A method and system for standardizing monitoring of chain stores

Abstract

The embodiment of the specification provides a chain store standardization monitoring method and system, the method comprises: acquiring standardization business processes and personnel information of a chain store; constructing a standardization image set based on the standardization business processes and the personnel information; acquiring scene features of a monitoring area; determining monitoring parameters based on the scene features; performing real-time monitoring on the monitoring area based on the monitoring parameters to obtain at least one monitoring image; and determining non-standard behaviors based on the standardization image set and the at least one monitoring image.

Description

Technical Field

[0001] This manual relates to the field of image processing, and in particular to a standardized monitoring method and system for chain stores. Background Technology

[0002] Chain stores refer to stores that operate under the organization and leadership of a headquarters, adopting a common business strategy and consistent marketing actions, and organically combining centralized procurement and decentralized sales. Chain store employees generally require standardized training and their behavior is subject to standardized supervision. However, as the number of chain stores increases, the number of employees also increases, making employee behavior management more difficult. Using surveillance equipment for standardized monitoring of employee behavior and performing frame-by-frame analysis of surveillance video leads to a large amount of data processing required.

[0003] Therefore, it is desirable to provide a standardized monitoring method and system for chain stores to achieve effective supervision of employee behavior while reducing computational load. Summary of the Invention

[0004] This specification provides one or more embodiments of a standardized monitoring method for chain stores. The method includes: acquiring standardized business processes and personnel information of the chain stores; constructing a standardized image set based on the standardized business processes and personnel information; acquiring scene features of the monitoring area; determining monitoring parameters based on the scene features; performing real-time monitoring of the monitoring area based on the monitoring parameters to obtain at least one monitoring image; and determining non-standard behavior based on the standardized image set and at least one monitoring image.

[0005] This specification provides one or more embodiments of a standardized monitoring system for chain stores. The system includes: a construction module for acquiring standardized business processes and personnel information of chain stores; constructing a standardized image set based on the standardized business processes and personnel information; a monitoring module for acquiring scene features of the monitoring area; determining monitoring parameters based on the scene features; performing real-time monitoring of the monitoring area based on the monitoring parameters to obtain at least one monitoring image; and a judgment module for judging non-standard behaviors based on the standardized image set and at least one monitoring image.

[0006] This specification provides one or more embodiments of a standardized monitoring device for chain stores, the device including a processor, the processor being used to execute any of the standardized monitoring methods for chain stores described in the above embodiments.

[0007] This specification provides one or more embodiments of a computer-readable storage medium that stores computer instructions. When a computer reads the computer instructions in the storage medium, the computer executes the standardized monitoring method for chain stores as described in any of the above embodiments. Attached Figure Description

[0008] This specification will be further described by way of exemplary embodiments, which will be described in detail with reference to the accompanying drawings. These embodiments are not limiting; in these embodiments, the same reference numerals denote the same structures, wherein:

[0009] Figure 1 This is a schematic diagram illustrating the application scenarios of a standardized monitoring system for chain stores, based on some embodiments of this specification.

[0010] Figure 2 This is an exemplary block diagram of a standardized monitoring system for chain stores, as shown in some embodiments of this specification;

[0011] Figure 3 This is an exemplary flowchart of a standardized monitoring method for chain stores, as shown in some embodiments of this specification.

[0012] Figure 4 This is an exemplary flowchart illustrating the determination of the capture frequency according to some embodiments of this specification;

[0013] Figure 5 This is an exemplary schematic diagram illustrating the determination of capture resolution based on a decision model according to some embodiments of this specification. Detailed Implementation

[0014] To more clearly illustrate the technical solutions of the embodiments in this specification, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are merely some examples or embodiments of this specification. For those skilled in the art, these drawings can be applied to other similar scenarios without creative effort. Unless obvious from the context or otherwise specified, the same reference numerals in the drawings represent the same structures or operations.

[0015] It should be understood that the terms “system,” “device,” “unit,” and / or “module” used herein are one way to distinguish different components, elements, parts, sections, or assemblies at different levels. However, if other terms can achieve the same purpose, they may be replaced by other expressions.

[0016] As indicated in this specification and claims, unless the context clearly indicates otherwise, the words "a," "an," "an," and / or "the" do not specifically refer to the singular and may also include the plural. Generally speaking, the terms "comprising" and "including" only indicate the inclusion of expressly identified steps and elements, which do not constitute an exclusive list, and the method or apparatus may also include other steps or elements.

[0017] Flowcharts are used in this specification to illustrate the operations performed by the system according to embodiments of this specification. It should be understood that the preceding or following operations are not necessarily performed in exact order. Instead, the steps can be processed in reverse order or simultaneously. Furthermore, other operations can be added to these processes, or one or more steps can be removed from them.

[0018] Figure 1 These are schematic diagrams illustrating application scenarios of a standardized monitoring system for chain stores, based on some embodiments of this specification. For example... Figure 1 As shown, the standardized monitoring system 100 for chain stores may include a processing device 110, a storage device 120, a network 130, a user terminal 140, and a monitoring device 150.

[0019] Processing device 110 can be used to process information and / or data related to the standardized monitoring system for chain stores. For example, processing device 110 can access information and / or data stored in storage device 120, user terminal 140, and / or monitoring device 150 via network 130.

[0020] In some embodiments, the processing device 110 can access standardized business processes and personnel information stored in the user terminal 140 via the network 130, and construct a standardized image set based on the standardized business processes and personnel information. In some embodiments, the processing device 110 can access monitoring images acquired by the monitoring device 150 via the network 130.

[0021] In some embodiments, the processing device 110 may be a single server or a group of servers. The group of servers may be a centralized group of servers connected to the network 130 via an access point, or a distributed group of servers connected to the network 130 via one or more access points.

[0022] Storage device 120 can be used to store data and / or instructions. In some embodiments, storage device 120 can store personnel information acquired from user terminal 140. In some embodiments, storage device 120 can store surveillance images acquired from monitoring device 150. In some embodiments, storage device 120 can store data and / or instructions used by processing device 110 to execute or use in order to complete the exemplary standardized monitoring system for chain stores described herein. Storage device 120 may include one or more storage components, each of which may be a separate device or part of other devices. In some embodiments, storage device 120 may include random access memory (RAM), read-only memory (ROM), mass storage, removable memory, volatile read-write memory, etc., or any combination thereof. Exemplarily, mass storage may include disks, optical disks, solid-state drives, etc. In some embodiments, storage device 120 may be implemented on a cloud platform. By way of example only, the cloud platform may include private cloud, public cloud, hybrid cloud, community cloud, distributed cloud, internal cloud, multi-tiered cloud, etc., or any combination thereof.

[0023] Network 130 can connect the various components of the system and / or connect the system to external resources. Network 130 enables communication between the components and with other parts outside the system, facilitating the exchange of data and / or information.

[0024] In some embodiments, the processing device 110 can obtain standardized business processes and personnel information from the user terminal 140 via the network 130. In some embodiments, the processing device 110 can send control commands to the monitoring device 150 via the network 130. For example, the processing device 110 can control the capture frequency of the monitoring device 150 via the network 130.

[0025] In some embodiments, network 130 can be any one or more of wired or wireless networks. In some embodiments, the network can be a point-to-point, shared, centralized, or other topologies, or a combination of multiple topologies. In some embodiments, network 130 may include one or more network access points. For example, network 130 may include wired or wireless network access points, such as base stations and / or network switching points, through which one or more components of the chain store standardized monitoring system 100 can connect to network 130 to exchange data and / or information.

[0026] User terminal 140 may include one or more terminals or software used by the user. In some embodiments, the user may be the owner of user terminal 140. In some embodiments, user terminal 140 may include mobile device 140-1, laptop computer 140-2, tablet computer 140-3, etc., or any combination thereof. In some embodiments, the user may obtain monitoring images and non-compliant behavior determination results through user terminal 140.

[0027] The monitoring device 150 can be a device used to monitor a monitored area in real time and acquire monitoring images. In some embodiments, the monitoring device 150 may include cameras, etc. In some embodiments, the monitoring device 150 may include multiple cameras to achieve real-time monitoring of multiple monitored areas.

[0028] It should be noted that the standardized monitoring system 100 for chain stores is provided for illustrative purposes only and is not intended to limit the scope of this specification. Those skilled in the art can make various modifications or variations based on the description in this specification. For example, the standardized monitoring system 100 for chain stores can be implemented on other devices to achieve similar or different functions. However, such changes and modifications will not depart from the scope of this specification.

[0029] Figure 2 This is an exemplary block diagram of a standardized monitoring system for chain stores, as shown in some embodiments of this specification. In some embodiments, the standardized monitoring system 200 for chain stores may include a construction module 210, a monitoring module 220, and a decision module 230.

[0030] Module 210 can be used to obtain standardized business processes and personnel information for chain stores. More information on obtaining standardized business processes and personnel information for chain stores can be found in step 310 and its related description.

[0031] In some embodiments, the construction module 210 can also be used to construct a standardized image set based on standardized business processes and personnel information. For more information on constructing the standardized image set, please refer to step 320 and its related description.

[0032] In some embodiments, the construction module 210 is further configured to construct a standardized image subset corresponding to the standardized business sub-process based on the standardized business sub-process and personnel information. Each standardized image subset includes an action start point determination set and an action result determination set. For more details on constructing the standardized image subset corresponding to the standardized business sub-process, please refer to step 320 and its related description.

[0033] The monitoring module 220 can be used to acquire scene features of the monitored area. For more information on acquiring scene features of the monitored area, please refer to step 330 and its related description.

[0034] In some embodiments, the monitoring module 220 can also be used to determine monitoring parameters based on scene characteristics. For more information on determining monitoring parameters, please refer to step 340 and its related description.

[0035] In some embodiments, the monitoring module 220 can also be used to perform real-time monitoring of the monitored area based on monitoring parameters to obtain at least one monitoring image. For more information on obtaining at least one monitoring image, please refer to step 350 and its related description.

[0036] In some embodiments, the monitoring module 220 is further configured to identify the current business sub-process and at least one main object in the monitored area based on scene characteristics; and determine the capture frequency based on the operational characteristics of the at least one main object. More information on determining the capture frequency can be found in [link to relevant documentation]. Figure 4 And its related descriptions.

[0037] In some embodiments, the monitoring module 220 is further configured to process at least one monitoring image based on a decision model, wherein the decision model is a machine learning model; determine whether the confidence level meets a preset condition; and in response, increase the capture resolution within a preset time period. More information on increasing the capture resolution within a preset time period can be found at [link to relevant documentation]. Figure 5 And its related descriptions.

[0038] The determination module 230 can be used to determine non-standard behavior based on a standardized image set and at least one surveillance image. More details on determining non-standard behavior can be found in step 360 and its related description.

[0039] It should be noted that the above description of the standardized monitoring system and its modules for chain stores is for convenience only and should not be construed as limiting this specification to the scope of the illustrated embodiments. It is understood that those skilled in the art, after understanding the principles of the system, may arbitrarily combine the various modules or construct subsystems connected to other modules without departing from these principles. In some embodiments, Figure 2 The construction module 210, monitoring module 220, and determination module 230 disclosed herein can be different modules within a single system, or a single module can implement the functions of two or more of the aforementioned modules. For example, the modules can share a single storage module, or each module can have its own separate storage module. Such variations are all within the scope of protection of this specification.

[0040] Figure 3This is an exemplary flowchart of a standardized monitoring method for chain stores according to some embodiments of this specification. Process 300 can be executed by the standardized monitoring system 100 or the standardized monitoring system 200 for chain stores. For example, process 300 can be stored as instructions in a storage medium (e.g., storage device 120), processing device 110, and / or Figure 2 The module in the process can execute the instruction, and when executing the instruction, the processing device 110 and / or the module can be configured to execute process 300. The operation of the process shown below is for illustrative purposes only. In some embodiments, process 300 may be accomplished using one or more additional operations not described and / or without one or more operations discussed. Additionally, Figure 3 The order of operations shown in the diagram and described below for process 300 is not restrictive.

[0041] Step 310: Obtain standardized business processes and personnel information for the chain stores. In some embodiments, step 310 may be performed by the construction module 210.

[0042] In some embodiments, chain stores may include, but are not limited to, chain supermarkets, chain milk tea shops, and chain restaurants.

[0043] Standardized business processes refer to operational procedures related to the operation of chain stores that have certain standardized requirements. For example, standardized business processes may include, but are not limited to, the procedures for making milk tea or baking pizza. The standardization requirements refer to the requirements that ensure operators (e.g., chain store employees) receive the same or similar services and / or products when repeatedly following the same standardized business process.

[0044] In some embodiments, the standardized business processes of chain stores can be pre-defined by management personnel. The construction module 210 can obtain the standardized business processes of chain stores in various ways. For example, each chain store and its standardized business processes can be pre-saved in a database, and the construction module 210 can retrieve the corresponding standardized business processes from the database based on the chain store's enterprise information.

[0045] Personnel information refers to information related to the identity and physical characteristics of relevant operational personnel in chain stores. For example, personnel information may include information such as the operator's height, facial features (e.g., face shape), and body characteristics (e.g., shoulder width, arm length).

[0046] In some embodiments, the construction module 210 can acquire personnel information in various ways. For example, the construction module 210 can prompt the operator to manually input relevant information (e.g., personnel identity) at the user terminal. Another example is that the construction module 210 can acquire image data including the operator using a camera device, and determine the operator's personnel information (e.g., physical characteristics) by recognizing the image data.

[0047] Step 320: Construct a standardized image set based on standardized business processes and personnel information. In some embodiments, step 320 may be performed by the construction module 210.

[0048] A standardized image set can refer to a collection of images taken for each operator during standardized operations within a standardized business process. For example, a standardized image set could include a collection of images taken for operator A during standardized operations in a standardized business process for making milk tea.

[0049] In some embodiments, different standardized image sets can be used for different operators. For example, the standardized image set corresponding to operator A in the standardized business process of making milk tea may be different from the standardized image set corresponding to operator B in the same process. In some embodiments, different standardized image sets can be used for different standardized business processes. For example, the standardized image set corresponding to operator A in the standardized business process of making milk tea may be different from the standardized image set corresponding to operator A in the standardized business process of taking a milk tea order.

[0050] In some embodiments, different operators may be responsible for different standardized business processes. For example, operators can be categorized according to the standardized business processes they are responsible for, with each category of operators responsible for one standardized business process. Correspondingly, each category of operators may correspond to a standardized image set.

[0051] In some embodiments, the construction module 210 can construct a standardized image set in various ways. For example, the construction module 210 can capture and score the operation process of each operator multiple times in accordance with the standardized business process, and determine the image set captured during the operation process where the scoring result meets the scoring threshold or the highest scoring result as the standardized image set.

[0052] In some embodiments, a standardized business process may include at least one standardized business sub-process, and a standardized image set may include at least one standardized image subset. Each standardized business sub-process may correspond to one standardized image subset.

[0053] In some embodiments, a standardized business process can consist of multiple operational steps. For example, a standardized business process for making milk tea may include steps such as preparing the tea base, adding milk, weighing the ingredients, and adding the toppings. Each operational step, with its own standardized requirements, can be referred to as a standardized sub-process. For instance, the operational step of "preparing the tea base" in the above example, with its own standardized requirements, constitutes a standardized sub-process.

[0054] A standardized image subset can refer to the set of images taken for each operator during a standardized operation in a standardized business sub-process. For example, a standardized image subset could include the set of images taken for operator A during a standardized operation in the "making tea base" standardized business sub-process.

[0055] In some embodiments, the construction module 210 can construct a standardized image subset corresponding to the standardized business sub-process based on the standardized business sub-process and personnel information. The construction method of the standardized image subset is similar to that of the standardized image set; for details, please refer to the construction process of the standardized image set described above, which will not be repeated here.

[0056] In some embodiments, each standardized image subset may include an action start point determination set and an action result determination set.

[0057] An action start point determination set refers to the set of images captured when an operator performs a standardized business sub-process, specifically the multiple actions performed at the beginning of the process (hereinafter referred to as the start action sequence). For example, when operator A performs the standardized business sub-process of "using seasonings," the multiple actions performed at the beginning of the process include "going to the seasoning area," "opening the lid of the seasoning box," and "taking the seasonings out of the seasoning box with a spoon." The corresponding action start point determination set includes images captured when operator A goes to the seasoning area, images captured when operator A opens the lid of the seasoning box, and images captured when operator A takes the seasonings out of the seasoning box with a spoon.

[0058] An action result determination set refers to the set of images captured after an operator completes a standardized business sub-process, representing multiple actions (hereinafter referred to as the result action sequence). For example, in the standardized business sub-process of "using seasonings," the multiple actions performed by operator A after completing the entire process include "putting the spoon back into the seasoning box" and "closing the lid of the seasoning box." The corresponding action start point determination set includes images captured when operator A puts the spoon back into the seasoning box and images captured when operator A closes the lid of the seasoning box.

[0059] In some embodiments, the action start point determination set and the action result determination set may include positive samples and negative samples, wherein positive samples are images corresponding to correct operations that conform to standardized operations, and negative samples are images corresponding to non-standard operations that do not conform to standardized operations. In some embodiments, the action start point determination set and the action result determination set can be used to train a determination model. More information on training the determination model can be found in [link to documentation]. Figure 5 And its related descriptions.

[0060] In some embodiments, the construction module 210 can construct the action start point determination set and the action result determination set in various ways. For example, the construction module 210 can determine the action start point determination set and the action result determination set by manual annotation during the construction of a standardized image subset.

[0061] In some embodiments described herein, by breaking down standardized business processes into standardized business sub-processes, and further into action start-point determination sets and action result determination sets, the difficulty of constructing image sets can be simplified. At the same time, it can quickly determine whether the operator has started or completed the standardized business process, effectively improving the accuracy of subsequent judgments on non-standard behaviors.

[0062] In some embodiments, the number of images in a standardized image subset may be related to the importance of a standardized business subprocess and the number of main objects in the standardized business subprocess.

[0063] The importance of a standardized business sub-process refers to the degree of importance of that standardized business sub-process to the entire standardized business process. In some embodiments, the importance of a standardized business sub-process can be represented by a real number between 0 and 1, with a larger value indicating higher importance.

[0064] In some embodiments, the importance of a standardized business sub-process can be obtained in various ways. For example, the importance of a standardized business sub-process can be determined based on its time consumption (e.g., the longer the time consumption, the higher the importance). Alternatively, the importance of a standardized business sub-process can be preset based on prior knowledge or historical data.

[0065] The primary object of a standardized business sub-process refers to the operator performing the operation within that sub-process. For example, in the standardized business sub-process of "using seasonings," the primary object could be operator A, who is currently using seasonings. Similarly, in the business sub-process of "kitchen cleaning," the primary object could include multiple operators currently cleaning.

[0066] In some embodiments, the number of main objects in a standardized business sub-process can be determined in various ways. For example, the number of main objects in a standardized business sub-process can be determined based on a pre-defined number of personnel requirements within that sub-process. Another example is that the number of main objects in a standardized business sub-process can be determined based on image recognition of a corresponding standardized image subset.

[0067] In some embodiments, the number of images in a standardized image subset can be positively correlated with the importance of the standardized business sub-process and the number of main objects in the standardized business sub-process. For example, the higher the importance of the standardized business sub-process, the more images are in the corresponding standardized image subset. Similarly, the more main objects a standardized business sub-process has, the more images are in the corresponding standardized image subset.

[0068] In some embodiments described herein, by relating the number of images in a standardized image subset to the importance of a standardized business subprocess and the number of main objects in the standardized business subprocess, the standardized image subset of a standardized business subprocess with high importance and / or a large number of main objects has a larger number of images, which is beneficial to improving the accuracy of subsequent non-standard behavior judgment.

[0069] In some embodiments, the number of images in the action start point determination set and the action result determination set may be related to the first misjudgment rate and the second misjudgment rate.

[0070] The first false positive rate (FPR) refers to the probability of an error occurring when determining the current business sub-process using the judgment model. The current business sub-process refers to the business process being performed by an operator as monitored by the monitoring equipment. For example, the current business sub-process could be the process monitored by the monitoring equipment where operator A is performing the "using seasonings" task. More information about the judgment model and the FPR can be found in [link to relevant documentation]. Figure 5 And related explanations.

[0071] The second false positive rate refers to the probability of making an error when identifying non-compliant behavior using a decision model. More information on the second false positive rate can be found at [link to relevant documentation]. Figure 5 And its related descriptions.

[0072] In some embodiments, the number of images in the action start point determination set and the action result determination set can be positively correlated with the first misclassification rate and the second misclassification rate. For example, when the first misclassification rate is higher, the number of images in the action start point determination set can be increased. Similarly, when the second misclassification rate is higher, the number of images in the action start point determination set and the action result determination set can be increased.

[0073] In some embodiments of this specification, by relating the number of images in the action start point determination set and the action result determination set to the first misjudgment rate and the second misjudgment rate, the first misjudgment rate and the second misjudgment rate can be effectively reduced, thereby improving the accuracy of the final non-standard behavior determination.

[0074] Step 330: Obtain scene features of the monitored area. In some embodiments, step 330 may be performed by the monitoring module 220.

[0075] A monitored area refers to an area that monitoring equipment can capture and monitor in real time. In some embodiments, a chain store may include multiple monitored areas, and each monitored area may include one or more monitoring devices. For example, when the chain store is a chain hair salon, the monitored areas may include the haircutting area, the shampooing area, the reception area, the rest area, etc.

[0076] Scene features refer to the relevant characteristics of the monitored area. For example, scene features may include the light intensity of the monitored area and the frequency of image changes. The frequency of image changes refers to the number of times the monitored image changes within a unit of time. An image change refers to a change in the content of two consecutive frames. For example, the frequency of image changes could be 5 times within 1 minute.

[0077] In some embodiments, the monitoring module 220 can acquire scene features in various ways. For example, the monitoring module 220 can determine the light intensity of the monitored area by analyzing the brightness of the monitoring images acquired by the monitoring device (such as a camera). Another example is that the monitoring module 220 can determine the frequency of image changes in the monitored area by comparing the content of monitoring images captured by the monitoring device within a preset time period. The length of the preset time period can be manually set.

[0078] Step 340: Determine monitoring parameters based on scene characteristics. In some embodiments, step 340 may be performed by monitoring module 220.

[0079] Monitoring parameters refer to the shooting parameters of monitoring equipment. For example, monitoring parameters may include the capture frequency and capture resolution of the monitoring equipment. The capture frequency refers to the number of times the monitoring equipment captures images per unit of time. For example, the capture frequency could be 100 images captured by the monitoring equipment in 1 second. The capture resolution refers to the amount of information stored in the monitoring image captured by the monitoring equipment. For example, the capture resolution could be 1.2 million pixels.

[0080] Monitoring parameters can be determined in several ways. For example, when the light intensity in the monitored area is low, the capture resolution of the monitoring equipment can be increased. Or, when the frequency of changes in the monitored area's image is high, the capture frequency of the monitoring equipment can be increased.

[0081] In some embodiments, the monitoring module 220 can identify the current business sub-process and at least one main object in the monitored area based on scene characteristics; and determine the capture frequency based on the operation characteristics of at least one main object. For more information on determining the capture frequency, please refer to [link to relevant documentation]. Figure 4 And its related descriptions.

[0082] In some embodiments, the monitoring module 220 can process at least one monitoring image based on a decision model to determine the confidence level of the irregular behavior; determine whether the confidence level meets a preset condition; and then, increase the capture resolution within a preset time period. More information on determining the capture resolution can be found in [link to relevant documentation]. Figure 5 And its related descriptions.

[0083] Step 350: Perform real-time monitoring of the monitored area based on monitoring parameters to obtain at least one monitoring image. In some embodiments, step 350 can be performed by the monitoring module 220.

[0084] Surveillance images refer to images captured in real time by monitoring equipment. For example, a surveillance image could be an image taken by a camera of operator A during the process of "making milk tea".

[0085] In some embodiments, the monitoring device continuously or intermittently captures images based on monitoring parameters to obtain at least one monitoring image. In some embodiments, multiple monitoring devices may be installed in a monitoring area to capture images of that area and obtain at least one monitoring image.

[0086] In some embodiments, the monitoring images can be captured by a monitoring device, and the monitoring module 220 can communicate with the monitoring device to obtain at least one monitoring image captured by the monitoring device.

[0087] Step 360: Determine non-standard behavior based on a standardized image set and at least one monitoring image. In some embodiments, step 360 may be performed by the determination module 230.

[0088] Non-standard behavior refers to actions taken by operators that do not follow the standardized operating procedures corresponding to the standardized business process. For example, in the operation of "using seasonings," the standardized procedure should be to use a spoon to take the seasonings, then return the spoon to the seasoning container and close the lid. If the operator fails to return the spoon to the seasoning container and / or fails to close the lid after taking the seasonings, this is considered non-standard behavior.

[0089] In some embodiments, the determination module 230 can determine non-standard behavior in a variety of ways. For example, the determination module 230 can compare at least one monitoring image with a standardized image set one by one, and when a monitoring image that does not conform to the content of the standardized image set is found, it is determined to be non-standard behavior.

[0090] In some embodiments, the determination module 230 can further process at least one monitoring image based on the determination model to determine irregular behavior. More information on determining irregular behavior based on a determination model can be found at [link to relevant documentation]. Figure 5 And its related descriptions.

[0091] In some embodiments of this specification, non-standard behavior of operators is determined based on captured surveillance images and standardized image sets constructed for each operator. This can improve the accuracy of determination while reducing the amount of data processing, which is conducive to standardized supervision of operator behavior.

[0092] Figure 4 This is an exemplary flowchart illustrating the determination of capture frequency according to some embodiments of this specification. Process 400 can be executed by the chain store standardized monitoring system 100 or the chain store standardized monitoring system 200. For example, process 400 can be stored as instructions in a storage medium (e.g., storage device 120), processing device 110, and / or Figure 2 The module in the process can execute the instruction, and when executing the instruction, the processing device 110 and / or the module can be configured to execute process 400. The operation of the process shown below is for illustrative purposes only. In some embodiments, process 400 may be accomplished using one or more additional operations not described and / or without one or more operations discussed. Additionally, Figure 4 The order of operations shown in the diagram and described below for process 300 is not restrictive.

[0093] Step 410: Based on scene characteristics, identify the current business sub-process and at least one main object in the monitored area.

[0094] In some embodiments, scene features may further include object features. Object features may refer to features related to the main object in the surveillance image. In some embodiments, object features may include object action features related to the actions performed by the main object and object appearance features related to the appearance of the main object.

[0095] In some embodiments, object motion features and object physical features can be determined using image recognition algorithms. For example, image recognition algorithms can be used to process surveillance images to determine object motion features and object physical features. Exemplary image recognition algorithms may include neural network image recognition, nonlinear dimensionality reduction image recognition, etc.

[0096] In some embodiments, the monitoring module 220 can determine the current business sub-process and at least one main object in the monitored area by searching a database based on object action features and object appearance features. The database includes multiple business sub-processes corresponding to reference object action features and multiple objects corresponding to reference object appearance features. By searching the database, reference object action features that match the object action features and reference object appearance features that match the object appearance features can be obtained. Then, the business sub-process corresponding to the reference object action features can be determined as the current business sub-process, and the object corresponding to the reference object appearance features can be determined as the main object.

[0097] Step 420: Determine the capture frequency based on the operational characteristics of at least one main object.

[0098] Operational characteristics refer to features related to an operator's actions when executing the current business sub-process. For example, an operational characteristic could be the operator's proficiency in executing the current business sub-process. In some embodiments, an operator's proficiency may include average proficiency and primary proficiency. Average proficiency refers to the operator's proficiency across all business sub-processes. Primary proficiency refers to the operator's proficiency in the current business sub-process. Average proficiency and primary proficiency can be represented by values ​​between 0 and 1; a higher value indicates a higher level of proficiency.

[0099] In some embodiments, the monitoring module 220 can obtain the average proficiency level based on various methods. For example, the average proficiency level can be determined based on the total number of times the operator exhibits non-standard behavior within a preset time period. The length of the preset time period can be set manually based on experience. For example, if an operator performs business sub-process A 10 times in a day or month, and the total number of non-standard behaviors is 1, then their average proficiency level could be (10-1) / 10 = 0.9.

[0100] In some embodiments, the primary proficiency level can be determined based on the percentage distribution of non-standard behaviors exhibited by an operator. This percentage distribution refers to the proportion of times a non-standard behavior occurs in a specific business sub-process among the multiple business sub-processes the operator is responsible for, relative to the total number of non-standard behaviors exhibited by the operator within a preset time period. For example, if operator A is responsible for business sub-processes A, B, and C, and the number of times non-standard behaviors occur are 10, 7, and 8 respectively, then when operator A is currently executing business sub-process A, the corresponding primary proficiency level could be 10 / (10+7+8) = 0.4.

[0101] In some embodiments, the monitoring module 220 can determine the capture frequency based on the average proficiency of the main object and a preset rule. In some embodiments, an exemplary preset rule may be: determining the capture frequency according to the proficiency range in which the average proficiency of the main object falls. Each proficiency range can be preset (e.g., based on prior knowledge or historical data) with a capture frequency. For example, assuming that the proficiency ranges include proficiency range A[0,0.2], proficiency range B[0.2,0.4], proficiency range C[0.4,0.6], proficiency range D[0.6,0.8], and proficiency range E[0.8,1], and the capture frequencies corresponding to proficiency ranges E and E are 60 times / min, 40 times / min, 30 times / min, 20 times / min, and 10 times / min, respectively, when the proficiency range in which the average proficiency of the main object A falls is proficiency range D, the capture frequency can be determined to be 20 times / min.

[0102] In some embodiments, after determining the capture frequency based on the average proficiency of the main object and preset rules, the monitoring module 220 can further adjust the capture frequency based on the average proficiency of the main object and its master proficiency in the current business sub-process. For example, when the master proficiency of the main object in the current business sub-process is greater than its average proficiency, the capture frequency of the monitoring device can be reduced, and vice versa.

[0103] In some embodiments, the monitoring module 220 may also determine the capture frequency based on the stability of the confidence level of the irregular behavior.

[0104] The confidence level of non-standard behavior refers to the accuracy with which the model determines that the operator's behavior is non-standard. In some embodiments, the confidence level of non-standard behavior can be represented by a value between 0 and 1; the higher the value, the more accurate the model is in determining that the operator's behavior is non-standard.

[0105] In some embodiments, the confidence level of the non-compliant behavior can be obtained based on the processing of at least one surveillance image by a decision model. More information on determining the confidence level of non-compliant behavior can be found at [link to relevant documentation]. Figure 5 And its related descriptions.

[0106] The stability of the confidence level of non-standard behavior refers to the degree of change in the confidence level of non-standard behavior across multiple judgments by the judgment model.

[0107] In some embodiments, the monitoring module 220 can determine the stability of the confidence level of non-standard behavior in various ways. For example, the monitoring module 220 can process the confidence levels of multiple non-standard behaviors output by the judgment model within a preset time period based on a preset formula to determine the stability of the confidence level of the non-standard behavior. An exemplary preset formula can be a formula for calculating variance and / or standard deviation. For example, if the confidence levels of multiple non-standard behaviors output by the judgment model within 2 hours are 0.8, 0.8, 0.9, and 0.7, respectively, the stability of the confidence level of the non-standard behavior can be determined to be 0.005 using the variance calculation formula.

[0108] In some embodiments, when the stability of the confidence level of the irregular behavior exceeds a stability threshold, the monitoring module 220 can reduce the capture frequency of the monitoring device. The stability threshold can be a system default value, an empirical value, a manually preset value, or any combination thereof, and can be set according to actual needs; this specification does not impose any restrictions on it.

[0109] In some embodiments of this specification, the higher the stability of the confidence level of non-standard behavior in images captured multiple times consecutively, the more stable the features in the image related to the determination of non-standard behavior are. Therefore, the capture frequency can be appropriately reduced to save monitoring costs and reduce the amount of data processing.

[0110] In some embodiments of this specification, the current business sub-process and main object of the monitoring area are determined by scene characteristics, and the capture frequency is determined by the operation characteristics of the main object. The capture frequency is increased for operators with low proficiency to obtain more monitoring images, which facilitates standardized supervision of their behavior. This is beneficial for accurately judging whether the behavior of operators with low proficiency is standardized, and can effectively reduce the amount of data processing.

[0111] It should be noted that the above descriptions of processes 300 and 400 are for illustrative purposes only and do not limit the scope of this specification. Those skilled in the art can make various modifications and changes to processes 300 and 400 under the guidance of this specification. However, such modifications and changes remain within the scope of this specification.

[0112] Figure 5 This is an exemplary schematic diagram illustrating the determination of capture resolution based on a decision model according to some embodiments of this specification.

[0113] In some embodiments, such as Figure 5 As shown, the determination module 230 can process at least one monitoring image 510 based on the determination model 530 to determine the confidence level 550 of the non-standard behavior.

[0114] The decision model can be a machine learning model used to determine the confidence level of non-standard behavior. For example, the decision model can include one or any combination of neural network (NN) models, convolutional neural network (CNN) models, etc.

[0115] In some embodiments, such as Figure 5 As shown, the input to the judgment model 530 may include at least one surveillance image 510, and the output includes a confidence level 550 for the non-standard behavior. More information about the surveillance images can be found in step 350 and its related description, and more information about the confidence level for the non-standard behavior can be found in step 420 and its related description.

[0116] In some embodiments, the input to the decision model 530 may further include monitoring parameters 520. More information about the monitoring parameters can be found in step 340 and its related description.

[0117] In some embodiments of this specification, monitoring parameters are used as inputs to the judgment model, and capture frequency and capture resolution are introduced as computational quantities, making the confidence level of the final output of non-standard behavior more valuable for reference.

[0118] In some embodiments, the decision model may consist of multiple processing layers. For example... Figure 5 As shown, the judgment model 530 may include a business sub-process determination layer 531 and an irregular behavior determination layer 532.

[0119] The business sub-process determination layer can be used to identify the current business sub-process corresponding to the monitored image. In some embodiments, the business sub-process determination layer can be a deep neural network model, etc.

[0120] In some embodiments, such as Figure 5 As shown, the input to the business sub-process determination layer 531 may include at least one monitoring image 510, and the output may include the current business sub-process 540. In some embodiments, such as Figure 5 As shown, the inputs to the business subprocess determination layer 531 may also include monitoring parameters 520.

[0121] The non-standard behavior determination layer can be used to identify non-standard behaviors. In some embodiments, the non-standard behavior determination layer can be a deep neural network model, etc.

[0122] In some embodiments, such as Figure 5 As shown, the input to the non-standard behavior determination layer 532 may include at least one monitoring image 510 and the current business sub-process 540, and the output may include the confidence level of the non-standard behavior 550. In some embodiments, such as Figure 5As shown, the input to the non-standard behavior determination layer 532 may also include monitoring parameters 520.

[0123] In some embodiments, the business sub-process determination layer and the non-standard behavior determination layer can be obtained by training them separately, and a judgment model can be obtained based on the trained business sub-process determination layer and non-standard behavior determination layer.

[0124] In some embodiments, the business sub-process determination layer can be obtained by training multiple first training samples with first labels. For example, multiple first training data with first labels can be input into the initial business sub-process determination layer, and a loss function can be constructed using the first labels and the output of the initial business sub-process determination layer. The parameters of the initial business sub-process determination layer are then iteratively updated based on the loss function. When the loss function of the initial business sub-process determination layer meets a preset condition, the model training is complete, and the trained business sub-process determination layer is obtained. The preset condition can be loss function convergence, the number of iterations reaching a threshold, etc.

[0125] In some embodiments, the non-standard behavior determination layer can be obtained by training multiple second training samples with second labels. The training process of the non-standard behavior determination layer is similar to that of the business sub-process determination layer, and will not be described in detail here.

[0126] In some embodiments, the first training sample may include sample images from the action start point determination set of each sample standardized image subset in the sample standardized image set. These sample images may include images of different resolutions. Images of different resolutions can be images captured at different capture resolutions, or images captured at the same capture resolution and then manually processed to obtain images of different resolutions. In some embodiments, the first label may be the business sub-process corresponding to the sample image in the action start point determination set of each sample standardized image subset. The first label may be obtained through manual annotation. For example, the business sub-processes corresponding to each sample image can be labeled with different classification numbers.

[0127] In some embodiments, the second training sample may include multiple sample business sub-processes and sample images from the action result judgment set within the sample standardized image subset corresponding to that sample business sub-process. These sample images may also include sample images of different resolutions. The acquisition method for images of different resolutions is similar to that described above and will not be repeated here. In some embodiments, the second label may be manually assigned to the images. For example, images exhibiting non-standard behavior are labeled as 1, and images without non-standard behavior are labeled as 0.

[0128] In some embodiments, when the judgment model processes at least one monitoring image to determine its current business sub-process and whether there is any irregular behavior, there is a first misjudgment rate and a second misjudgment rate.

[0129] The first false positive rate refers to the probability that an error occurs when the decision model determines the current business sub-process. For example, the actual current business sub-process in the surveillance image is "using seasonings," but the decision model determines that the business sub-process is another business sub-process (e.g., "making pizza") based on the processing of the surveillance image. In this case, the decision model has made an error in determining the current business sub-process.

[0130] In some embodiments, the first misjudgment rate can be determined based on the historical judgment data of the judgment model. In some embodiments, the historical judgment data includes the number of times the judgment model misjudged the current business sub-process (hereinafter referred to as the first misjudgment count) and the total number of historical judgments by the judgment model. The historical judgment data can be determined by manually reviewing the monitoring images (e.g., playing back the monitoring images and manually confirming whether the judgment model misjudged), comparing the current business sub-process output by the judgment model multiple times with the actual current business sub-process. For example, the first misjudgment count of the judgment model in the historical judgment data can be divided by the total number of historical judgments by the judgment model to obtain the first misjudgment rate. As an example, if the first misjudgment count of the judgment model is 2 and the total number of historical judgments is 20, the first misjudgment rate can be determined to be 2 / 20 = 0.1.

[0131] The second misjudgment rate refers to the probability of an error occurring when the judgment model determines that an operator has engaged in non-standard behavior. For example, the judgment model might determine, based on surveillance footage, that operator A failed to close the lid of the condiment box while using seasonings, thus constituting non-standard behavior. However, in reality, the employee's operation was perfectly standard and no non-standard behavior occurred. In this case, the judgment model has made an error in determining that the operator engaged in non-standard behavior.

[0132] In some embodiments, similar to the determination of the first misjudgment rate, the second misjudgment rate can be determined based on the historical judgment data of the judgment model. In some embodiments, the historical judgment data also includes the number of times the judgment model misjudged the operator's non-standard behavior (hereinafter referred to as the second misjudgment count) and the total number of historical judgments of the judgment model. The historical judgment data can be determined by manually reviewing the monitoring images (e.g., playing back the monitoring images and manually confirming whether the judgment model has made a misjudgment), comparing the multiple outputs of the judgment model regarding whether the operator has non-standard behavior with the actual occurrence of non-standard behavior by the operator. For example, the second misjudgment count of the judgment model in the historical judgment data can be divided by the total number of historical judgments of the judgment model to obtain the second misjudgment rate. As an example, when the second misjudgment count of the judgment model is 1 and the total number of historical judgments is 20, the second misjudgment rate can be determined to be 1 / 20 = 0.05.

[0133] In some embodiments, such as Figure 5 As shown, the judgment module 230 can determine whether the confidence level 550 of the non-standard behavior meets the preset conditions; in response, the capture resolution within the preset time period is increased.

[0134] In some embodiments, the preset condition may be that the confidence level of the non-compliant behavior is greater than a first threshold and less than a second threshold, wherein the first threshold is less than the second threshold. For example, assuming the first threshold is 0.5 and the second threshold is 0.8, when the confidence level of the non-compliant behavior is 0.7, it can be determined that the confidence level of the non-compliant behavior meets the preset condition.

[0135] The first and second thresholds can be determined in various ways. For example, the first and second thresholds can be system default values, empirical values, manually preset values, or any combination thereof, and can be set according to actual needs.

[0136] In some embodiments, the first threshold and the second threshold can be determined based on a second false positive rate. For example, when the second false positive rate is high, the first threshold and the second threshold can be increased, and vice versa.

[0137] In some embodiments, different current business sub-processes may have different first thresholds and second thresholds. For example, for the current business sub-process of "using seasonings", the first threshold and the second threshold may be set to 0.5 and 0.8, respectively. As another example, for the current business sub-process of "making pizza", the first threshold and the second threshold may be set to 0.6 and 0.9, respectively.

[0138] A preset time period refers to a future period of time. The length of the preset time period can be set manually. For example, the preset time period could be the next month.

[0139] In some embodiments, when the confidence level of the non-standard behavior meets a preset condition, the determination module 230 can increase the capture resolution within a preset time period based on the increase in capture resolution. The increase in capture resolution can be related to the operational characteristics of the main object. For example, the lower the proficiency of the main object, the greater the increase in capture resolution of the monitoring device.

[0140] In some embodiments of this specification, by relating the increase in capture resolution to the operational characteristics of the main object, the monitoring images of less skilled operators can be made clearer, which helps to improve the accuracy of judging non-standard behavior.

[0141] In some embodiments of this specification, a judgment model is used to process monitoring images to determine the operator's current business sub-process and non-standard behavior, which helps improve supervision efficiency. Furthermore, based on the relationship between the confidence level of non-standard behavior and preset conditions, the capture resolution is adjusted, improving the accuracy of non-standard behavior judgment.

[0142] The basic concepts have been described above. Obviously, for those skilled in the art, the detailed disclosure above is merely illustrative and does not constitute a limitation of this specification. Although not explicitly stated herein, those skilled in the art may make various modifications, improvements, and corrections to this specification. Such modifications, improvements, and corrections are suggested in this specification and therefore remain within the spirit and scope of the exemplary embodiments described herein.

[0143] Furthermore, this specification uses specific terms to describe embodiments thereof. For example, "an embodiment," "one embodiment," and / or "some embodiments" refer to a particular feature, structure, or characteristic associated with at least one embodiment of this specification. Therefore, it should be emphasized and noted that references to "an embodiment," "one embodiment," or "an alternative embodiment" in different locations throughout this specification do not necessarily refer to the same embodiment. Moreover, certain features, structures, or characteristics in one or more embodiments of this specification can be appropriately combined.

[0144] Furthermore, unless expressly stated in the claims, the order of processing elements and sequences, the use of numbers and letters, or other names described in this specification are not intended to limit the order of the processes and methods described herein. Although various examples have been discussed in the foregoing disclosure of some embodiments of the invention that are currently considered useful, it should be understood that such details are for illustrative purposes only, and the appended claims are not limited to the disclosed embodiments; rather, the claims are intended to cover all modifications and equivalent combinations that conform to the spirit and scope of the embodiments described herein. For example, while the system components described above can be implemented using hardware devices, they can also be implemented solely using software solutions, such as installing the described system on existing servers or mobile devices.

[0145] Similarly, it should be noted that, in order to simplify the description disclosed herein and thus aid in the understanding of one or more embodiments of the invention, the foregoing description of embodiments in this specification may sometimes combine multiple features into a single embodiment, drawing, or description thereof. However, this method of disclosure does not imply that the subject matter of this specification requires more features than those mentioned in the claims. In fact, the embodiments contain fewer features than all the features of a single embodiment disclosed above.

[0146] In some embodiments, numbers describing the quantity of components and attributes are used. It should be understood that such numbers used in the description of embodiments are modified in some examples with the terms "approximately," "approximately," or "generally." Unless otherwise stated, "approximately," "approximately," or "generally" indicates that the numbers are allowed to vary by ±20%. Accordingly, in some embodiments, the numerical parameters used in the specification and claims are approximate values, which may be changed depending on the characteristics required by individual embodiments. In some embodiments, numerical parameters should take into account specified significant digits and employ a general method of digit reservation. Although the numerical ranges and parameters used to confirm their breadth of range in some embodiments of this specification are approximate values, in specific embodiments, such values ​​are set as precisely as feasible.

[0147] For each patent, patent application, patent application publication, and other material, such as articles, books, specifications, publications, and documents, referenced in this specification, the entire contents of which are incorporated herein by reference. This excludes historical application documents that are inconsistent with or conflict with the content of this specification, as well as documents that limit the broadest scope of the claims in this specification (currently or subsequently appended to this specification). It should be noted that in the event of any inconsistency or conflict between the descriptions, definitions, and / or terminology used in the supplementary materials to this specification and the content of this specification, the descriptions, definitions, and / or terminology used in this specification shall prevail.

[0148] Finally, it should be understood that the embodiments described in this specification are merely illustrative of the principles of the embodiments described herein. Other variations may also fall within the scope of this specification. Therefore, alternative configurations of the embodiments described herein are intended to be illustrative rather than limiting, and should be considered consistent with the teachings of this specification. Accordingly, the embodiments described herein are not limited to those explicitly introduced and described herein.

Claims

1. A standardized monitoring method for chain stores, characterized in that, include: Obtain standardized business processes and personnel information for chain stores; Based on the standardized business processes and personnel information, a standardized image set is constructed; Obtain scene features of the monitored area; Based on the aforementioned scene characteristics, the monitoring parameters are determined; The monitoring parameters include capture frequency and capture resolution; This includes: identifying the current business sub-process and at least one main object in the monitored area based on the scene characteristics; determining the capture frequency based on the operation characteristics of the at least one main object; the operation characteristics are the operator's proficiency in performing the current business sub-process, the proficiency including average proficiency and main proficiency; the main proficiency refers to the operator's proficiency in the current business sub-process; the average proficiency refers to the operator's proficiency in all business sub-processes; the main object refers to the operator performing standardized business sub-process operations; determining the capture frequency based on the operation characteristics of the at least one main object includes: The capture frequency is determined based on the average proficiency of the at least one main object and a preset rule, wherein the preset rule is to determine the capture frequency according to the proficiency range in which the average proficiency of the at least one main object is located; Based on the monitoring parameters, the monitoring area is monitored in real time to obtain at least one monitoring image; Based on the standardized image set and the at least one surveillance image, irregular behavior is determined, and The at least one monitoring image is processed based on a decision model to determine the confidence level of the irregular behavior; the decision model is a machine learning model. Determine whether the confidence level meets a preset condition; the preset condition is that the confidence level of the non-standard behavior is greater than a first threshold and less than a second threshold; the first threshold and the second threshold are determined based on a second false positive rate; different current business sub-processes have different first thresholds and second thresholds; the second false positive rate refers to the probability of an error occurring when the non-standard behavior is determined by the judgment model; In response, the capture resolution is increased within a preset time period based on the increase in capture resolution, wherein the increase in capture resolution is related to the operational features of the at least one main object.

2. The method according to claim 1, characterized in that, The standardized business process includes at least one of the standardized business sub-processes, and the standardized image set includes at least one standardized image subset. The number of images in the standardized image subset is positively correlated with the importance of the standardized business sub-process and the number of main objects in the standardized business sub-process. The construction of a standardized image set based on the standardized business process and the personnel information includes: Based on the standardized business sub-process and the personnel information, a standardized image subset corresponding to the standardized business sub-process is constructed, wherein each standardized image subset includes an action start point determination set and an action result determination set; the number of images in the action start point determination set and the action result determination set are positively correlated with the first misjudgment rate and the second misjudgment rate; the first misjudgment rate refers to the probability of an error occurring when determining the current business sub-process through the determination model.

3. A standardized monitoring system for chain stores, characterized in that, include: Build modules, used for Obtain standardized business processes and personnel information for chain stores; Based on the standardized business process and the personnel information, a standardized image set is constructed; the standardized image set includes at least one standardized image subset. Monitoring module, used for Obtain scene features of the monitored area; Based on the scene characteristics, monitoring parameters are determined; the monitoring parameters include capture frequency and capture resolution. This includes: identifying the current business sub-process and at least one main object in the monitored area based on the scene characteristics; determining the capture frequency based on the operation characteristics of the at least one main object; the operation characteristics are the operator's proficiency in performing the current business sub-process, the proficiency including average proficiency and main proficiency; the main proficiency refers to the operator's proficiency in the current business sub-process; the average proficiency refers to the operator's proficiency in all business sub-processes; the main object refers to the operator performing standardized business sub-process operations; determining the capture frequency based on the operation characteristics of the at least one main object includes: determining the capture frequency based on the average proficiency of the at least one main object and a preset rule, wherein the preset rule is to determine the capture frequency according to the proficiency range in which the average proficiency of the at least one main object falls; Based on the monitoring parameters, the monitoring area is monitored in real time to obtain at least one monitoring image; The determination module is used to determine irregular behavior based on the standardized image set and the at least one surveillance image, and The at least one monitoring image is processed based on a decision model to determine the confidence level of the irregular behavior; the decision model is a machine learning model. Determine whether the confidence level meets a preset condition; the preset condition is that the confidence level of the non-standard behavior is greater than a first threshold and less than a second threshold; the first threshold and the second threshold are determined based on a second false positive rate; different current business sub-processes have different first thresholds and second thresholds; the second false positive rate refers to the probability of an error occurring when the non-standard behavior is determined by the judgment model; In response, the capture resolution is increased within a preset time period based on the increase in capture resolution, wherein the increase in capture resolution is related to the operational features of the at least one main object.

4. The system according to claim 3, characterized in that, The standardized business process includes at least one of the standardized business sub-processes, and the standardized image set includes at least one standardized image subset. The number of images in the standardized image subset is positively correlated with the importance of the standardized business sub-process and the number of main objects in the standardized business sub-process. The construction of a standardized image set based on the standardized business process and the personnel information includes: Based on the standardized business sub-process and the personnel information, a standardized image subset corresponding to the standardized business sub-process is constructed, wherein each standardized image subset includes an action start point determination set and an action result determination set; the number of images in the action start point determination set and the action result determination set are positively correlated with the first misjudgment rate and the second misjudgment rate; the first misjudgment rate refers to the probability of an error occurring when determining the current business sub-process through the determination model.

5. A standardized monitoring device for chain stores, characterized in that, The device includes at least one processor and at least one memory; The at least one memory is used to store computer instructions; The at least one processor is configured to execute at least a portion of the computer instructions to implement the method as described in any one of claims 1 to 2.

6. A computer-readable storage medium, characterized in that, The storage medium stores computer instructions that, when executed by a processor, implement the method as described in any one of claims 1 to 2.

Images