Cash operation monitoring method and related device

Abstract

The invention discloses a cash operation monitoring method and a related device, which can be applied to the field of artificial intelligence or finance. And the terminal equipment receives the monitoring image of the target cash operation in real time, performs local image enhancement of the target local area on the monitoring image, inputs the enhanced monitoring image into a pre-configured risk identification model to obtain a risk identification result output by the risk identification model, and sends the risk identification result to the terminal equipment according to the risk identification result. And judging whether the target cash operation accords with a safe operation rule, and if not, generating alarm information. According to the scheme, the monitoring image of the cash operation is subjected to risk identification through the risk identification model deployed in the terminal equipment, the risk behavior is warned in time, the real-time performance of risk identification of the cash operation is improved, the cash identifiability is improved in a targeted manner through local image enhancement of the target local area, and the user experience is improved. The complexity of local image enhancement is reduced, the recognition efficiency is improved, and the accuracy and robustness of the recognition result of the risk recognition model are improved.

Description

Technical Field

[0001] The present application relates to the field of computer vision technology, and in particular to a cash operation monitoring method and related devices. Background Art

[0002] The vault is a key security area in financial scenarios and is equipped with multiple cameras. Taking the bank vault as an example, the bank security system uses multiple monitors to perform round-the-clock monitoring of the vault environment. Cash operations are key regulatory targets of the bank system. Usually, the bank system's regulatory strategy for cash operations is to manually observe the monitoring screen in real time to find out whether there are any cash operation risks. After discovering possible risks, manually trigger the alarm device of a bank security system.

[0003] The method of issuing alarms through manual observation of monitoring images may result in missed or false detections. The cause of the risk can only be checked after a security incident occurs using the historical monitoring video stream stored in the vault security system. This makes it impossible to ensure the real-time and accuracy of the risk alarms of the bank security system. Summary of the invention

[0004] In view of the above problems, the present application provides a cash operation monitoring method and related devices to achieve the purpose of improving the real-time and accuracy of cash operation monitoring. The specific scheme is as follows:

[0005] The first aspect of the present application provides a method for monitoring cash operations, comprising:

[0006] receiving a monitoring image of a target cash operation in real time, performing local image enhancement of a target local area on the monitoring image to obtain an enhanced monitoring image, wherein the target local area is a specified area of ​​cash in the monitoring image when the target cash operation is performed;

[0007] Inputting the enhanced surveillance image into a pre-configured risk identification model to obtain a risk identification result output by the risk identification model, wherein the risk identification model includes a cash identification model and a personnel identification model, and the risk identification result includes a cash identification result output by the cash identification model and a personnel identification result output by the personnel identification model;

[0008] Obtaining a safety operation rule for the target cash operation, wherein the safety operation rule at least includes a prescribed number of personnel and prescribed types of personnel within a cash area;

[0009] According to the risk identification result, determining whether the target cash operation complies with the safety operation rules;

[0010] If the safety operation rules are not met, an alarm message is generated, and the alarm message includes an alarm image, the risk identification result and the safety operation rules.

[0011] In a possible implementation, performing local image enhancement of a target local area on a surveillance image to obtain an enhanced surveillance image includes:

[0012] extracting the target local area from the surveillance image based on a specified execution area of ​​the target cash operation and a shooting angle of the surveillance image;

[0013] The contrast enhancement and edge sharpening are performed on the local area of ​​the target to obtain the enhanced monitoring image.

[0014] In a possible implementation, the enhanced surveillance image is input into a preconfigured risk identification model to obtain a risk identification result output by the risk identification model, including:

[0015] After the enhanced monitoring image is input into the cash recognition model to obtain the cash recognition result, if the cash recognition result indicates that a cash image exists, the enhanced monitoring image is input into the personnel recognition model to obtain the personnel recognition result.

[0016] In a possible implementation, the cash operation monitoring method further includes:

[0017] Obtain feedback information of historical alarm information, wherein the feedback information of the historical alarm information includes whether the cash recognition result and the person recognition result are correct, as well as the corrected cash recognition result and the corrected person recognition result

[0018] Annotating the warning image based on the feedback information of the warning information to obtain a retraining data set;

[0019] The retraining data set includes a cash recognition retraining data set and a personnel recognition retraining data set, the cash recognition retraining data set includes at least a first type of warning image and a cash label, wherein the first type of warning image is a warning image of an incorrect cash recognition result, and the cash label of the first type of warning image is a corrected cash recognition result, and the personnel recognition retraining data set includes at least a second type of warning image and a personnel label, wherein the second type of warning image is a warning image of an incorrect personnel recognition result, and the personnel label of the second type of warning image is a corrected personnel recognition result;

[0020] When a preset retraining time is reached, the risk identification model is retrained based on the retraining data set to obtain retraining parameter information; or, the retraining data set and the retraining instruction are sent to the server, so that the server retrains the risk identification model based on the retraining data set and returns the retraining parameter information; the retraining parameter information includes a hyperparameter group and a model parameter;

[0021] The risk identification model is optimized based on the retraining parameter information.

[0022] In a possible implementation, retraining the risk identification model based on the retraining data set includes:

[0023] Obtaining pre-configured hyperparameter search information, wherein the hyperparameter search information includes a preset type and a corresponding preset value range;

[0024] If the number of training times does not reach a preset number threshold, the risk identification model is configured using the preset hyperparameter group, and the risk identification model is retrained based on the retraining data set, and retraining parameter information is obtained after a preset training completion condition is met;

[0025] The preset hyperparameter group includes preset values ​​of multiple preset types of hyperparameters obtained by randomly selecting values ​​in the hyperparameter search information;

[0026] If the number of training times reaches the threshold, a target hyperparameter group is obtained according to the hyperparameter search results, the risk identification model is configured using the target hyperparameter group, and the risk identification model is trained based on the retraining data set, and after the training completion condition is met, retraining parameter information is obtained; wherein the target hyperparameter group includes target values ​​of hyperparameters of multiple target types.

[0027] In a possible implementation, the cash operation monitoring method further includes, before inputting the enhanced monitoring image into a preconfigured risk identification model:

[0028] The receiving server trains the machine learning model based on the training data set to obtain initial configuration information of the cash recognition model and the personnel recognition model, and deploys the cash recognition model and the personnel recognition model based on the initial configuration information, wherein the initial configuration information includes an initial hyperparameter group and initial model parameters.

[0029] A second aspect of the present application provides a cash operation monitoring device, comprising:

[0030] A local enhancement unit is used to receive a monitoring image of a target cash operation in real time, and perform local image enhancement of a target local area on the monitoring image to obtain an enhanced monitoring image, wherein the target local area is a specified area of ​​cash in the monitoring image when the target cash operation is performed;

[0031] a risk identification unit, configured to input the enhanced surveillance image into a pre-configured risk identification model to obtain a risk identification result output by the risk identification model, wherein the risk identification model includes a cash identification model and a personnel identification model, and the risk identification result includes a cash identification result output by the cash identification model and a personnel identification result output by the personnel identification model;

[0032] An operation rule determination unit is used to obtain a safety operation rule for the target cash operation, wherein the safety operation rule at least includes: The required number of personnel and the required types of personnel within the cash area;

[0033] A risk determination unit, configured to determine whether the target cash operation complies with safety operation rules according to the risk identification result;

[0034] The risk warning unit is used to generate warning information if the safety operation rules are not met. The warning information includes a warning image, the risk identification result and the safety operation rules.

[0035] A third aspect of the present application provides a computer program product, comprising computer-readable instructions, which, when executed on an electronic device, enables the electronic device to implement the method for monitoring cash operations of the first aspect or any implementation of the first aspect.

[0036] A fourth aspect of the present application provides an electronic device, comprising at least one processor and a memory connected to the processor, wherein:

[0037] The memory is used to store computer programs;

[0038] The processor is used to execute the computer program so that the electronic device can implement the cash operation monitoring method of the above-mentioned first aspect or any implementation manner of the first aspect.

[0039] A fifth aspect of the present application provides a computer storage medium, which carries one or more computer programs. When the one or more computer programs are executed by an electronic device, the electronic device can perform the method for monitoring cash operations according to the first aspect or any implementation of the first aspect.

[0040] By means of the above technical scheme, the present application provides a cash operation monitoring method and related devices, wherein the terminal device receives the monitoring image of the target cash operation in real time, performs local image enhancement of the target local area on the monitoring image, and obtains an enhanced monitoring image, wherein the target local area is the specified area of ​​cash in the monitoring image when the target cash operation is performed. The enhanced monitoring image is input into a pre-configured risk identification model to obtain a risk identification result output by the risk identification model, wherein the risk identification model includes a cash identification model and a personnel identification model, and the risk identification result includes a cash identification result output by the cash identification model and a personnel identification result output by the personnel identification model. The safety operation rules for the target cash operation are obtained, wherein the safety operation rules at least include the specified number of personnel and the specified type of personnel in the cash area. According to the risk identification results, it is determined whether the target cash operation complies with the safety operation rules. If the safety operation rules are not met, an alarm message is generated, and the alarm message includes an alarm image, a risk identification result, and a safety operation rule. This solution uses the risk identification model deployed on the terminal device to perform risk identification on the monitoring images of cash operations in real time, and promptly warns of risk behaviors that do not comply with safe operation rules, thereby improving the real-time nature of risk identification of cash operations. Specifically, through local enhancement of the target local area, the cash recognizability is improved in a targeted manner, the complexity of local image enhancement is reduced, the recognition efficiency is improved, and the accuracy and robustness of the recognition results of the risk identification model are improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] The above and other features, advantages and aspects of the embodiments of the present disclosure will become more apparent with reference to the following detailed description in conjunction with the accompanying drawings. Throughout the accompanying drawings, the same or similar reference numerals represent the same or similar elements. It should be understood that the drawings are schematic and the originals and elements are not necessarily drawn to scale.

[0042] Figure 1 An architectural diagram of a distributed cash operation monitoring system provided for this application;

[0043] Figure 2 A schematic diagram of the hardware structure of a terminal provided in this application;

[0044] Figure 3 A schematic diagram of the hardware structure of a server provided for this application;

[0045] Figure 4 A schematic diagram of the specific structure of a cash operation monitoring system provided in this application;

[0046] Figure 5 A specific implementation flow chart of a cash operation monitoring method provided in an embodiment of the present application;

[0047] Figure 6A flowchart of a cash operation monitoring method provided in an embodiment of the present application;

[0048] Figure 7 A schematic diagram of the structure of a cash operation monitoring device provided in an embodiment of the present application;

[0049] Figure 8 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0050] The following describes the embodiments of the present application in conjunction with the drawings in the embodiments of the present application. The terms used in the implementation method section of the present application are only used to explain the specific embodiments of the present application, and are not intended to limit the present application.

[0051] The embodiments of the present application are described below in conjunction with the accompanying drawings. Those skilled in the art will appreciate that, with the development of technology and the emergence of new scenarios, the technical solutions provided in the embodiments of the present application are also applicable to similar technical problems.

[0052] The terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and need not be used to describe a specific order or sequential order. It should be understood that the terms used in this way can be interchangeable under appropriate circumstances, which is only to describe the distinction mode adopted by the objects of the same attributes when describing in the embodiments of the present application. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, so that the process, method, system, product or equipment comprising a series of units need not be limited to those units, but may include other units that are not clearly listed or inherent to these processes, methods, products or equipment.

[0053] The present application can be applied in the technical field of risk monitoring of bank vaults, and specifically can be applied in the monitoring scenario of cash operations in the vault.

[0054] The importance of cash management in the vault cannot be ignored. It is directly related to the asset security, operational efficiency and risk prevention of banks or financial institutions, and the cash operation of staff is the key supervision object of cash management. Cash in the vault needs to be strictly managed to ensure the safety of bank assets. At present, through a strict cash management system, it can effectively prevent cash from being stolen, damaged or lost, and ensure the safety and integrity of assets. The cash management system includes but is not limited to multiple security protection measures that require strict implementation, such as regular inventory and limiting the number and authority of personnel contact, as well as formulating cash operation specifications for various cash operations. For example, the management system for limiting the number and authority of personnel contact specifically includes: for the actions involving cash operations in the vault, it is required that multiple people are present and the types of personnel are specified. It is understandable that the presence of multiple people of specified personnel types can effectively ensure the fairness and accuracy of cash operations. In addition, the presence of multiple people can also improve the transparency and traceability of cash operations. With the participation of multiple people, every link in the cash operation process can be clearly recorded and effectively monitored, which helps to discover and solve problems in a timely manner and ensure that cash operations comply with safety operation rules and regulations.

[0055] The existing monitoring methods mainly rely on manual monitoring of the cash operation area in the vault, which is mainly done by manual inspection through monitoring large screens or by replaying the monitoring video afterwards. The naked eye subjectively judges whether the type and number of personnel in the cash operation area during the cash operation comply with the cash management system, and whether there are any behaviors that do not comply with the cash operation specifications, and artificially triggers the bank security system's warning device for vault risks to achieve the purpose of risk warning. Limited by the subjectivity and ability differences of manual observation, the existing bank security system has the technical disadvantages of low real-time, low accuracy, and high lag in the warning of financial control risk monitoring.

[0056] In order to solve the above technical shortcomings, this solution proposes to collect monitoring images in real time based on machine vision technology and use the pre-trained risk monitoring model to identify whether the video image has risks. If so, it will automatically trigger the bank security system's warning device for vault risks. Among them, the risk monitoring model is built and optimized based on the deep learning model, thereby improving the real-time monitoring of vault cash management and cash operations. Furthermore, this solution can solve the problems of large manual workload and inspection lag, and can cover all vault environments and all time periods, reducing omissions in manual inspections and subjective errors.

[0057] The research found that the existing deep learning models have some technical shortcomings in the monitoring scenario of cash operations in vault cash management, as follows:

[0058] 1. Data dependency and quality issues: The effectiveness of deep learning models depends largely on the quantity and quality of input data. If the data of cash operations in the vault is insufficient, the labels are inaccurate, or there is noise, the training effect of the model will be affected, which may lead to false positives or false negatives. At the same time, due to the nature of the vault, it is impossible to provide a large amount of business operation data, which will result in poor training results.

[0059] 2. The accuracy of the model. After the initial deployment, the accuracy was low due to the incompleteness of the training data and the complexity of the vault scenario.

[0060] 3. The bank security system is deployed in the form of a central server, and the deep learning model is deployed on the server in the bank security system (such as a cloud server). The image sensitive information involved in various operations in the vault cash management involves security and privacy protection issues during the transmission process to the server, which may lead to the leakage of image sensitive information or malicious use.

[0061] In this regard, this solution further optimizes the cash operation monitoring solution, and proposes a distributed cash operation monitoring system through the combination of machine vision technology and edge computing technology. In the cash operation monitoring system, the terminal equipment configured on the edge side deploys the machine learning model using the trained model parameters, collects monitoring images in real time through the camera, and uses the pre-deployed machine learning model to identify the risks of cash operations on the monitoring images. Specifically, by performing real-time analysis of the types and numbers of people performing cash operations in the monitoring images, it is determined whether there are any violations. The distributed deployment of edge-side terminals reduces the computing pressure of the server and avoids the leakage or malicious use of sensitive image information, thereby improving data security.

[0062] Furthermore, the cash operation monitoring system trains and optimizes the parameters of the machine learning model through the server, and sends the trained parameters to the terminal device deployed on the edge side, so that the terminal device deploys and optimizes the machine learning model. This solution can improve the recognition accuracy of the machine learning model by training and optimizing the parameters of the machine learning model through the server.

[0063] The present application embodiment provides a distributed cash operation monitoring system, see Figure 1 , Figure 1 A schematic diagram of a system architecture is shown. The system may include a terminal 100 and a server 200. The server 200 may include one or more servers ( Figure 1 In the example, a server is included, and the server 200 can provide the method provided in the embodiment of the present application for one or more terminals.

[0064] Among them, a cash operation monitoring application can be installed on the terminal 100. The above application and web page can provide an interface. The terminal 100 can receive relevant parameters entered by the user on the cash operation monitoring interface and send the above parameters to the server 200. The server 200 can obtain processing results based on the received parameters and return the processing results to the terminal 100.

[0065] It should be understood that in some optional implementations, the terminal 100 can also complete the action of obtaining the processing result based on the received parameters by itself without the cooperation of the server, and the embodiments of the present application are not limited to this.

[0066] Next describe Figure 1 The product form of the mid-terminal 100;

[0067] The terminal 100 in the embodiment of the present application can be a mobile phone, a tablet computer, a wearable device, a vehicle-mounted device, an augmented reality (AR) / virtual reality (VR) device, a laptop computer, an ultra-mobile personal computer (UMPC), a netbook, a personal digital assistant (PDA), etc., and the embodiment of the present application does not impose any limitation on this.

[0068] Figure 2 An optional hardware structure diagram of the terminal 100 is shown.

[0069] refer to Figure 2 As shown, the terminal 100 may include a radio frequency unit 110, a memory 120, an input unit 130, a display unit 140, a camera 150 (optional), an audio circuit 160 (optional), a speaker 161 (optional), a microphone 162 (optional), an earphone jack 163 (optional), a processor 170, an external interface 180, a power supply 190 and other components. Those skilled in the art will appreciate that Figure 2 These are merely examples of terminals or multi-function devices and do not constitute limitations on the terminals or multi-function devices, which may include more or fewer components than those shown in the figures, or combinations of certain components, or different components.

[0070] The input unit 130 can be used to receive input digital or character information, and generate key signal input related to the user settings and function control of the portable multifunctional device. Specifically, the input unit 130 may include a touch screen 131 (optional) and / or other input devices 132. The touch screen 131 can collect the user's touch operations on or near it (such as the user's operation on or near the touch screen using any suitable object such as fingers, joints, stylus, etc.), and drive the corresponding connection device according to a pre-set program. The touch screen can detect the user's touch action on the touch screen, convert the touch action into a touch signal and send it to the processor 170, and can receive and execute the command sent by the processor 170; the touch signal at least includes the touch point coordinate information. The touch screen 131 can provide an input interface and an output interface between the terminal 100 and the user. In addition, the touch screen can be implemented using multiple types such as resistive, capacitive, infrared and surface acoustic wave. In addition to the touch screen 131, the input unit 130 can also include other input devices. Specifically, other input devices 132 may include, but are not limited to, one or more of a physical keyboard, function keys (such as a volume control key, a switch key, etc.), a trackball, a mouse, a joystick, and the like.

[0071] Among them, the input device 132 can receive input data and the like.

[0072] The display unit 140 may be used to display information input by the user or provided to the user, various menus of the terminal 100, interactive interfaces, file displays, and / or playback of any multimedia files. In the embodiment of the present application, the display unit 140 may be used to display an interface for monitoring cash operations, processing results, and the like.

[0073] The memory 120 can be used to store instructions and data. The memory 120 can mainly include an instruction storage area and a data storage area. The data storage area can store various data, such as multimedia files, texts, etc.; the instruction storage area can store software units such as operating systems, applications, instructions required for at least one function, or their subsets and extensions. It can also include a non-volatile random access memory; provide the processor 170 with hardware, software and data resources including management of computing and processing equipment, and support control software and applications. It is also used for the storage of multimedia files, and the storage of running programs and applications.

[0074] The processor 170 is the control center of the terminal 100. It uses various interfaces and lines to connect various parts of the entire terminal 100. By running or executing instructions stored in the memory 120 and calling data stored in the memory 120, it executes various functions of the terminal 100 and processes data, thereby controlling the terminal device as a whole. Optionally, the processor 170 may include one or more processing units; preferably, the processor 170 may integrate an application processor and a modem processor, wherein the application processor mainly processes the operating system, user interface and application program, and the modem processor mainly processes wireless communication. It is understandable that the above-mentioned modem processor may not be integrated into the processor 170. In some embodiments, the processor and the memory may be implemented on a single chip, and in some embodiments, they may also be implemented separately on separate chips. The processor 170 may also be used to generate corresponding operation control signals, send them to corresponding components of the computing and processing device, read and process data in the software, especially read and process data and programs in the memory 120, so that each functional module therein performs corresponding functions, thereby controlling the corresponding components to act according to the requirements of the instructions.

[0075] Among them, the memory 120 can be used to store software codes related to the cash operation monitoring method, the processor 170 can execute the steps of the cash operation monitoring method, and can also schedule other units (such as the above-mentioned input unit 130 and display unit 140) to achieve corresponding functions.

[0076] The radio frequency unit 110 (optional) can be used for receiving and sending information or receiving and sending signals during a call, for example, after receiving the downlink information of the base station, it is sent to the processor 170 for processing; in addition, the designed uplink data is sent to the base station. Generally, the RF circuit includes but is not limited to an antenna, at least one amplifier, a transceiver, a coupler, a low noise amplifier (Low Noise Amplifier, LNA), a duplexer, etc. In addition, the radio frequency unit 110 can also communicate with network devices and other devices through wireless communication. The wireless communication can use any communication standard or protocol, including but not limited to Global System of Mobile communication (Global System of Mobile communication, GSM), General Packet Radio Service (General Packet Radio Service, GPRS), Code Division Multiple Access (Code Division Multiple Access, CDMA), Wideband Code Division Multiple Access (Wideband Code Division Multiple Access, WCDMA), Long Term Evolution (Long Term Evolution, LTE), email, Short Messaging Service (SMS), etc.

[0077] In this embodiment of the present application, the RF unit 110 can send data to the server 200 and receive processing results sent by the server 200.

[0078] It should be understood that the radio frequency unit 110 is optional and can be replaced by other communication interfaces, such as a network port.

[0079] The terminal 100 also includes a power supply 190 (such as a battery) for supplying power to various components. Preferably, the power supply can be logically connected to the processor 170 through a power management system, so that the power management system can manage functions such as charging, discharging, and power consumption.

[0080] The terminal 100 further includes an external interface 180 , which may be a standard Micro USB interface or a multi-pin connector, and may be used to connect the terminal 100 to communicate with other devices, or to connect a charger to charge the terminal 100 .

[0081] Although not shown, the terminal 100 may also include a flashlight, a wireless fidelity (WiFi) module, a Bluetooth module, sensors with different functions, etc., which are not described in detail here. Some or all of the methods described below may be applied in the following embodiments. Figure 2 In the terminal 100 shown.

[0082] Next describe Figure 1 The product form of the server 200;

[0083] Figure 3 A structural diagram of a server 200 is provided, such as Figure 3 As shown, the server 200 includes a bus 201, a processor 202, a communication interface 203, and a memory 204. The processor 202, the memory 204, and the communication interface 203 communicate with each other via the bus 201.

[0084] The bus 201 may be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus. The bus may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 3 Only one thick line is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0085] The processor 202 may be any one or more of a central processing unit (CPU), a graphics processing unit (GPU), a microprocessor (MP), or a digital signal processor (DSP).

[0086] The memory 204 may include a volatile memory, such as a random access memory (RAM). The memory 204 may also include a non-volatile memory, such as a read-only memory (ROM), a flash memory, a hard drive (HDD), or a solid state drive (SSD).

[0087] The memory 204 may be used to store software codes related to the cash operation monitoring method, and the processor 202 may execute the steps of the cash operation monitoring method of the chip, and may also schedule other units to implement corresponding functions.

[0088] It should be understood that the above-mentioned terminal 100 and server 200 can be centralized or distributed devices, and the processors in the above-mentioned terminal 100 and server 200 (such as processor 170 and processor 202) can be hardware circuits (such as application specific integrated circuit (ASIC), field-programmable gate array (FPGA), general-purpose processor, digital signal processor (DSP), microprocessor or microcontroller, etc.), or a combination of these hardware circuits. For example, the processor can be a hardware system with an instruction execution function, such as a CPU, DSP, etc., or a hardware system without an instruction execution function, such as an ASIC, FPGA, etc., or a combination of the above-mentioned hardware systems without an instruction execution function and hardware systems with an instruction execution function.

[0089] Figure 4 A specific structural diagram of a distributed cash operation monitoring system provided in an embodiment of the present application, the monitoring system includes a camera, a model detection module, an algorithm reasoning module, an edge management module and a model optimization module. Among them. The model detection module, the algorithm reasoning module, and the edge management module are deployed on the terminal device on the edge side, the terminal device is connected to the camera pre-deployed in the cash operation area in the vault, and the model optimization module is deployed on the server.

[0090] The camera is used to capture the pre-configured cash operation area in real time, and transmit the resulting monitoring stream to the model detection module in the edge computing box.

[0091] The model detection module is used to call the risk identification model deployed in the edge computing box to detect the cash operation image in the vault in real time, obtain the risk identification result, and send the risk identification result to the algorithm reasoning module. Among them, the risk identification model includes a cash identification model and a personnel identification model, and the risk identification result includes the cash identification result output by the cash identification model and the personnel identification result output by the personnel identification model.

[0092] The algorithm reasoning module is used to receive risk identification results and analyze whether to send an alarm to the edge management module based on the risk identification results.

[0093] The edge management module is used to configure the camera's data stream and detection area during deployment, and is also used to display alarm information and alarm images. According to the image marking rules, it marks whether the alarm image is a valid image and whether the image detection result is correct and corrected. The edge management module is also used to send the marked image information to the model optimization module after the upload time is reached, where the marked image information includes the alarm image and the mark, and the mark includes whether the image detection result is correct.

[0094] The model optimization module is used for model training and optimization. The model optimization process includes receiving the labeled image information from the edge management module, performing model hyperparameter search training, and selecting the best performing model hyperparameters to send to the model detection module for updating the model hyperparameters.

[0095] The embodiment of the present application provides a distributed cash operation monitoring system, which monitors cash operations in real time through various functional modules deployed in the terminal devices on the edge side, and identifies in real time whether there are risks in the cash operation images, that is, behaviors that do not comply with cash management regulations, and issues timely warnings for risky behaviors. The functional modules deployed on the server side are used to search for model hyperparameters and train and optimize model parameters, and the model hyperparameter group and model parameters are sent to each terminal device. Thus, the risk identification model is used to improve the real-time identification of cash operation risks in the vault, the high computing power of the server is used to train and optimize the model parameters, and the recognition accuracy of the risk identification model is improved. The edge computing capabilities of each terminal device are used to improve data security.

[0096] Furthermore, the model detection module improves the recognizability of smaller objects such as banknotes through data enhancement methods such as local contrast enhancement and image edge sharpening, thereby improving the accuracy and robustness of the recognition results of the risk identification model.

[0097] Furthermore, the edge management module obtains sample data for optimizing the risk identification model by marking whether the warning image is a valid image and whether the image detection result is correct. The model optimization module automatically optimizes the risk identification model based on the marked image information, completes continuous iterative updating of the model, and improves the accuracy of the risk identification results of the risk identification model.

[0098] Specifically, in the monitoring scenario of cash operations, the model detection module of the edge computing box decodes, preprocesses, and performs model detection on the received real-time video stream, detects the cash operation scene in the vault in real time, obtains the information of banknotes and items and the category of personnel, and finally sends the reasoning results to the algorithm reasoning module. The algorithm reasoning module receives the results of the model detection, and analyzes and determines whether the designated cash operator is less than the specified number when cash appears in the cash operation area based on these results. If there are less than the specified number of personnel, an alarm message is sent to the edge management module. The edge management module can display the alarm information and corresponding images in real time, and send these alarm messages to relevant personnel through channels such as the network for reminder. At the same time, the staff can also mark whether these alarm images are valid images, and whether the image detection results are correct and corrected. The module regularly sends the above-mentioned marked images to the model optimization module. The actual alarm data collected regularly constitutes a data set and is imported into the model optimization module for model hyperparameter search and training. The best performing model is regularly selected and sent to the model detection module for model replacement, and the detection model is continuously optimized autonomously.

[0099] Figure 5 A specific implementation flow chart of a cash operation monitoring method provided in an embodiment of the present application is as follows: Figure 5 As shown, this method specifically includes:

[0100] S501. Collect original data.

[0101] In this embodiment, the original data includes vault scene data and public resource data. The vault scene data includes multiple vault cash operation area background images, multiple cash images in vault cash operation scenes, and multiple vault staff images. The public resource data includes personnel data, cash data, and negative sample data.

[0102] S502: Process the original data to obtain a training data set.

[0103] In this embodiment, the process of processing the original data includes:

[0104] 1. Preprocess the data, including pre-processing operations such as deduplication, standardization and screening.

[0105] 2. Perform data labeling on the preprocessed data. In particular, each image in the data is labeled with personnel type and personnel information, wherein the personnel information includes the number of personnel and personnel type, and each image in the data is labeled with cash, wherein the cash labeling includes whether there is cash in the image.

[0106] 3. Perform data enhancement operations on the labeled data to obtain a training data set, which includes a cash recognition training data set and a personnel recognition training data set. The cash recognition training data set includes sample images and cash annotations, and the personnel recognition training data set includes sample images and personnel information annotations. Data enhancement includes preset enhancement operations such as image stitching, scaling, color conversion, and cropping.

[0107] S503: Perform image enhancement on the images in the training data set.

[0108] In this embodiment, the image enhancement method includes contrast enhancement and image edge sharpening.

[0109] Specifically, the contrast enhancement method includes:

[0110] The surveillance image is adaptively divided into multiple local areas, and the contrast of the internal pixels of each local area is calculated. The enhancement parameters are determined according to the contrast of the local area, and the local area with lower contrast is enhanced more strongly. The contrast of each local area is enhanced according to the enhancement parameters and merged into a complete image.

[0111] Image edge sharpening methods include:

[0112] The Laplace operator is applied to identify the edge map of the surveillance image and the parameters are continuously adjusted to achieve the best effect. Finally, the edge map of the image is directly superimposed on the surveillance image.

[0113] S504: Through the pre-built model automation training platform, the machine learning model is trained based on the training data set to obtain the cash recognition model and the personnel recognition model respectively, and the cash recognition model and the personnel recognition model are deployed to the edge computing box.

[0114] In this embodiment, the cash recognition model is used to output cash recognition results, which are used to indicate whether there is cash in the image. The person recognition model outputs person recognition results, which include the type and number of people in the image.

[0115] In this embodiment, the cash recognition model is trained based on the cash recognition training data set until the training completion condition is met, and the various parameters of the cash recognition model are obtained, and the cash recognition model is deployed on the edge computing box based on the hyperparameter group and parameters. The personnel recognition model is trained based on the personnel recognition training data set until the training completion condition is met, and the various hyperparameters of the personnel recognition model are obtained, and the personnel recognition model is deployed on the edge computing box based on the various hyperparameters and parameters. Among them, the values ​​of various hyperparameters in the hyperparameter groups of the cash recognition model and the personnel recognition model are all initial hyperparameters randomly obtained from the pre-configured hyperparameter types and values.

[0116] In this embodiment, the edge computing box is configured on the terminal device at the edge side.

[0117] It should be noted that S501 to 504 are executed by a server, which can connect to multiple bank systems, collect as much raw data as possible, rely on powerful computing power to improve the quality of the raw data, obtain large-volume and high-quality training data, and improve the accuracy of the risk identification model.

[0118] S505: Receive the monitoring image of the target cash operation in real time, and perform local image enhancement of the target local area on the monitoring image.

[0119] In this embodiment, the monitoring images are collected and transmitted in real time by a camera arranged in the cash operation area.

[0120] In this embodiment, the target local area is a specified area of ​​cash in the monitoring image when performing the target cash operation. Optionally, based on the specified execution area of ​​the target cash operation and the shooting angle of the monitoring image, the target local area is predicted and extracted from the monitoring image.

[0121] For example, when the target cash operation is a cash distribution operation, the target local area is the cash distribution box area, and the cash distribution box area is the area of ​​the cash distribution box in the surveillance image. It should be noted that the spatial position relationship between the fixed cash distribution box and the camera is fixed, so the cash distribution box area can be determined by information such as the camera's shooting angle, and the recognition of the mobile cash distribution box can be achieved based on image recognition. Since the cash distribution box is relatively large in size and has simple features, the local image area to be enhanced is determined based on the easily determined cash distribution box area, without the need to perform local image enhancement of each available area on the surveillance image.

[0122] In this embodiment, local image enhancement includes contrast enhancement and edge sharpening.

[0123] It should be noted that due to the long distance between the camera and the cash operation area and the fact that cash has many subtle features as the target detection object, the difficulty of cash recognition is increased. For this reason, local image enhancement of the target local area is performed on the surveillance image before inputting the cash detection model.

[0124] The contrast of the banknote part is selectively enhanced through the local contrast enhancement strategy, thereby reducing the interference caused by background information to the recognition. The edge part of the image is enhanced through image edge sharpening, so that the edge of the banknote is clearer and more prominent, greatly improving the recognizability of the banknote. Therefore, the subtle features of the possible area are enhanced through local contrast enhancement and image edge sharpening, and the details of the cash are highlighted to improve the accuracy of cash detection. By localizing the target, the complexity of local image enhancement is reduced and the image processing speed is improved.

[0125] S506. Input the enhanced surveillance image into the cash recognition model and the personnel recognition model in the edge computing box to obtain the cash recognition result output by the cash recognition model and the personnel recognition result output by the personnel recognition model.

[0126] In this embodiment, after the enhanced surveillance image is input into the cash recognition model in the edge computing box to obtain the cash recognition result, if the cash recognition result indicates the presence of a cash image, the enhanced surveillance image is input into the personnel recognition model in the edge computing box for personnel recognition.

[0127] This step configures the recognition order of the cash recognition model and the personnel recognition model, and starts the personnel recognition model after the cash recognition result indicates that the monitored image area is a cash area, thereby improving the model detection efficiency and reducing the consumption of model computing resources.

[0128] S507. Determine whether the cash operation complies with the pre-configured cash operation specification based on the cash identification result and the personnel identification result.

[0129] In this embodiment, the cash operation specification includes safety operation rules corresponding to each cash operation, and the safety operation rules at least include a specified number of personnel and specified types of personnel in the cash area.

[0130] In this embodiment, the corresponding security operation rules are searched from the cash operation specification according to the target cash operation as the target security operation rules, and whether the cash operation complies with the target security operation rules is determined according to the cash recognition result and the personnel recognition result. The target cash operation is the cash operation currently being executed, and the cash operation can be determined by other monitoring systems or business systems.

[0131] S508. If not, a cash operation warning message is issued.

[0132] In this embodiment, the alarm information of cash operation includes an alarm image, a cash recognition result, a personnel recognition result, and a target safety operation rule, wherein the target safety operation rule is a safety operation rule that the cash recognition result and the personnel recognition result do not meet. For example, in a cash dispensing scenario, if the model detection result indicates that the number of personnel in the cash dispensing area is less than 2, the monitoring image is used as an alarm image to generate alarm information of cash dispensing.

[0133] In this embodiment, the specific method of issuing alarm information for cash operations includes: sending the alarm information to a pre-configured security supervision client, the security supervision client includes a text message client, an email client, and an application client, etc., so that the supervisor can promptly check the cash operation risks through the alarm information received and displayed by the security supervision client.

[0134] S509: Obtain feedback information of historical alarm information.

[0135] In this embodiment, feedback information of the historical alarm information is obtained according to a preset period, and the feedback information of the historical alarm information includes whether the cash recognition result and the person recognition result are correct, as well as the corrected cash recognition result and the corrected person recognition result.

[0136] The alarm information generated in actual applications is collected regularly, and the staff can mark whether these alarm images are valid images and whether the image detection results need to be corrected. Then the above marked images are regularly sent to the model optimization part for model optimization iteration to continuously enhance the accuracy and robustness of the algorithm.

[0137] S510 , labeling each warning image based on feedback information of historical warning information to obtain a retraining data set.

[0138] In this embodiment, the retraining data set includes a cash recognition retraining data set and a personnel recognition retraining data set. The cash recognition retraining data set includes at least a first type of warning image and a cash label, wherein the first type of warning image is a warning image of an incorrect cash recognition result, and the cash label of the first type of warning image is a corrected cash recognition result. The personnel recognition retraining data set includes at least a second type of warning image and a personnel label, wherein the second type of warning image is a warning image of an incorrect personnel recognition result, and the personnel label of the second type of warning image is a corrected personnel recognition result.

[0139] It should be noted that S505~510 are executed by the terminal device on the edge side. The terminal device automatically identifies the personnel information and cash area of ​​the monitoring image through the risk identification model deployed on the edge, avoids the risk of leakage of sensitive image information, improves the real-time performance of risk monitoring, and reduces resource consumption on the server side.

[0140] S511. When the time for retraining is reached, the risk identification model is retrained based on the retraining data set through the pre-built model automation training platform to achieve optimization of the risk identification model.

[0141] In this embodiment, the retraining opportunity includes when the performance evaluation result of at least one risk detection model does not meet the model performance condition, or when a preset retraining interval is reached.

[0142] In this embodiment, the cash recognition model is retrained based on the cash recognition retraining data set until the retraining completion conditions are met, the optimized parameters of the cash recognition model are obtained, and the cash recognition model is updated in the edge computing box based on the optimized parameters. The cash recognition model is retrained based on the personnel recognition retraining data set until the retraining completion conditions are met, the optimized parameters of the personnel recognition model are obtained, and the personnel recognition model is updated in the edge computing box based on the optimized parameters.

[0143] In this embodiment, the automated training platform has a hyperparameter search training function and a data receiving autonomous training function. The specific method of retraining the machine learning model based on the retraining data set through the pre-built model automated training platform includes:

[0144] Get pre-configured hyperparameter search information, which includes preset types and corresponding preset value ranges.

[0145] After receiving the training start instruction, if the number of training times does not reach the preset threshold, the risk identification model is configured using the preset hyperparameter group, and the risk identification model is trained based on the retraining data set. After the training completion conditions are met, the trained risk identification model and its parameters are obtained.

[0146] The preset hyperparameter group includes preset values ​​of multiple preset types of hyperparameters, and the preset hyperparameter group is obtained by randomly selecting values ​​in the hyperparameter search information.

[0147] If the number of training times reaches the threshold, the target hyperparameter group is obtained according to the hyperparameter search results, the risk identification model is configured using the target hyperparameter group, and the risk identification model is trained based on the retraining data set. After the training completion conditions are met, the trained risk identification model and its parameters are obtained.

[0148] The target hyperparameter group includes target values ​​of hyperparameters of multiple target types.

[0149] It should be noted that hyperparameters are parameters that need to be manually set before model training, and they have a crucial impact on the final performance of the model. These parameters include but are not limited to learning rate, regularization strength, network hierarchy, training batch size, number of training times, etc. By adjusting these hyperparameters through hyperparameter search, the accuracy and generalization ability of the model can be effectively improved.

[0150] The goal of hyperparameter search is to find the optimal hyperparameter group to minimize the model error or maximize the model performance. Hyperparameter search analyzes the influence of different types of hyperparameters on the model results based on the model results, and analyzes whether the value of the hyperparameter has a positive or negative correlation with the model. Targeted selection is made within the range of hyperparameters specified during configuration, and then training is performed to ultimately obtain the best model.

[0151] It should be noted that S511 can be implemented by the server. When the retraining time is reached, the terminal device sends a retraining instruction and a retraining data set to the server. The server retrains the risk identification model configured in the terminal device based on at least the retraining data set of the terminal device, and sends down the retrained parameters and the hyperparameter group to be updated. The terminal device updates the parameters and hyperparameters to optimize the risk identification model.

[0152] It can be seen from the above technical solutions that the cash operation monitoring method provided by the embodiment of the present application can identify risks of cash operation monitoring images in real time through the risk identification model deployed on the terminal device, and timely warn of risky behaviors, thereby improving the real-time and accuracy of risk identification of cash operations. The model hyperparameters are searched and the model parameters are trained and optimized through the server, and the model hyperparameter group and model parameters are sent to each terminal device. The high computing power of the server is used to train and optimize the model parameters, thereby improving the recognition accuracy of the risk identification model, and utilizing the edge computing capabilities of each terminal device to improve data security.

[0153] Furthermore, through image enhancement methods such as local contrast enhancement in the target local area and image edge sharpening, the cash recognizability can be improved in a targeted manner, the complexity of local image enhancement can be reduced, the recognition speed can be increased, and the accuracy and robustness of the recognition results of the risk identification model can be improved.

[0154] Furthermore, by marking whether the warning image is a valid image and whether the image detection result is correct by the terminal device, a retraining data set for optimizing the local risk identification model is obtained. Therefore, the server can retrain the risk identification model corresponding to the terminal device based on the retraining data set, thereby improving the matching degree between the risk identification model and the vault environment corresponding to the terminal device, and improving the accuracy of the risk identification model.

[0155] This solution can significantly reduce the risk and manpower cost of manual supervision in this scenario by identifying the risks of surveillance images in real time, and can also provide real-time warnings to reduce time loss and solve the lag problem. The model automation training platform is used to regularly train the real data collected from actual applications, screen out the best performing models, and perform self-updating iterations of the models, further improving the accuracy of risk identification.

[0156] It should be noted that the above is only an optional specific implementation process of a cash operation monitoring method provided in this embodiment. The present application can also be implemented through other specific implementation methods. For example, the hyperparameter search method includes but is not limited to grid search, random search, Bayesian optimization search and other optional search methods. For another example, another implementation method of S505 is: receiving the monitoring image in real time, inputting the monitoring image into the cash recognition model and the personnel recognition model in the edge computing box at the same time, and obtaining the cash recognition result output by the cash recognition model and the personnel recognition result output by the personnel recognition model. For another example, S511 can be implemented by the server or automatically by the terminal device. In another optional embodiment, the terminal device locally implements the retraining of the risk identification model and completes the optimization of the risk identification model by itself. Targeted model optimization is implemented locally, so that the risk identification model has a high degree of matching with the current vault monitoring scenario.

[0157] In summary, Figure 6 A cash operation monitoring method provided in an embodiment of the present application is as follows: Figure 6 As shown, the method includes:

[0158] S601. Receive a monitoring image of a target cash operation in real time, and perform local image enhancement of a target local area on the monitoring image to obtain an enhanced monitoring image.

[0159] In this embodiment, the monitoring images are collected and transmitted in real time by a camera arranged in the cash operation area.

[0160] In this embodiment, the target local area is a specified area of ​​cash in the monitoring image when performing the target cash operation. Optionally, based on the specified execution area of ​​the target cash operation and the shooting angle of the monitoring image, the target local area is predicted and extracted from the monitoring image.

[0161] S602: Input the enhanced monitoring image into a pre-configured risk identification model to obtain a risk identification result output by the risk identification model.

[0162] In this embodiment, the risk identification model includes a cash identification model and a personnel identification model, and the risk identification result includes a cash identification result output by the cash identification model and a personnel identification result output by the personnel identification model.

[0163] In this embodiment, after the enhanced surveillance image is input into the cash recognition model in the edge computing box to obtain the cash recognition result, if the cash recognition result indicates the presence of a cash image, the enhanced surveillance image is input into the personnel recognition model in the edge computing box for personnel recognition.

[0164] S603: Obtain the security operation rules of the target cash operation.

[0165] In this embodiment, the security operation rules at least include specified personnel, quantity and specified personnel types in the cash area.

[0166] S604. Determine whether the target cash operation complies with safety operation rules based on the risk identification result.

[0167] S605: If the safety operation rules are not met, an alarm message is generated.

[0168] In this embodiment, the warning information includes a warning image, the risk identification result and a safe operation rule, wherein the warning image is a monitoring image corresponding to the target cash operation not complying with the safe operation rule.

[0169] It can be seen from the above technical scheme that a cash operation monitoring method provided in an embodiment of the present application performs risk identification on the monitoring images of cash operations in real time through a risk identification model deployed on a terminal device, and issues timely warnings for risk behaviors that do not comply with safety operation rules, thereby improving the real-time risk identification of cash operations. Specifically, by enhancing the local image of the target local area, the cash recognizability is improved in a targeted manner, the complexity of local image enhancement is reduced, the recognition speed is increased, and the accuracy and robustness of the recognition results of the risk identification model are improved.

[0170] It should be noted that the cash operation monitoring method and related devices provided by the present invention can be used in the field of artificial intelligence or finance. The above is only an example and does not limit the application field of the cash operation monitoring method and related devices provided by the present invention.

[0171] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant laws, regulations and standards of relevant countries and regions.

[0172] A cash operation monitoring method provided in an embodiment of the present application is introduced above. A device for executing the cash operation monitoring method will be introduced below.

[0173] See also Figure 7 , Figure 7 This is a schematic diagram of the structure of a cash operation monitoring device provided in an embodiment of the present application. Figure 7 As shown, the cash operation monitoring device 700 includes:

[0174] The local enhancement unit 701 is used to receive the monitoring image of the target cash operation in real time, and perform local image enhancement of the target local area on the monitoring image to obtain an enhanced monitoring image, wherein the target local area is a specified area of ​​cash in the monitoring image when the target cash operation is performed;

[0175] The risk identification unit 702 is used to input the enhanced surveillance image into a pre-configured risk identification model to obtain a risk identification result output by the risk identification model, wherein the risk identification model includes a cash identification model and a personnel identification model, and the risk identification result includes a cash identification result output by the cash identification model and a personnel identification result output by the personnel identification model;

[0176] An operation rule determination unit 703 is used to obtain a safety operation rule for the target cash operation, wherein the safety operation rule at least includes a specified number of personnel and a specified type of personnel in the cash area;

[0177] A risk determination unit 704 is used to determine whether the target cash operation complies with the safety operation rules according to the risk identification result;

[0178] The risk warning unit 705 is used to generate warning information if the safety operation rules are not met, and the warning information includes a warning image, the risk identification result and the safety operation rules.

[0179] In a possible implementation, the local enhancement unit is used to perform local image enhancement of a target local area on the surveillance image, and when the enhanced surveillance image is obtained, it is specifically used to:

[0180] extracting the target local area from the surveillance image based on a specified execution area of ​​the target cash operation and a shooting angle of the surveillance image;

[0181] The contrast enhancement and edge sharpening are performed on the local area of ​​the target to obtain the enhanced monitoring image.

[0182] In a possible implementation, the risk identification unit is used to input the enhanced monitoring image into a pre-configured risk identification model, and when obtaining the risk identification result output by the risk identification model, specifically to:

[0183] After the enhanced monitoring image is input into the cash recognition model to obtain the cash recognition result, if the cash recognition result indicates that a cash image exists, the enhanced monitoring image is input into the personnel recognition model to obtain the personnel recognition result.

[0184] In a possible implementation, the cash operation monitoring device further includes: a retraining unit, configured to:

[0185] Obtain feedback information of historical alarm information, wherein the feedback information of the historical alarm information includes whether the cash recognition result and the person recognition result are correct, as well as the corrected cash recognition result and the corrected person recognition result

[0186] Annotating the warning image based on the feedback information of the warning information to obtain a retraining data set;

[0187] The retraining data set includes a cash recognition retraining data set and a personnel recognition retraining data set, the cash recognition retraining data set includes at least a first type of warning image and a cash label, wherein the first type of warning image is a warning image of an incorrect cash recognition result, and the cash label of the first type of warning image is a corrected cash recognition result, and the personnel recognition retraining data set includes at least a second type of warning image and a personnel label, wherein the second type of warning image is a warning image of an incorrect personnel recognition result, and the personnel label of the second type of warning image is a corrected personnel recognition result;

[0188] When a preset retraining time is reached, the risk identification model is retrained based on the retraining data set to obtain retraining parameter information; or, the retraining data set and the retraining instruction are sent to the server, so that the server retrains the risk identification model based on the retraining data set and returns the retraining parameter information; the retraining parameter information includes a hyperparameter group and a model parameter;

[0189] The risk identification model is optimized based on the retraining parameter information.

[0190] In a possible implementation, when the retraining unit is used to retrain the risk identification model based on the retraining data set, it is specifically used to:

[0191] Obtaining pre-configured hyperparameter search information, wherein the hyperparameter search information includes a preset type and a corresponding preset value range;

[0192] If the number of training times does not reach a preset number threshold, the risk identification model is configured using the preset hyperparameter group, and the risk identification model is retrained based on the retraining data set, and retraining parameter information is obtained after a preset training completion condition is met;

[0193] The preset hyperparameter group includes preset values ​​of multiple preset types of hyperparameters obtained by randomly selecting values ​​in the hyperparameter search information;

[0194] If the number of training times reaches the threshold, a target hyperparameter group is obtained according to the hyperparameter search results, the risk identification model is configured using the target hyperparameter group, and the risk identification model is trained based on the retraining data set, and after the training completion condition is met, retraining parameter information is obtained; wherein the target hyperparameter group includes target values ​​of hyperparameters of multiple target types.

[0195] In a possible implementation, the cash operation monitoring device further includes a model deployment unit, which is used to: before inputting the enhanced monitoring image into a pre-configured risk identification model:

[0196] The receiving server trains the machine learning model based on the training data set to obtain initial configuration information of the cash recognition model and the personnel recognition model, and deploys the cash recognition model and the personnel recognition model based on the initial configuration information, wherein the initial configuration information includes an initial hyperparameter group and initial model parameters.

[0197] The present application also provides an electronic device in an embodiment. Figure 8 As shown, it shows a schematic diagram of the structure of an electronic device suitable for implementing the embodiments of the present application. The electronic device in the embodiments of the present application may include but is not limited to fixed terminals such as mobile phones, notebook computers, PDAs (personal digital assistants), PADs (tablet computers), desktop computers, etc. Figure 8 The electronic device shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0198] like Figure 8 As shown, the electronic device may include a processing device (e.g., a central processing unit, a graphics processing unit, etc.) 801, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 802 or a program loaded from a storage device 808 to a random access memory (RAM) 803. When the electronic device is powered on, various programs and data required for the operation of the electronic device are also stored in the RAM 803. The processing device 801, the ROM 802, and the RAM 803 are connected to each other via a bus 804. An input / output (I / O) interface 805 is also connected to the bus 804.

[0199] Typically, the following devices may be connected to the I / O interface 805: input devices 806 including, for example, a touch screen, a touch pad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; output devices 807 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; storage devices 808 including, for example, a memory card, a hard disk, etc.; and communication devices 809. The communication device 809 may allow the electronic device to communicate with other devices wirelessly or by wire to exchange data. Although Figure 6An electronic device having various devices is shown, but it should be understood that it is not required to implement or possess all the devices shown. More or fewer devices may be implemented or possessed instead.

[0200] Also provided in an embodiment of the present application is a computer program product including computer-readable instructions. When the computer-readable instructions are executed on an electronic device, the electronic device implements any cash operation monitoring method provided in the embodiment of the present application.

[0201] A computer-readable storage medium is also provided in an embodiment of the present application. The storage medium carries one or more computer programs. When the one or more computer programs are executed by an electronic device, the electronic device can implement any cash operation monitoring method provided in the embodiment of the present application.

[0202] It should also be noted that the device embodiments described above are merely schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed over multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. In addition, in the drawings of the device embodiments provided by the present application, the connection relationship between the modules indicates that there is a communication connection between them, which may be specifically implemented as one or more communication buses or signal lines.

[0203] Through the description of the above implementation mode, the technicians in the field can clearly understand that the present application can be implemented by means of software plus necessary general hardware, and of course, it can also be implemented by special hardware including special integrated circuits, special CPUs, special memories, special components, etc. In general, all functions completed by computer programs can be easily implemented by corresponding hardware, and the specific hardware structure used to implement the same function can also be various, such as analog circuits, digital circuits or special circuits. However, for the present application, software program implementation is a better implementation mode in more cases. Based on such an understanding, the technical solution of the present application is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which is stored in a readable storage medium, such as a computer floppy disk, a U disk, a mobile hard disk, a ROM, a RAM, a disk or an optical disk, etc., including a number of instructions to enable a computer device (which can be a personal computer, a training device, or a network device, etc.) to execute the methods described in each embodiment of the present application.

[0204] In the above embodiments, all or part of the embodiments may be implemented by software, hardware, firmware or any combination thereof. When implemented by software, all or part of the embodiments may be implemented in the form of a computer program product.

[0205] The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions may be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions may be transmitted from a website site, a computer, a training device, or a data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode to another website site, computer, training device, or data center. The computer-readable storage medium may be any available medium that a computer can store or a data storage device such as a training device, a data center, etc. that includes one or more available media integrations. The available medium may be a magnetic medium, (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid-state drive (SSD)), etc.

Claims

1. A method for monitoring cash operations, characterized in that: Applied to terminals, including: receiving a monitoring image of a target cash operation in real time, performing local image enhancement of a target local area on the monitoring image to obtain an enhanced monitoring image, wherein the target local area is a specified area of ​​cash in the monitoring image when the target cash operation is performed; Inputting the enhanced surveillance image into a pre-configured risk identification model to obtain a risk identification result output by the risk identification model, wherein the risk identification model includes a cash identification model and a personnel identification model, and the risk identification result includes a cash identification result output by the cash identification model and a personnel identification result output by the personnel identification model; Obtaining a safety operation rule for the target cash operation, wherein the safety operation rule at least includes a prescribed number of personnel and prescribed types of personnel within a cash area; According to the risk identification result, determining whether the target cash operation complies with the safety operation rules; If the safety operation rules are not met, an alarm message is generated, and the alarm message includes an alarm image, the risk identification result and the safety operation rules.

2. The cash operation monitoring method according to claim 1, characterized in that: The step of performing local image enhancement on a target local area of ​​a surveillance image to obtain an enhanced surveillance image includes: extracting the target local area from the surveillance image based on a specified execution area of ​​the target cash operation and a shooting angle of the surveillance image; The contrast enhancement and edge sharpening are performed on the local area of ​​the target to obtain the enhanced monitoring image.

3. The cash operation monitoring method according to claim 1, characterized in that: The step of inputting the enhanced monitoring image into a pre-configured risk identification model to obtain a risk identification result output by the risk identification model includes: After the enhanced monitoring image is input into the cash recognition model to obtain the cash recognition result, if the cash recognition result indicates that a cash image exists, the enhanced monitoring image is input into the personnel recognition model to obtain the personnel recognition result.

4. The cash operation monitoring method according to claim 1, characterized in that: The cash operation monitoring method further includes: Obtain feedback information of historical alarm information, wherein the feedback information of the historical alarm information includes whether the cash recognition result and the person recognition result are correct, as well as the corrected cash recognition result and the corrected person recognition result Annotating the warning image based on the feedback information of the warning information to obtain a retraining data set; The retraining data set includes a cash recognition retraining data set and a personnel recognition retraining data set, the cash recognition retraining data set includes at least a first type of warning image and a cash label, wherein the first type of warning image is a warning image of an incorrect cash recognition result, and the cash label of the first type of warning image is a corrected cash recognition result, and the personnel recognition retraining data set includes at least a second type of warning image and a personnel label, wherein the second type of warning image is a warning image of an incorrect personnel recognition result, and the personnel label of the second type of warning image is a corrected personnel recognition result; When a preset retraining time is reached, the risk identification model is retrained based on the retraining data set to obtain retraining parameter information; or, the retraining data set and the retraining instruction are sent to the server, so that the server retrains the risk identification model based on the retraining data set and returns the retraining parameter information; the retraining parameter information includes a hyperparameter group and a model parameter; The risk identification model is optimized based on the retraining parameter information.

5. The cash operation monitoring method according to claim 4, characterized in that: The retraining of the risk identification model based on the retraining data set includes: Obtaining pre-configured hyperparameter search information, wherein the hyperparameter search information includes a preset type and a corresponding preset value range; If the number of training times does not reach a preset number threshold, the risk identification model is configured using the preset hyperparameter group, and the risk identification model is retrained based on the retraining data set, and retraining parameter information is obtained after a preset training completion condition is met; The preset hyperparameter group includes preset values ​​of multiple preset types of hyperparameters obtained by randomly selecting values ​​in the hyperparameter search information; If the number of training times reaches the threshold, a target hyperparameter group is obtained according to the hyperparameter search results, the risk identification model is configured using the target hyperparameter group, and the risk identification model is trained based on the retraining data set, and after the training completion condition is met, retraining parameter information is obtained; wherein the target hyperparameter group includes target values ​​of hyperparameters of multiple target types.

6. The cash operation monitoring method according to claim 1, characterized in that: The cash operation monitoring method further includes, before inputting the enhanced monitoring image into a pre-configured risk identification model: The receiving server trains the machine learning model based on the training data set to obtain initial configuration information of the cash recognition model and the personnel recognition model, and deploys the cash recognition model and the personnel recognition model based on the initial configuration information, wherein the initial configuration information includes an initial hyperparameter group and initial model parameters.

7. A cash operation monitoring device, characterized in that: include: A local enhancement unit is used to receive a monitoring image of a target cash operation in real time, and perform local image enhancement of a target local area on the monitoring image to obtain an enhanced monitoring image, wherein the target local area is a specified area of ​​cash in the monitoring image when the target cash operation is performed; a risk identification unit, configured to input the enhanced surveillance image into a pre-configured risk identification model to obtain a risk identification result output by the risk identification model, wherein the risk identification model includes a cash identification model and a personnel identification model, and the risk identification result includes a cash identification result output by the cash identification model and a personnel identification result output by the personnel identification model; An operation rule determination unit, used to obtain a safety operation rule for the target cash operation, the safety operation rule at least including a specified number of personnel and a specified type of personnel in the cash area; A risk determination unit, configured to determine whether the target cash operation complies with safety operation rules according to the risk identification result; The risk warning unit is used to generate warning information if the safety operation rules are not met. The warning information includes a warning image, the risk identification result and the safety operation rules.

8. A computer program product, characterized in that It comprises computer-readable instructions, and when the computer-readable instructions are executed on an electronic device, the electronic device implements the cash operation monitoring method as claimed in any one of claims 1 to 6.

9. An electronic device, characterized in that: The method comprises at least one processor and a memory connected to the processor, wherein: The memory is used to store computer programs; The processor is used to execute the computer program so that the electronic device can implement the cash operation monitoring method as described in any one of claims 1 to 6.

10. A computer storage medium, characterized in that: The storage medium carries one or more computer programs, which, when executed by an electronic device, enable the electronic device to implement the cash operation monitoring method as described in any one of claims 1 to 6.

Images