the preset ratio have an intersection. If so, it is determined that the cash target is within a preset range around the staff member; and / or, if a receipt appears in the associated time period of the second transaction event, and the cash box target does not pop up within a fourth preset time period after the second transaction event, an alarm instruction is output.

[0134] In one embodiment, the first image acquisition device is configured to be able to identify cash. The first image acquisition device is used to capture a first image of a first target environment area. When the first image acquisition device does not identify a cash target in the first detection area, the first image acquisition device captures the first image at a first time interval; when the first image acquisition device identifies a cash target in the first detection area, the first image acquisition device continuously captures the first image at a second time interval; wherein the second time interval is smaller than the first time interval.

[0135] In one embodiment, Fig.10As shown, the present application also provides a system for identifying private cashiers, the system comprising:

[0136] The second image acquisition device is used to acquire a second image of the second target environment area, the second image at least includes a fourth detection area, and the fourth detection area includes a cashier counter.

[0137] The second server is used to identify whether there are at least two portable terminal devices and at least two human targets in the fourth detection area; if at least two human targets are identified in the fourth detection area, and one of the portable terminal devices is located within a preset range around the other portable terminal device, it is determined that a third transaction event has occurred; it is also used to determine whether a receipt appears in a time period associated with the third transaction event, and if not, output an alarm instruction, wherein the associated time period is a preset time period before and / or a preset time period after the time when the second transaction event occurs.

[0138] In one embodiment, the second image also includes a fifth detection area, and the fifth detection area overlaps or intersects with the fourth detection area. The second server is also used to identify whether there is a staff member in the fifth detection area, and the positional relationship between the staff member and the portable terminal device; if the portable terminal device is located within a preset range around the staff member, or there is an intersection between the staff member and the portable terminal device, determine that a fourth transaction event has occurred; determine whether a receipt appears in the time period associated with the fourth transaction event, and if not, output an alarm instruction, wherein the associated time period is the fifth preset time period before the fourth transaction event occurs and / or the sixth preset time period after.

[0139] In one embodiment, the second server is further used to enlarge the identification frame of at least one portable terminal device by a preset ratio, determine whether there is an intersection between the identification frames of the two portable terminal devices after the enlargement by the preset ratio, and if so, determine that one of the portable terminal devices is located within a preset range around the other portable terminal device.

[0140] In one embodiment, the second server is also used to define the two portable terminal devices that are closer to the second image acquisition device as the first device and the device that is farther away from the second image acquisition device as the second device, and to enlarge the first device by a preset multiple; determine whether the first device after being enlarged by the first preset multiple overlaps with the staff among the two human targets; if so, enlarge the second device by a second preset multiple, and the second preset multiple is smaller than the first preset multiple; determine whether the second device after being enlarged by the second preset multiple overlaps with the customer among the two human targets, and if so, determine that a third transaction event has occurred.

[0141] In one embodiment, the second server is also used to determine that there are staff members in the fifth detection area if it is identified that the human target in the fifth detection area is wearing a uniform and / or carrying specific items; and / or, predetermine the working area in the fifth detection area, and if a human target is identified in the working area, determine that there are staff members in the fifth detection area; and / or, when a human target is identified in the fifth detection area, determine whether the human target is facing the second image acquisition device, and determine the relative positions of the second image acquisition device, the human target, and the cash register; determine whether the human target is a staff member based on the orientation of the human target relative to the second image acquisition device and the relative position of the human target and the cash register; wherein the second image acquisition device is used to capture a second image of the second target environment area.

[0142] In one embodiment, whether the human target is a staff member is determined based on the orientation of the human target relative to the second image acquisition device and the relative position of the human target and the cash register, including: if the human target is away from the second image acquisition device and is closer to the second image acquisition device relative to the cash register, the human target is determined to be a staff member; or, if the human target is facing the second image acquisition device and is farther away from the second image acquisition device relative to the cash register, the human target is determined to be a staff member.

[0143] In one embodiment, the second image also includes a sixth detection area, which includes an operation area; the second server is also used to output an alarm instruction when a receipt such as a POS receipt appears in a time period associated with the second transaction event, if there is an overlap between two portable terminal devices in the fourth detection area, and the staff moves away from the fourth detection area to the sixth detection area within a third preset time.

[0144] In one embodiment, the second server is also used to, when determining that a third transaction event has occurred, not output an alarm instruction if the facial image of one of the human targets is identified to match a facial image in a preset database, or if the uniform of one of the human targets is identified to be a specific work uniform.

[0145] Each module in the above-mentioned system for identifying private cashiers can be implemented in whole or in part by software, hardware, or a combination thereof. Each of the above-mentioned modules can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in a computer device in the form of software, so that the processor can call and execute operations corresponding to each of the above modules.

[0146] In one embodiment, a computer device is provided. The computer device includes a memory and a processor. The memory stores a computer program. When the processor executes the computer program, the steps in the above method embodiments are implemented. The internal structure diagram of the computer device can be as follows: Fig.11 As shown, the computer device includes a processor, a memory and a network interface connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store data. The network interface of the computer device is used to communicate with an external terminal through a network connection.

[0147] Those skilled in the art will understand that Fig.11 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0148] In one embodiment, a computer-readable storage medium is further provided, on which a computer program is stored. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented.

[0149] A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory may include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive memory, magnetic memory, ferroelectric memory, phase change memory, graphene memory, etc. Volatile memory may include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in the embodiments provided in this application may include at least one of a relational database and a non-relational database. The non-relational database may include a distributed database based on blockchain, etc., but is not limited thereto. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic unit, a data processing logic unit based on quantum computing, etc., but is not limited thereto.

[0150] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0151] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A method for identifying private cashiers, characterized in that: The method comprises: Acquire a first image of a first target environment area; wherein the first image at least includes a first detection area, and the first detection area includes a cashier counter; Identify the cash target and the cash box target in the first detection area, determine whether the cash target is located in the cash box target according to the position information of the cash target and the cash box target, and determine that a first transaction event occurs if the cash target is not located in the cash box target; Determine whether a receipt voucher appears in a time period associated with the first transaction event, and if not, output an alarm instruction, wherein the associated time period is a first preset time period before the occurrence of the first transaction event and / or a second preset time period after the occurrence of the first transaction event; Wherein, the first image also includes a second detection area and a third detection area, the second detection area overlaps or intersects with the first detection area, and the third detection area includes an operation area, and the method further includes: Identify whether there is a staff member in the second detection area, and the positional relationship between the staff member and the cash target; if the cash target is within a preset range around the staff member, or there is an intersection between the staff member and the cash target, determine that a second transaction event occurs; If the receipt does not appear within a third preset time period after the second transaction event occurs, and the staff member is identified in the third detection area, an alarm instruction is output.

2. The method for identifying private cashier according to claim 1, characterized in that: The method further comprises: Determine whether a receipt appears in the time period associated with the second transaction event. If not, output an alarm instruction, wherein the associated time period is a first preset time period before the occurrence of the second transaction event and / or a second preset time period after the occurrence of the second transaction event.

3. The method for identifying private cashier according to claim 2, characterized in that: Identifying whether there is a staff member in the second detection area includes: If it is identified that a human target in the second detection area is wearing a uniform and / or carrying a specific item, it is determined that there is a staff member in the second detection area; and / or, Predetermine a working area in the second detection area, and if a human target is identified in the working area, determine that there is a worker in the second detection area; and / or, When a human target is identified in the second detection area, determine whether the human target is facing the first camera, and determine the relative positions of the first camera, the human target and the cashier counter; based on the orientation of the human target relative to the first camera and the relative positions of the human target and the cashier counter, determine whether the human target is a staff member; wherein the first camera is used to capture a first image of the first target environment area.

4. The method for identifying private cashier according to claim 3, characterized in that: Determining whether the human target is a staff member based on the orientation of the human target relative to the first camera and the relative position of the human target and the cashier counter includes: If the human target is facing away from the first camera and is closer to the first camera than the cashier counter, it is determined that the human target is the staff member; or, If the human target is facing the first camera and is farther away from the first camera than the cashier counter, it is determined that the human target is a staff member.

5. The method for identifying private cashier according to claim 2, characterized in that: Determining that the cash target is within a preset range around the staff member includes: Enlarging the identification frame of the cash target and / or the staff member by a preset ratio, determining whether the cash target identification frame after the enlargement by the preset ratio and the identification frame of the staff member intersect, and if so, determining that the cash target is within a preset range around the staff member; and / or, If a receipt appears in the time period associated with the second transaction event, and the cash box target does not pop up within a fourth preset time period after the second transaction event occurs, an alarm instruction is output.

6. The method for identifying private cashiering according to claim 1, characterized in that: The first camera is configured to be able to recognize the cash, and the first camera is used to capture a first image of a first target environment area. The method further includes: When the first camera fails to identify the cash object in the first detection area, the first camera captures the first image at a first time interval; When the first camera recognizes the cash object in the first detection area, the first camera continuously captures the first image at a second time interval; The second time interval is smaller than the first time interval.

7. A method for identifying private cashiers, characterized in that: Acquire a second image of a second target environment area, wherein the second image at least includes a fourth detection area, and the fourth detection area includes a cashier counter; Identify whether there are at least two portable terminal devices and at least two human targets in the fourth detection area; if it is identified that there are at least two human targets in the fourth detection area, and one of the portable terminal devices is located within a preset range around the other portable terminal device, determine that a third transaction event occurs; Determine whether a receipt voucher appears in the time period associated with the third transaction event, and if not, output an alarm instruction, wherein the associated time period is a preset time period before and / or a preset time period after the occurrence of the third transaction event; Wherein, the second image also includes a fifth detection area, and the fifth detection area overlaps or intersects with the fourth detection area, and the method further includes: Identify whether there is a staff member in the fifth detection area, and the positional relationship between the staff member and one of the portable terminal devices; If the portable terminal device is within a preset range around the staff member, or there is an intersection between the staff member and the portable terminal device, it is determined that a fourth transaction event occurs; Determine whether a receipt appears in the time period associated with the fourth transaction event. If not, output an alarm instruction. The associated time period is the fifth preset time period before the occurrence of the fourth transaction event and / or the sixth preset time period after the occurrence of the fourth transaction event.

8. The method for identifying private cashiering according to claim 7, characterized in that: One of the portable terminal devices is located within a preset range around another portable terminal device, including: The identification frame of at least one of the portable terminal devices is enlarged by a preset ratio, and it is determined whether there is an intersection between the identification frames of the two portable terminal devices after being enlarged by the preset ratio. If so, it is determined that one of the portable terminal devices is located within a preset range around the other portable terminal device.

9. The method for identifying private cashiering according to claim 8, characterized in that: If at least two human targets are identified in the fourth detection area, and one of the portable terminal devices is located within a preset range around the other portable terminal device, determining that a third transaction event occurs includes: The portable terminal device closer to the second camera is defined as the first device, and the portable terminal device farther from the second camera is defined as the second device, and the first device is enlarged by a preset multiple; wherein the second camera is used to capture a second image of the second target environment area; Determining whether the first device after being enlarged by a first preset multiple overlaps with a worker among the two human targets; If yes, enlarging the second device by a second preset multiple, the second preset multiple being smaller than the first preset multiple; It is determined whether the second device after being enlarged by a second preset multiple overlaps with the customer among the two human targets, and if so, it is determined that a third transaction event occurs.

10. The method for identifying private cashiering according to claim 9, characterized in that: Identifying whether there is a staff member in the fifth detection area includes: If it is identified that a human target in the fifth detection area is wearing a uniform and / or carrying a specific item, it is determined that there is a staff member in the fifth detection area; and / or, Predetermine a working area in the fifth detection area, and if a human target is identified in the working area, determine that there is a worker in the fifth detection area; and / or, When a human target is identified in the fifth detection area, determine whether the human target is facing the second camera, and determine the relative positions of the second camera, the human target and the cash register; based on the orientation of the human target relative to the second camera and the relative positions of the human target and the cash register, determine whether the human target is a staff member; wherein the second camera is used to capture a second image of the second target environment area.

11. The method for identifying private cashiering according to claim 10, characterized in that: Determining whether the human target is a staff member based on the orientation of the human target relative to the second camera and the relative position of the human target and the cashier counter includes: If the human target is facing away from the second camera and is closer to the second camera than the cashier counter, it is determined that the human target is the staff member; or, If the human target is facing the second camera and the human target is farther away from the second camera than the cashier counter, it is determined that the human target is a staff member.

12. The method for identifying private cashiers according to claim 9, characterized in that: The second image also includes a sixth detection area, and the sixth detection area includes an operation area; The method also includes: when a receipt appears in the time period associated with the third transaction event, if there is overlap between two portable terminal devices in the fourth detection area, and the staff member drives away from the fourth detection area to the sixth detection area within the third preset time, an alarm instruction is output.

13. The method for identifying private cashiers according to claim 7, characterized in that: The method further comprises: When it is determined that a third transaction event has occurred, if the facial image of one of the human targets is identified to match a facial image in a preset database, or if the uniform of one of the human targets is identified to be a specific work uniform, no alarm instruction is output.

14. A system for identifying private cashiers, characterized in that: The system comprises: A first image acquisition device, used to acquire a first image of a first target environment area; wherein the first image at least includes a first detection area, and the first detection area includes a cashier counter; The first server is used to identify the cash target and the cash box target in the first detection area, determine whether the position of the cash target is located in the cash box target according to the position information of the cash target and the cash box target, and determine that a first transaction event occurs if the position of the cash target is not located in the cash box target; and is also used to determine whether a receipt voucher appears in a time period associated with the first transaction event, and if not, output an alarm instruction, wherein the associated time period is a first preset time period before the occurrence of the first transaction event and / or a second preset time period after the occurrence of the first transaction event; Among them, the first image also includes a second detection area and a third detection area, the second detection area overlaps or intersects with the first detection area, and the third detection area includes an operation area. The first server is also used to identify whether there is a staff member in the second detection area, and the positional relationship between the staff member and the cash target; if the cash target is within a preset range around the staff member, or there is an intersection between the staff member and the cash target, it is determined that a second transaction event has occurred; if the receipt does not appear within a third preset time period after the second transaction event occurs, and the staff member is identified in the third detection area, an alarm instruction is output.

15. A system for identifying private cashiers, characterized in that: The system comprises: A second image acquisition device is used to acquire a second image of a second target environment area, wherein the second image at least includes a fourth detection area, and the fourth detection area includes a cashier counter; The second server is used to identify whether there are at least two portable terminal devices and at least two human targets in the fourth detection area; if it is identified that there are at least two human targets in the fourth detection area, and one of the portable terminal devices is located within a preset range around the other portable terminal device, it is determined that a third transaction event occurs; and it is also used to determine whether a receipt voucher appears in a time period associated with the third transaction event, and if not, output an alarm instruction, wherein the associated time period is a preset time period before and / or a preset time period after the occurrence of the third transaction event; Among them, the second image also includes a fifth detection area, and the fifth detection area overlaps or intersects with the fourth detection area. The second server is also used to identify whether there is a staff member in the fifth detection area, and the positional relationship between the staff member and one of the portable terminal devices; if the portable terminal device is located within a preset range around the staff member, or there is an intersection between the staff member and the portable terminal device, it is determined that a fourth transaction event has occurred; it is determined whether a receipt appears in a time period associated with the fourth transaction event, and if not, an alarm instruction is output, and the associated time period is the fifth preset time period before the occurrence of the fourth transaction event and / or the sixth preset time period after the occurrence.

16. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 or the steps of the method according to any one of claims 7 to 13 are implemented.

17. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 or the steps of the method according to any one of claims 7 to 13 are implemented.

Images