Camera state prediction method, apparatus, device, and readable storage medium

Abstract

The application discloses a camera state prediction method, device and equipment and a readable storage medium, and the method comprises the following steps: acquiring camera core data of a permission verification camera; performing camera element evaluation on the camera core data to obtain a part image quality parameter corresponding to a part image of a permission verification part, a working performance parameter corresponding to camera working data, and a core device performance parameter corresponding to core device usage data; performing available state prediction processing on the permission verification camera according to the part image quality parameter, the working performance parameter, the core device performance parameter and a camera element parameter weight to obtain a predicted available state corresponding to the permission verification camera; and the predicted available state is used for evaluating the availability of the permission verification camera. The application improves the prediction accuracy of the available state of the permission verification camera.

Description

Technical Field

[0001] This application relates to the field of computer technology, and in particular to a method, apparatus, device, and readable storage medium for predicting the state of a camera. Background Technology

[0002] With the development of computer technology, in many business scenarios requiring permission verification, dedicated permission verification cameras can be used to capture partial images of business objects, and then the validity of those images can be determined to confirm whether the object has the relevant permissions. If a permission verification camera malfunctions or fails, the captured partial images will be inaccurate, leading to a decline in core metrics such as user experience. Therefore, during business maintenance, it is usually necessary to prepare corresponding inventory and plans in advance to replace malfunctioning permission verification cameras. Thus, predicting the availability of permission verification cameras is a core technical problem that needs to be solved during business maintenance.

[0003] Existing methods for predicting the availability of access control cameras rely on the availability of their core components. For example, if the core component is a laser with a lifespan of two years, maintenance might assume the camera's lifespan is also two years. However, in practice, the lifespan of the access control camera is affected by other components, including the laser itself, which is influenced by usage patterns and environment. Excessive usage frequency or overheating can shorten its lifespan. Therefore, relying solely on the availability of core components to determine the availability of access control cameras is not very accurate. Summary of the Invention

[0004] This application provides a camera status prediction method, apparatus, device, and readable storage medium, which can improve the accuracy of predicting the availability status of permission verification cameras.

[0005] One embodiment of this application provides a camera state prediction method, including:

[0006] Obtain the core data of the permission verification camera; the permission verification camera refers to the camera used to capture images of the permission verification area of ​​the first object; the permission verification area images are used to verify the business permissions of the first object; the core data of the camera includes the permission verification area images, camera working data, and core component usage data;

[0007] The camera's core data is evaluated to obtain the image quality parameters of the authorized verification area, the working performance parameters of the camera's working data, and the core device performance parameters of the core device usage data.

[0008] Based on the image quality parameters of the location, the operating performance parameters, the performance parameters of the core components, and the weighted parameters of the camera element, the access control camera is processed to predict its availability status, thus obtaining the predicted availability status of the access control camera. The predicted availability status is used to evaluate the availability of the access control camera.

[0009] One embodiment of this application provides a camera state prediction device, including:

[0010] The data acquisition module is used to acquire the core data of the permission verification camera; the permission verification camera refers to the camera used to capture images of the permission verification area of ​​the first object; the permission verification area images are used to verify the business permissions of the first object; the core camera data includes the permission verification area images, camera working data, and core component usage data;

[0011] The element evaluation module is used to evaluate the core data of the camera to obtain the image quality parameters of the area corresponding to the permission verification area, the working performance parameters of the camera working data, and the core device performance parameters of the core device usage data.

[0012] The status prediction module is used to predict the availability status of the access verification camera based on the image quality parameters of the location, the working performance parameters, the performance parameters of the core components, and the weighted parameters of the camera element parameters. The predicted availability status is used to evaluate the availability of the access verification camera.

[0013] In one possible implementation, the feature evaluation module is used to perform camera feature evaluation on the core camera data. When obtaining the image quality parameters corresponding to the permission verification area image, the working performance parameters corresponding to the camera working data, and the core device performance parameters corresponding to the core device usage data, it is specifically used to perform the following operations:

[0014] Image quality assessment is performed on the images of the authorization verification area to obtain the corresponding image quality parameters of the authorization verification area.

[0015] The camera's working data is used to evaluate its performance, and the corresponding performance parameters are obtained.

[0016] The performance of core components is evaluated based on the usage data, and the corresponding performance parameters of the core components are obtained.

[0017] In one possible implementation, the part image quality parameters include conventional image quality parameters;

[0018] The element evaluation module is used to evaluate the image quality of the permission verification area image. When obtaining the image quality parameters of the corresponding area image, it is specifically used to perform the following operations:

[0019] Perform edge detection on the image of the permission verification area to obtain the edge of the area image;

[0020] Perform image contrast calculation on the image of the authorization verification area to obtain the image contrast of the area;

[0021] Brightness distribution detection is performed on the image of the permission verification area to obtain the brightness distribution of the area image;

[0022] The image sharpness of the permission verification area is calculated to obtain the image sharpness of the area.

[0023] The traditional image quality parameters corresponding to the permission verification area image are determined based on the area image edge, area image contrast, area image brightness distribution, and area image clarity.

[0024] In one possible implementation, the permission verification area image includes a color area image and an infrared area image; the area image quality parameters include intelligent image quality parameters.

[0025] The element evaluation module is used to evaluate the image quality of the permission verification area image. When obtaining the image quality parameters of the corresponding area image, it is specifically used to perform the following operations:

[0026] Image preprocessing is performed on the color region image to obtain a standardized color region image;

[0027] Image preprocessing is performed on the infrared region image to obtain a standardized infrared region image;

[0028] The texture sharpness of standardized color area images is evaluated to obtain texture sharpness parameters;

[0029] The vein clarity was evaluated using standardized infrared images of specific areas to obtain vein clarity parameters.

[0030] The texture clarity parameter and the vein clarity parameter are weighted and summed to obtain the intelligent image quality parameter corresponding to the image of the authorization verification area.

[0031] In one possible implementation, the feature evaluation module is used to preprocess the color region image to obtain a standardized color region image, specifically by performing the following operations:

[0032] Image normalization is performed on the color part image to obtain a normalized color part image;

[0033] Denoising is performed on the normalized color region image to obtain a denoised color region image;

[0034] The image size of the denoised color region image is adjusted to a standardized image size to obtain a standardized color region image.

[0035] In one possible implementation, the camera's operating data includes device temperature data; the operating performance parameters include device temperature performance parameters.

[0036] The element evaluation module is used to evaluate the working performance of the camera's working data. When obtaining the working performance parameters corresponding to the camera's working data, it is specifically used to perform the following operations:

[0037] If the device temperature data is less than or equal to the standard temperature data, then the standard temperature performance parameter shall be determined as the device temperature performance parameter.

[0038] If the device temperature data is greater than the standard temperature data, then determine the temperature data difference between the device temperature data and the standard temperature data, and determine the loss temperature performance parameters based on the temperature data difference;

[0039] The difference between the standard temperature performance parameter and the loss temperature performance parameter is determined as the device temperature performance parameter.

[0040] In one possible implementation, the camera's operating data includes image brightness data; the operating performance parameters include image brightness performance parameters.

[0041] The element evaluation module is used to evaluate the working performance of the camera's working data. When obtaining the working performance parameters corresponding to the camera's working data, it is specifically used to perform the following operations:

[0042] Determine the magnitude of the deviation between the image brightness data and the standard brightness data;

[0043] If the deviation is less than or equal to the deviation threshold, the standard brightness performance parameter is determined as the image brightness performance parameter.

[0044] If the deviation amplitude is greater than the deviation amplitude threshold, then the amplitude difference between the deviation amplitude and the deviation amplitude threshold is determined, and the loss brightness performance parameter is determined based on the amplitude difference.

[0045] The difference between the standard brightness performance parameter and the lost brightness performance parameter is determined as the image brightness performance parameter.

[0046] In one possible implementation, the camera operating data includes device frame rate data; the operating performance parameters include device frame rate performance parameters.

[0047] The element evaluation module is used to evaluate the working performance of the camera's working data. When obtaining the working performance parameters corresponding to the camera's working data, it is specifically used to perform the following operations:

[0048] If the device frame rate data is greater than or equal to the standard frame rate data, then the standard frame rate performance parameter is determined as the device frame rate performance parameter.

[0049] If the device frame rate data is less than the standard frame rate data, then the frame rate data difference between the standard frame rate data and the device frame rate data is determined, and the loss frame rate performance parameter is determined based on the frame rate data difference.

[0050] The difference between the standard frame rate performance parameter and the lossy frame rate performance parameter is determined as the device frame rate performance parameter.

[0051] In one possible implementation, the core device usage data includes the number of times the core device is used;

[0052] The element evaluation module is used to evaluate the performance of core components based on their usage data. When obtaining the core component performance parameters corresponding to the usage data, it is specifically used to perform the following operations:

[0053] The usage time of the core component is determined based on the number of times the core component is used and the duration of a single use of the core component.

[0054] Obtain the total available time of core components, determine the ratio between the used time of core components and the total available time of core components, and determine the core component performance parameters corresponding to the core component usage data based on the ratio of the used time.

[0055] In one possible implementation, the predicted available state includes a predicted available score;

[0056] The status prediction module is used to predict the availability status of the access control camera based on the image quality parameters, operating performance parameters, core component performance parameters, and camera element parameter weights. Specifically, when obtaining the predicted availability status of the access control camera, it performs the following operations:

[0057] A camera element fusion feature vector is generated based on the image quality parameters of the location, the working performance parameters, the performance parameters of the core components, and the weights of the camera element parameters.

[0058] Obtain a camera availability state prediction model; the camera availability state prediction model contains N camera availability state decision trees; different camera availability state decision trees are used to represent different camera availability state decision strategies; N is a positive integer;

[0059] Traverse the available state decision trees of N cameras to obtain the available state decision tree of the i-th camera; i is a positive integer less than N;

[0060] The camera availability status decision score is obtained by processing the camera element fusion feature vector through the availability status decision tree of the i-th camera.

[0061] When the decision tree of available states of N cameras is completed, the decision scores of available states of N cameras are weighted and summed to obtain the predicted available score of the permission verification camera.

[0062] In one possible implementation, predicting available states includes predicting available scores; the state prediction module is also used to perform the following operations:

[0063] Based on the predicted available score and the total available time of the cameras, determine the predicted remaining usage time of the permission verification camera;

[0064] If the predicted remaining usage time is less than the remaining usage time threshold, a maintenance notification is sent to the camera management terminal; the maintenance notification is used to instruct the permission verification camera to undergo maintenance.

[0065] One embodiment of this application provides a computer device, including: a processor, a memory, and a network interface;

[0066] The processor is connected to the memory and the network interface. The network interface is used to provide a data communication network element, the memory is used to store a computer program, and the processor is used to call the computer program to execute the method in the embodiments of this application.

[0067] One aspect of this application provides a computer-readable storage medium storing a computer program adapted for loading by a processor and executing the methods described in this application.

[0068] One aspect of this application provides a computer program product or computer program that includes computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the method described in this application.

[0069] In this embodiment, core camera data of an access control camera can be obtained. This core data includes images of the access control area, camera operating data, and core component usage data. Then, camera element evaluation is performed on the obtained core data to obtain image quality parameters corresponding to the access control area images, operating performance parameters corresponding to the camera operating data, and core component performance parameters corresponding to the core component usage data. Finally, based on the image quality parameters, operating performance parameters, core component performance parameters, and camera element parameter weights, the access control camera's availability status is predicted to obtain the predicted availability status of the access control camera. The predicted availability status is used to assess the usability of the access control camera. The method provided in this embodiment allows for dynamic evaluation of camera elements in the core data generated during actual use of the access control camera, obtaining multiple camera element parameters that reflect the actual usage of the access control camera. Then, based on these multiple camera element parameters and the influence weight of each camera element parameter on the availability status of the access control camera, the availability status of the access control camera is predicted, improving the accuracy of the prediction of the access control camera's availability status. Attached Figure Description

[0070] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0071] Figure 1 This is a schematic diagram of a network architecture provided in an embodiment of this application;

[0072] Figure 2 This is a schematic diagram of a camera state prediction method provided in an embodiment of this application;

[0073] Figure 3 This is a flowchart illustrating a camera state prediction method provided in an embodiment of this application;

[0074] Figure 4 This is a schematic diagram of a camera availability status decision tree provided in an embodiment of this application;

[0075] Figure 5 This is a flowchart illustrating a camera state prediction method provided in an embodiment of this application;

[0076] Figure 6 This is an overall flowchart of a method for predicting the availability status of a palm-swiping camera provided in an embodiment of this application;

[0077] Figure 7 This application provides a schematic diagram of the structure of a camera state prediction device;

[0078] Figure 8 This is a schematic diagram of the structure of a computer device provided in an embodiment of this application. Detailed Implementation

[0079] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0080] To facilitate understanding, the following brief explanations are provided for some of the terms:

[0081] A webcam (also known as a computer camera, computer eye, or electronic eye) is a video input device widely used in video conferencing, telemedicine, and real-time monitoring. Users communicate with each other online using images and sound through the webcam. Additionally, users can utilize it in various popular digital imaging and audio-visual processing fields.

[0082] Color image: A color image captured by the camera's color sensor in natural light. In "face / palm payment", it is generally used for: the selection of facial / palm images and the comparison and recognition of facial / palm features.

[0083] Infrared image: An infrared image captured by the infrared sensor of a camera, which is formed by the image of the infrared light. In "face recognition / palm payment", it is generally used for: liveness detection of the object to be inspected.

[0084] Optimization: Select a set of color images, depth images, and infrared images that meet the prerequisites for liveness detection and comparison recognition algorithms (face scanning: color, depth, infrared; palm scanning: color, infrared). Optimization is achieved by selecting the color image based on face / palm angle, size, centering, and clarity; the infrared image based on brightness; and the depth image based on completeness (face scanning).

[0085] Threshold: T, the algorithm score that satisfies a certain false recognition rate and pass rate (rejection rate), generally an integer between 0 and 100 (some use a 10-point system or a 1000-point system, but this application recommends a uniform 100-point system).

[0086] Image quality inspection: Detects the quality of facial or hand images by assessing factors such as lighting, posture, and angle.

[0087] Deep learning technology: Deep learning is a machine learning technique that utilizes deep neural network systems. The concept of deep learning originated from research on artificial neural networks; for example, the multilayer perceptron with multiple hidden layers is a type of deep learning architecture. Deep learning discovers distributed feature representations of data by combining low-level features to form more abstract high-level representations of attribute categories or features.

[0088] Please see Figure 1 , Figure 1 This is a schematic diagram of a network architecture provided in an embodiment of this application. Please refer to [link / reference]. Figure 1 , Figure 1 This is a schematic diagram of a network architecture provided in an embodiment of this application. For example... Figure 1 As shown, the network architecture may include a server 100 and a cluster of authentication devices, which may include authentication devices 10a, 10b, ..., 10n. Any authentication device in the cluster may have a communication connection with the server 100; for example, authentication device 10a and authentication device 10b may have a communication connection with the server 100. This communication connection is not limited to a specific method; it can be established directly or indirectly via wired communication, wireless communication, or other methods. This application does not impose any limitations on this method. Both the server 100 and the authentication device cluster are exemplary embodiments of computer equipment.

[0089] like Figure 1 As shown, any permission verification device in the permission verification device cluster can be integrated with a permission verification camera to support business permission verification of business objects. For example, permission verification device 10a can be integrated with a permission verification camera 110. Permission verification device 10a can capture images of the permission verification parts of business objects through the permission verification camera. If permission verification device 10a has a local cache of part image feature library, it can complete the business permission verification of business objects locally. Alternatively, permission verification device 10a can compress, encrypt, and encapsulate the permission verification part images to generate a business verification request, and send the business verification request to server 100 to request server 100 to complete the business permission verification of business objects.

[0090] Among them, such as Figure 1The permission verification device 10a shown may include, but is not limited to, payment devices, smartphones, computers, smart voice interaction devices, smart home appliances, vehicle terminals, aircraft, and other smart devices capable of installing permission verification cameras.

[0091] In one feasible embodiment, taking palm-scanning payment authorization verification as an example, the authorization verification device 10a refers to a computer device that integrates an authorization verification camera and can collect the user's palm image before making payment. For example, the authorization verification device 10a can be a terminal providing palm-scanning payment functionality, a payment collection device provided by a merchant, or a vending machine. Illustratively, after the user performs a trigger operation on the payment option on the authorization verification device 10a and selects the palm-scanning payment method, the authorization verification device 10a will, after obtaining full authorization from the user, open the authorization verification camera to collect the user's palm image. Optionally, after the authorization verification device 10a completes palmprint recognition locally, if the palmprint recognition is successful, it interacts with the server 100 for value transfer; or, the authorization verification device 10a requests the server 100 to perform palmprint recognition in the cloud, and if the palmprint recognition is successful, the server 100 performs value transfer.

[0092] Among them, such as Figure 1 The server 100 shown can be an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN (Content Delivery Network), and big data and artificial intelligence platforms.

[0093] It is understandable that the availability of the aforementioned access control camera directly affects the success of the entire business access control verification process. Therefore, being able to predict the availability of the access control camera in real time and plan the maintenance of the access control equipment can better ensure user experience. Therefore, this application provides a camera status prediction method that can predict the availability of the access control camera based on its actual usage, thereby assessing the availability of the access control camera. Specifically, this camera status prediction method may include: acquiring the core data of the access control camera; wherein, the access control camera refers to a camera used to collect images of the access control area of ​​a first object, the access control area images are used to perform business access control verification on the first object, and the core data of the camera includes the access control area images, camera working data, and core component usage data; then, evaluating the camera elements of the core data to obtain the image quality parameters corresponding to the access control area images, the working performance parameters corresponding to the camera working data, and the core component performance parameters corresponding to the core component usage data; finally, performing availability status prediction processing on the access control camera based on the image quality parameters, working performance parameters, core component performance parameters, and the weights of the camera element parameters to obtain the predicted availability status of the access control camera; the predicted availability status is used to assess the availability of the access control camera.

[0094] It is understood that the camera state prediction method provided in this application embodiment can be executed by a computer device, which includes, but is not limited to, the server 100 and the authorization verification device mentioned above; this application embodiment can be applied to various scenarios, including but not limited to cloud technology, artificial intelligence, blockchain, smart driving and other scenarios.

[0095] It should be noted that this application may display prompt interfaces, pop-ups, or output voice prompts before and during the collection of user data. These prompt interfaces, pop-ups, or voice prompts are used to inform the user that their data is being collected. This ensures that the application only begins the steps for collecting user data after receiving confirmation from the user regarding the prompt interface or pop-up; otherwise (i.e., without user confirmation), the steps for collecting user data end, meaning no user data is collected. In other words, all user data collected in this application is collected with the user's consent and authorization, and the collection, use, and processing of related user data must comply with the relevant laws, regulations, and standards of the relevant regions.

[0096] It is understood that the above scenarios are merely examples and do not constitute a limitation on the application scenarios of the technical solutions provided in the embodiments of this application. The technical solutions of this application can also be applied to other scenarios. For example, as those skilled in the art will know, with the evolution of system architecture and the emergence of new business scenarios, the technical solutions provided in the embodiments of this application are also applicable to similar technical problems.

[0097] To better understand the above process of predicting the availability of the access control camera, please refer to [link / reference]. Figure 2 , Figure 2 This is a scene illustration of a camera state prediction method provided in an embodiment of this application. For example... Figure 2 As shown, the permission verification device 20 can be the one described above. Figure 1 Any of the permission verification devices in the permission verification device cluster shown, for example, permission verification device 20 can be permission verification device 10a; server 200 can be the aforementioned Figure 1 The server shown is 100.

[0098] like Figure 2 As shown, the authentication device 20 can be a palm-swiping device that integrates an authentication camera 210 and can collect images of the user's palm for payment. The core components of the authentication camera 210 have a lifespan of 2 years. Therefore, before the authentication camera 210 is put into use, its lifespan can be roughly estimated to be 2 years. However, due to the actual usage environment and the influence of other components, the actual lifespan of the authentication camera 210 may be reduced or increased. Therefore, the future availability of the authentication camera 210 can be predicted based on its actual usage. Assuming that the authentication camera 210 has started to be used normally, in order to perform preventative maintenance before the authentication camera 210 completely fails and to avoid authentication interruption affecting the user's payment experience, the core camera data of the authentication camera 210 can be obtained periodically to predict the latest availability of the authentication camera 210.

[0099] like Figure 2As shown, in a prediction of the availability status of the access verification camera 210, the access verification device 20 can first obtain the latest camera core data 220 corresponding to the access verification camera 210, and then send the camera core data 220 to the server 200. The camera core data 220 may include access verification area image 2201, camera working data 2202, and core component usage data 2203. The access verification area image 2201 may include the latest color palm print image, infrared palm vein image, etc., of the business object collected by the access verification camera 210; the camera working data 2202 may include the latest working data of the access verification camera 210, such as CPU (Central Processing Unit) temperature, laser temperature, average brightness of the color image, average brightness of the infrared image, original frame rate, and aligned frame rate; the core component usage data 2203 may include the number of times and duration of use of the core components of the access verification camera 210, such as the laser, since its inception.

[0100] like Figure 2 As shown, server 200 can evaluate camera elements in camera core data 220 to obtain part image quality parameters 2100 corresponding to the permission verification part image 2201, working performance parameters 2200 corresponding to the camera working data 2202, and core device performance parameters 2300 corresponding to the core device usage data 2203. The part image quality parameter 2100 can be a score of the image quality of the permission verification part image 2201; a higher part image quality parameter 2100 indicates higher image quality. The working performance parameter 2200 can be a score of the camera's working performance; a higher working performance parameter 2200 indicates higher camera working performance. The core device performance parameter 2203 can be a score of the core device's performance; a higher core device performance parameter 2203 indicates higher core device performance. Then, server 200 can obtain the camera element parameter weights. Each camera element parameter weight is used to characterize the importance of its corresponding camera element in predicting the availability status of the permission verification camera 210. Server 200 can adjust the obtained part image quality parameter 2100, working performance parameter 2200, and core component performance parameter 2300 based on the camera element parameter weights, and then perform the prediction processing of the availability status of the permission verification camera 210 based on the adjusted camera element parameters. Figure 2As shown, server 200 can perform predictive processing on the availability status of permission verification camera 210 based on image quality weight × part image quality parameter 2100, working performance weight × working performance parameter 2200, and core component performance weight × core component performance parameter 2300, to obtain predicted availability status 2400 of permission verification camera 210. The predicted availability status 2400 can be used to evaluate the availability level of permission verification camera. For example, the predicted availability status 2400 can include a predicted availability score. Assuming the predicted availability score takes the value [0, 100], a higher predicted availability score indicates a higher availability level of permission verification camera, and a longer remaining usable lifespan. Optionally, a maintenance score threshold can be set. When the predicted availability score is lower than this maintenance score threshold, business maintenance personnel can begin preparing a preventative maintenance plan for the predicted permission verification camera 210 or prepare a replacement camera.

[0101] Therefore, the camera status prediction method provided in this application embodiment can predict the future availability status of the permission verification camera by periodically using the latest core camera data generated during actual use. Based on the prediction results, maintenance plans and resource allocation can be dynamically adjusted to reduce unnecessary inventory backlog and premature equipment replacement, optimize resource allocation, and reduce operating costs.

[0102] It is understood that the above scenarios are merely examples and do not constitute a limitation on the application scenarios of the technical solutions provided in the embodiments of this application. The technical solutions of this application can also be applied to other scenarios. For example, as those skilled in the art will know, with the evolution of system architecture and the emergence of new business scenarios, the technical solutions provided in the embodiments of this application are also applicable to similar technical problems.

[0103] Further, please see Figure 3 , Figure 3 This is a flowchart illustrating a camera state prediction method provided in an embodiment of this application. The method can be implemented using a computer device (e.g., the one described above). Figure 1 The server 100 in the corresponding embodiment executes the method. The following description uses the example of the method being executed by a computer device as an example. The camera state prediction method may include at least the following steps S101-S103:

[0104] Step S101: Obtain the core data of the permission verification camera; the permission verification camera refers to a camera used to capture images of the permission verification area of ​​the first object; the permission verification area image is used to perform business permission verification on the first object; the core data of the camera includes the permission verification area image, camera working data, and core device usage data.

[0105] Specifically, business permission verification can include various business permission verifications such as payment permission verification and login permission verification. The permission verification area image can be an image of a part representing the biometric information of the first object, such as a face image, palm image, fingerprint image, iris image, etc. The permission verification camera is a data acquisition device or component used to acquire the permission verification area image of the first object during the business permission verification process.

[0106] Specifically, the core data of the authorization verification camera can include images of the authorization verification area, camera operating data, and core component usage data. The number and type of authorization verification area images can be one or more, and can be collected based on different verification rules. For example, in a palm payment scenario, the authorization verification area images will include color images and infrared images. Color images refer to color palmprint images, and infrared images refer to infrared palm vein images. Camera operating data refers to data collected by the authorization verification camera during operation, specifically including CPU temperature, laser temperature, average brightness of the color image, average brightness of the infrared image, original frame rate, and aligned frame rate. Core component usage data refers to the usage data of the authorization verification camera's core components, specifically including the number of times and duration of laser usage. It is understandable that the obtained permission verification area image and camera working data can be data corresponding to the permission verification camera's most recent use. For example, in a palm payment scenario, if the permission verification camera was last used to verify user A's payment permissions, then when obtaining the core camera data, the obtained permission verification area image could include user A's color palm print and infrared palm vein image, and the obtained camera working data would be the camera's working data when collecting user A's area image. Core device usage data refers to the historical cumulative usage data of the core device since the permission verification camera began use. Optionally, the obtained permission verification area image and camera working data can also be multiple sets of data generated by the permission verification camera within a recent period of use (the most recent ten minutes, the most recent day). When subsequently evaluating the permission verification area image and camera working data, each set of data can be evaluated separately and then averaged to obtain the area image quality parameters corresponding to the permission verification area image and the working performance parameters corresponding to the camera working data. For ease of understanding, this application will only describe the acquisition of one set of permission verification area image and camera working data as an example in the following process.

[0107] Step S102: Perform camera element evaluation on the core data of the camera to obtain the image quality parameters of the area corresponding to the permission verification area image, the working performance parameters corresponding to the camera working data, and the core device performance parameters corresponding to the core device usage data.

[0108] Specifically, camera elements refer to factors that reflect the availability of a camera. These elements can be categorized into location image quality parameters, operational performance parameters, and core component performance parameters. Location image quality parameters are scores reflecting the image quality of the access control area; therefore, evaluating location image quality parameters involves scoring the image quality of the access control area. Operational performance parameters are scores reflecting the performance of the access control camera during operation; therefore, evaluating operational performance parameters involves scoring the current operational performance of the camera based on its operational data. Core component performance parameters are scores reflecting the performance of the core components; therefore, evaluating core component performance parameters involves scoring the performance of the core components based on their usage data. It is understood that the camera element evaluation method can be chosen according to the actual situation. Taking the evaluation of location image quality parameters as an example, location image quality parameters can be determined through image quality scoring algorithms or by outputting location image quality parameters through a trained image quality evaluation model; this application does not impose any restrictions on this.

[0109] Step S103: Based on the image quality parameters of the location, the working performance parameters, the core device performance parameters, and the camera element parameter weights, perform availability status prediction processing on the permission verification camera to obtain the predicted availability status corresponding to the permission verification camera; the predicted availability status is used to evaluate the availability of the permission verification camera.

[0110] Specifically, the predicted availability status can include a predicted availability score, which can be used to characterize the predicted remaining available lifespan of the access control camera. For example, the predicted availability score can be set to 0-100. If the total available lifespan of the access control camera is 2 years, then when the predicted availability score is 0, the remaining available lifespan of the access control camera is predicted to be 0. When the predicted availability score is 50, the remaining available lifespan of the access control camera is predicted to be 1 year. In one feasible embodiment, a possible implementation process for predicting the availability status of an access verification camera based on location image quality parameters, operational performance parameters, core component performance parameters, and camera element parameter weights can be as follows: A camera element fusion feature vector is generated based on the location image quality parameters, operational performance parameters, core component performance parameters, and camera element parameter weights; a camera availability status prediction model is obtained, which may contain N camera availability status decision trees; different camera availability status decision trees are used to represent different camera availability status decision strategies; the N camera availability status decision trees are traversed, and the camera element fusion feature vector is processed through the i-th camera availability status decision tree to obtain the i-th camera availability status decision score; when the traversal of the N camera availability status decision trees is completed, the obtained N camera availability status decision scores are weighted and summed to obtain the predicted availability score corresponding to the access verification camera.

[0111] One feasible implementation process for generating a camera element fusion feature vector based on the location image quality parameters, working performance parameters, core component performance parameters, and camera element parameter weights can be as follows: First, the location image quality parameters, working performance parameters, and core component performance parameters are standardized to obtain standardized location image quality parameters, standardized working performance parameters, and standardized core component performance parameters. Then, feature extraction is performed on the standardized location image quality parameters, standardized working performance parameters, and standardized core component performance parameters to obtain location image quality feature vectors, working performance feature vectors, and core component performance feature vectors. Finally, the location image quality feature vectors, working performance feature vectors, and core component performance feature vectors are weighted and fused according to the camera element parameter weights to obtain the camera element fusion feature vector. Parameter standardization refers to standardizing each parameter to the same scoring system. For example, if the location image quality parameter corresponds to a 10-point scale (i.e., values ​​between [0, 10]), and assuming the standard scoring system is a percentage scale, and the location image quality parameter is 8, then after parameter standardization, the obtained standardized location image quality parameter is 80. It is understandable that the obtained parameters are media data, lacking information that computer devices can understand. Therefore, it is necessary to transform the parameters from an unstructured raw image into structured information that computers can recognize and process. This involves scientifically abstracting the parameters and establishing a mathematical model to describe and replace the camera fusion element parameters, enabling computer devices to recognize these parameters through calculations and operations on this mathematical model. The mathematical model can be a vector space model, extracting features from the parameters, that is, converting the parameters into feature vectors. The camera element parameter weights will include the weight corresponding to each parameter, representing the degree of influence of each parameter on the camera's lifespan or the importance of each parameter. The weighted fusion process can be either weighting each feature vector and then adding the vectors, or weighting and then concatenating the vectors, depending on the model settings. This application does not impose any restrictions here. For example, weighted fusion processing involves concatenating weighted vectors. Assuming the weight of the part image quality feature vector X1 is w1, the weight of the working performance feature vector X2 is w2, and the weight of the core component performance feature vector X3 is w3, then the camera element fusion feature vector X = [w1X1, w2X2, w3X3].

[0112] The camera availability prediction model can be a pre-trained XGBoost (eXtremeGradientBoosting) model, which can contain N pre-trained camera availability decision trees. Each camera availability decision tree is used to judge the feature data of the permission verification camera.

[0113] For ease of understanding, please refer to the following: Figure 4 , Figure 4 This is a schematic diagram of a camera availability status decision tree provided in an embodiment of this application. Figure 4 As shown, the camera availability state prediction model can include a camera availability state decision tree 4, which can contain three leaf nodes. Each leaf node corresponds to a prediction score. For example, leaf node 43 corresponds to a prediction score of 2, leaf node 44 corresponds to a prediction score of 10, and leaf node 45 corresponds to a prediction score of 20. The camera availability state decision tree 4 can also contain decision nodes that determine the branch to follow. Each decision node corresponds to a decision strategy, which is used to determine which branch to follow based on the relevant data in the input feature vector. For example, decision node 41 can be used to determine whether feature A (such as the CPU temperature score when the camera is working, which can be reflected by some data in the camera element fusion feature vector X) is greater than a certain threshold (such as 80 points). If it is, then follow decision node 42; otherwise, follow leaf node 43. Since leaf node 43 is the bottom node, the camera availability state decision score of the camera availability state decision tree 4 for the camera element fusion feature vector X can be determined to be the prediction score corresponding to leaf node 43, which is 2.

[0114] The process involves weighted summation of the available state decision scores for the N cameras to obtain the predicted available score for the access verification camera. This means adjusting the available state decision scores of the N cameras according to their respective weights, and then summing them to arrive at the predicted available score. It's understandable that the weights corresponding to the available state decision scores of the N cameras can be determined during model training and are related to the performance and importance of each decision tree.

[0115] Optionally, in a feasible embodiment, the process of training the camera availability prediction model can be as follows: Collect core camera data from cameras at different stages of use, and simultaneously determine the camera's availability status (e.g., predicted availability score) as the corresponding label. Then, divide the collected cameras into a training set, a validation set, and a test set, with a ratio of 7:2:1 or 8:1:1, which is not limited herein. The training set is used to train the model, the validation set is used to adjust the model's hyperparameters, and the test set is used to evaluate the final model's performance. Then, obtain an initial camera availability prediction model, define training model parameters such as max_depth (maximum tree depth), learning_rate (learning rate), and n_estimators (number of trees), and then use the defined training model parameters and the training set to train the initial camera availability prediction model, resulting in a trained camera availability prediction model.

[0116] Then, the performance of the trained camera availability prediction model on unseen data can be evaluated using a validation set, and the hyperparameters of the trained model can be adjusted to obtain the validation model. For example, the model can be judged as overfitting or underfitting by observing the loss function (such as mean squared error, if it is a numerical target such as lifespan prediction) or other evaluation metrics (such as mean absolute error, MAE) on the validation set.

[0117] Finally, the performance of the camera availability prediction model can be evaluated using a test set. If the performance evaluation passes, the trained camera availability prediction model is confirmed as the official camera availability prediction model. Performance evaluation of the trained model can include calculating error metrics and model validation. Error metrics can include mean squared error loss (MSE) and mean absolute error loss (MAE). Model validation can utilize cross-validation to verify the model's stability and generalization ability. MSE is the average of the squared differences between the predicted and true values; a smaller MSE value indicates more accurate predictions. MAE represents the average absolute error between the predicted and true values.

[0118] Optionally, the predicted availability status includes a predicted availability score; based on the predicted availability score and the total available time of the cameras, the predicted remaining usage time for the access control cameras is determined; if the predicted remaining usage time is less than the remaining usage time threshold, a maintenance notification is sent to the camera management terminal; the maintenance notification is used to instruct the access control cameras to undergo maintenance. For example, if the predicted availability score ranges from [0, 100], and the predicted availability score is 5, and the total available time of the cameras is 2 years, then the predicted remaining usage time is 2 * 0.05 years = 0.1 years. A remaining usage time threshold of 0.2 years can be specified. For access control cameras whose predicted remaining usage time is less than the remaining usage time threshold, a maintenance plan can be developed to arrange equipment repair and replacement in advance, ensuring the normal operation of the equipment.

[0119] The method provided in this application embodiment can dynamically evaluate the core camera data generated by the permission verification camera during actual use, obtain multiple camera element parameters that reflect the actual use of the permission verification camera, and then predict the availability status of the permission verification camera based on the multiple camera element parameters and the influence weight of each camera element parameter on the availability status of the permission verification camera, thereby improving the accuracy of predicting the availability status of the permission verification camera.

[0120] Further, please see Figure 5 , Figure 5 This is a flowchart illustrating a camera state prediction method provided in an embodiment of this application. The method can be implemented using a computer device (e.g., the one described above). Figure 1 The server 100 in the corresponding embodiment executes the method. The following description uses the example of the method being executed by a computer device as an example. The camera state prediction method may include at least the following steps S201-S205:

[0121] Step S201: Obtain the core data of the permission verification camera; the permission verification camera refers to a camera used to capture images of the permission verification area of ​​the first object; the permission verification area image is used to perform business permission verification on the first object; the core data of the camera includes the permission verification area image, camera working data, and core device usage data.

[0122] Specifically, the implementation process of step S201 can be found above. Figure 3 The description of step S101 in the corresponding embodiment will not be repeated here.

[0123] Step S202: Perform image quality assessment on the permission verification area image to obtain the area image quality parameters corresponding to the permission verification area image.

[0124] Specifically, the image quality parameters for the authorized verification area can include traditional image quality parameters. A feasible implementation process for evaluating the image quality of the authorized verification area to obtain the corresponding image quality parameters can be as follows: perform image edge detection on the authorized verification area to obtain the image edges; perform image contrast calculation on the authorized verification area to obtain the image contrast; perform brightness distribution detection on the authorized verification area to obtain the image brightness distribution; perform image sharpness calculation on the authorized verification area to obtain the image sharpness; and determine the traditional image quality parameters corresponding to the authorized verification area based on the image edges, image contrast, image brightness distribution, and image sharpness.

[0125] Image edge detection can utilize either the Sobel algorithm (a classic image processing method) or the Canny algorithm (a multi-stage edge detection algorithm). Sobel edge detection detects edges in an image by calculating the gradient of each pixel to determine the edge's location and direction. Canny edge detection, a multi-stage algorithm widely used in computer vision, accurately detects edges in an image through a series of steps. A feasible image edge detection process can be as follows: First, convert the image of the authorization verification area into a grayscale image. Then, use the Sobel operator (small matrices in both horizontal and vertical directions) to convolve with each pixel in the image, calculating the gradient values ​​in the horizontal and vertical directions. Next, calculate the edge intensity of each pixel using a formula (such as an approximate calculation). Finally, non-maximum suppression and double thresholding can be applied to refine and connect the edges, thus obtaining the edges of the authorization verification area image.

[0126] Image contrast calculation can utilize contrast calculation methods, such as local contrast evaluation. Local contrast evaluation is an image quality assessment method primarily used to evaluate the impact of contrast variations in local areas of an image on the overall visual effect. Local contrast evaluation typically measures image quality by calculating the contrast differences between different regions of the image. A feasible local contrast evaluation process can be as follows: First, divide the authorization verification area image into multiple small regions, such as square or rectangular regions; then calculate the region contrast, that is, calculate the contrast of each small region. A simple way is to measure the contrast by calculating the difference between the maximum and minimum pixel values ​​within the region; finally, combine the contrasts of all small regions to evaluate the contrast of the entire authorization verification area image. For example, the average contrast of all small regions can be calculated, or a weighted average can be used based on the importance of different regions.

[0127] Brightness distribution detection can be performed using histogram equalization. Histogram equalization is an image processing algorithm whose main purpose is to enhance image contrast. An image's histogram shows the distribution of pixel values. For an image with low contrast, its pixel values ​​may be concentrated in a small range. The algorithm first counts the frequency of each pixel value (0-255 for grayscale images) to obtain the original histogram. Then, a cumulative distribution function is used to calculate the mapping value for each pixel value, making the pixel value distribution more uniform. Finally, the original pixel values ​​are transformed according to the mapping relationship to obtain the image with enhanced contrast. For example, in a very dark image where most pixel values ​​are concentrated in the low-brightness range, histogram equalization will stretch the pixel values, making details in both dark and bright areas more apparent.

[0128] Image sharpness calculation can utilize Laplace transform or other high-frequency component analysis methods. Similarly, the image of the authorization verification area is first converted to grayscale. Then, the Laplace operator is used to perform convolution operations on each pixel in the image to obtain the Laplace-transformed image. Finally, sharpness metrics are calculated, such as the variance of the Laplace-transformed image. A larger variance indicates more dramatic changes in grayscale values, richer edges and details, and thus a sharper image.

[0129] One feasible implementation of determining the traditional image quality parameters corresponding to the permission verification part image based on the part image edge, part image contrast, part image brightness distribution, and part image sharpness can be as follows: scoring the part image edge, part image contrast, part image brightness distribution, and part image sharpness separately to obtain part image edge score, part image contrast score, part image brightness distribution score, and part image sharpness score; and then performing a weighted summation of the part image edge score, part image contrast score, part image brightness distribution score, and part image sharpness score to obtain the traditional image quality parameters corresponding to the permission verification part image. When scoring the edges of a part of the image, the number and sharpness of the edges in the image can be calculated. The more edges there are and the sharper they are, the higher the edge score of the corresponding part of the image. When scoring the contrast of a part of the image, the higher the contrast, the better the image sharpness, and the higher the contrast score of the corresponding part of the image. When scoring the brightness distribution of a part of the image, the uniformity and appropriateness of the brightness distribution can be evaluated. Too bright or too dark will reduce the brightness distribution score of the part of the image. When scoring the sharpness of a part of the image, the more high-frequency components there are, the richer the details of the image, and the higher the sharpness score of the corresponding part of the image.

[0130] Specifically, the image quality parameters for the authorized verification location can also include intelligent image quality parameters. The authorized verification location image can include both color and infrared images. A feasible implementation process for evaluating the image quality of the authorized verification location image to obtain its corresponding image quality parameters can be as follows: Preprocess the color image to obtain a standardized color image; preprocess the infrared image to obtain a standardized infrared image; evaluate the texture clarity of the standardized color image to obtain texture clarity parameters; evaluate the vein clarity of the standardized infrared image to obtain vein clarity parameters; and perform a weighted sum of the texture clarity parameters and vein clarity parameters to obtain the intelligent image quality parameters corresponding to the authorized verification location image. Here, the color image is a color image, and the infrared image is an infrared image. Their uses differ. For example, in palm print detection, the color image is generally a color palm print image, used to evaluate palm print clarity, while the infrared image is generally an infrared vein image, used to evaluate vein clarity.

[0131] One feasible implementation of image preprocessing to obtain a standardized color region image from a color region image can be as follows: Normalize the color region image to obtain a normalized color region image; denoise the normalized color region image to obtain a denoised color region image; and resize the denoised color region image to a standardized image size to obtain a standardized color region image. Image normalization refers to adjusting image brightness and contrast to ensure image data consistency. Denoising can use median filtering or other denoising algorithms to reduce image noise. The standardized image size can be a pre-defined size for easy model processing.

[0132] One feasible implementation process for evaluating the texture sharpness of standardized color area images and obtaining texture sharpness parameters can be as follows: First, feature extraction is performed on the standardized color area images to obtain image texture features. Then, the image texture features are input into a texture sharpness model, and the texture sharpness parameters are output through the texture sharpness model. For ease of understanding, the evaluation of palm print sharpness in a color palm print image is used as an example. During feature extraction, convolutional neural networks (CNNs) can be used to extract palm print features. Pre-trained models such as ResNet-50 (a deep residual network) and Inception (a deep learning model) can be used for palm feature extraction. Then, intermediate layer features are extracted through feature layers to capture palm print details. Finally, the extracted texture features are input into a pre-trained palm print sharpness model for palm print sharpness evaluation. The palmprint clarity model can use deep learning models, such as DenseNet (densely connected network) and EfficientNet (an efficient and scalable convolutional neural network architecture with automatic model scaling). The training data for training the palmprint clarity model can be a large number of labeled palmprint images, with the labels including clarity scores. The training process can use supervised learning methods to optimize the model parameters so that it can accurately evaluate palmprint clarity.

[0133] One feasible implementation process for evaluating vein sharpness in standardized infrared region images and obtaining vein sharpness parameters can be as follows: First, feature extraction is performed on the standardized infrared region images to obtain vein features. Then, these vein features are input into a vein sharpness model, which outputs vein sharpness parameters. For ease of understanding, the evaluation of vein sharpness in infrared vein images is used as an example. When extracting vein features, infrared image feature extraction methods can be used. Pre-trained models such as ResNet-50 (deep residual network) and VGG (Visual Geometry Group, convolutional neural network) can be used to fine-tune the infrared image, and then high-level features are extracted through feature layers to capture the vein structure. Finally, the extracted vein features are input into the pre-trained vein sharpness model for vein sharpness evaluation. The vein clarity model can also use deep learning models, such as DenseNet (densely connected network) and EfficientNet (an efficient and scalable convolutional neural network architecture with automatic model scaling). The training data for training the vein clarity model can be a large number of labeled vein images, with the labels including vein clarity scores. The training process can use supervised learning methods to optimize the model parameters so that it can accurately evaluate vein clarity.

[0134] One feasible implementation of the above-mentioned weighted summation of the texture clarity parameter and vein clarity parameter to obtain the intelligent image quality parameter corresponding to the permission verification area image can be as follows: First, standardize the texture clarity parameter and vein clarity parameter to ensure parameter consistency. Then, obtain the texture clarity weight corresponding to the texture clarity parameter and the vein clarity weight corresponding to the vein clarity parameter. The intelligent image quality parameter is then calculated as: Intelligent Image Quality Parameter = Texture Clarity Parameter × Texture Clarity Weight + Vein Clarity Parameter × Vein Clarity Weight. The vein clarity weight and texture clarity weight can be set according to the influence of palm print and vein clarity on the overall image quality, and their sum should be 1. For example, if the vein clarity weight is 0.4, then the texture clarity weight is 0.6.

[0135] Step S203: Evaluate the working performance of the camera working data to obtain the working performance parameters corresponding to the camera working data.

[0136] Specifically, camera operating data includes device temperature data; operating performance parameters include device temperature performance parameters. A feasible implementation process for evaluating the operating performance of camera operating data to obtain the corresponding operating performance parameters can be as follows: If the device temperature data is less than or equal to the standard temperature data, then the standard temperature performance parameter is determined as the device temperature performance parameter; if the device temperature data is greater than the standard temperature data, then the temperature difference between the device temperature data and the standard temperature data is determined, and the loss temperature performance parameter is determined based on the temperature difference; the difference between the standard temperature performance parameter and the loss temperature performance parameter is determined as the device temperature performance parameter. The device temperature data can include CPU temperature and laser temperature. For example, when scoring the CPU temperature, it can be considered that when the CPU temperature is less than temperature 'a', the CPU temperature score (i.e., the device temperature performance parameter) is 100. When the CPU temperature is greater than temperature 'a', the greater the increase, the lower the score. For example, when the CPU temperature increases by 1 degree, the score decreases by 5. Therefore, when the CPU temperature is (a+2) degrees, the CPU temperature score can be determined to be 90, and the lowest possible CPU temperature score is 0. The laser temperature scoring follows the same principle.

[0137] Specifically, camera operating data includes image brightness data; operating performance parameters include image brightness performance parameters. A feasible implementation process for evaluating the operating performance of camera operating data to obtain the corresponding operating performance parameters can be as follows: determine the deviation amplitude between image brightness data and standard brightness data; if the deviation amplitude is less than or equal to a deviation amplitude threshold, then the standard brightness performance parameter is determined as the image brightness performance parameter; if the deviation amplitude is greater than the deviation amplitude threshold, then the amplitude difference between the deviation amplitude and the deviation amplitude threshold is determined, and the loss brightness performance parameter is determined based on the amplitude difference; the difference between the standard brightness performance parameter and the loss brightness performance parameter is determined as the image brightness performance parameter. The image brightness data can include the average brightness of the color image and the average brightness of the infrared image. That is, the further the average brightness of the color image deviates from the average brightness of the standard color image (whether larger or smaller), the lower the final average brightness score of the color image, and the same applies to the average brightness of the infrared image.

[0138] Specifically, camera operating data includes device frame rate data; operating performance parameters include device frame rate performance parameters. A feasible implementation process for evaluating the operating performance of camera operating data to obtain the corresponding operating performance parameters can be as follows: If the device frame rate data is greater than or equal to the standard frame rate data, then the standard frame rate performance parameter is determined as the device frame rate performance parameter; if the device frame rate data is less than the standard frame rate data, then the frame rate difference between the standard frame rate data and the device frame rate data is determined, and the lost frame rate performance parameter is determined based on the frame rate difference; the difference between the standard frame rate performance parameter and the lost frame rate performance parameter is determined as the device frame rate performance parameter. The device frame rate data can include the original frame rate and the aligned frame rate. That is, the lower the frame rate, the lower the final score.

[0139] Step S204: Perform core device performance evaluation on the core device usage data to obtain the core device performance parameters corresponding to the core device usage data.

[0140] Specifically, core component usage data includes the number of times the core component is used. A feasible implementation process for evaluating the core component performance based on the core component usage data to obtain the corresponding core component performance parameters can be as follows: determine the used time of the core component based on the number of times the core component is used and the duration of each single use; obtain the total available time of the core component; determine the time ratio between the used time and the total available time; and determine the core component performance parameters corresponding to the core component usage data based on the time ratio. For example, the lifespan of a laser is 10,000 hours of continuous operation (i.e., the total available time of the core component). However, in reality, the laser is only turned on when swiping the palm. Assuming the single use time of the laser is 10 seconds, when the laser has been used 10,000 times, the actual usage time of the laser can be calculated as: 10,000 x 10 / 60 / 60 = approximately 28 hours. At this time, the performance parameter of the laser can be evaluated as (10,000 - 28) / 10,000.

[0141] Step S205: Based on the image quality parameters of the location, the working performance parameters, the core component performance parameters, and the camera element parameter weights, perform availability status prediction processing on the permission verification camera to obtain the predicted availability status corresponding to the permission verification camera; the predicted availability status is used to evaluate the availability of the permission verification camera.

[0142] Specifically, the implementation process of step S205 can be found in the detailed description of step S103 above, and will not be repeated here.

[0143] The method provided in this application allows for accurate evaluation of core camera elements, thereby improving the accuracy of the predicted availability status of the permission verification camera. By predicting the camera's lifespan in advance, preventative maintenance can be performed before the camera completely fails, avoiding the impact of business interruption on user experience. Furthermore, it enables the rapid identification and replacement of cameras that are about to fail, reducing system downtime and ensuring business continuity.

[0144] Further, please see Figure 6 , Figure 6 This is an overall flowchart of a method for predicting the availability status of a palm-scanning camera, provided in an embodiment of this application. The palm-scanning camera refers to a camera used in payment transactions to capture images of a user's palm to verify whether the user has palm-scanning authorization. Figure 6 As shown, the entire process includes the following steps S31 to S34.

[0145] Step S31: Collect sample camera data and sample camera core data.

[0146] Specifically, the collected sample cameras can include the following different stages of palm-swiping cameras: newly used, used for six months, used for one year, used for two years, used for three years, partially damaged, and completely damaged.

[0147] Specifically, the core data of the camera may include image data (i.e., the image of the permission verification area mentioned above), working data (i.e., the working data of the camera mentioned above), and core device usage data (i.e., the core device usage data mentioned above). The image data may specifically include color image palm print and infrared image palm vein image data; the working data may include CPU temperature, laser temperature, average brightness of the color image, average brightness of the infrared image, original frame rate, and frame rate after alignment; the core device usage data may specifically include the number of times and duration of laser usage.

[0148] Specifically, in order to ensure the normal execution of subsequent steps and avoid the randomness of data, the number of palm-scanning cameras collected at each stage should be multiple, for example, at least 20; the amount of core data collected should also be multiple days' worth of data, for example, at least 20 days' worth of data.

[0149] Step S32: Divide the core elements of the camera and set the weight of each core element of the camera.

[0150] Specifically, the core elements of a camera can be divided into image quality score (i.e., the image quality parameters of the above-mentioned parts), working data score (i.e., the working performance parameters of the above-mentioned parts), and core component usage data score (i.e., the performance parameters of the above-mentioned core components).

[0151] Specifically, image quality scoring refers to evaluating the quality of color palm print and infrared palm vein images using both traditional and AI (artificial intelligence) image quality scoring algorithms. This results in a traditional image quality algorithm score (the aforementioned traditional image quality parameters) and an AI image quality algorithm score (the aforementioned intelligent image quality parameters). The traditional image quality algorithm score includes ratings for image edges, contrast, brightness distribution, and sharpness index. Edge detection can use Sobel or Canny edge detection algorithms. When scoring edges, the number and sharpness of edges in the image are calculated; more and sharper edges result in a higher score. Contrast evaluation can be determined using contrast calculation methods, such as local contrast evaluation. Higher contrast generally indicates better image sharpness, leading to a higher score. Brightness distribution can be determined using histogram equalization. When scoring brightness, the uniformity and appropriateness of the brightness distribution are assessed; excessive brightness or darkness lowers the score. The sharpness index can be achieved using Laplace transform or other high-frequency component analysis methods. More high-frequency components indicate richer image detail, resulting in a higher score. The AI ​​image quality algorithm score includes rating the clarity of palmprint and vein images. First, the image data undergoes preprocessing. Then, palmprint features are extracted, and the palmprint clarity score is estimated based on these features. Similarly, vein features are extracted, and the vein clarity score is estimated. Finally, both the palmprint and vein clarity scores are standardized to ensure consistency. A weighted average is then calculated, assigning weights based on the impact of palmprint and vein clarity on overall image quality. For example, if the palmprint clarity weight is 0.6 and the vein clarity weight is 0.4, the final comprehensive AI image quality algorithm score is calculated as: Palmprint Clarity Score * 0.6 + Vein Clarity Score * 0.4.

[0152] Specifically, the performance data scoring refers to the scoring of "CPU temperature, laser temperature, average brightness of the color image, average brightness of the infrared image, original frame rate, and aligned frame rate". The higher the CPU temperature and laser temperature, the lower the score; the further the average brightness of the color image and infrared image deviates from the standard brightness, the worse the score; and for the original frame rate and aligned frame rate, the lower the frame rate, the lower the score.

[0153] Specifically, the core component usage data scoring refers to scoring the "number of times and duration of laser usage." The longer the number of times and duration of laser usage, the lower the corresponding score.

[0154] Specifically, based on the scoring method described above, the camera element scores corresponding to the sample cameras obtained in step S31 can be determined. Then, based on these camera element scores, the availability status of the sample cameras, and actual empirical values, the weight of each camera element can be set. For example, the image quality score weight is 30%, the working data score weight is 20%, and the core component usage data score weight is 50%. In a feasible embodiment, multiple sets of camera element weights can be set, and then the availability status of the sample cameras can be predicted according to each set of camera element weights, the camera element scores of the sample cameras, and the availability status of the sample cameras. The set of camera element weights with the smallest error between the predicted availability status and the actual availability status is then used as the set camera element weights.

[0155] Step S33: Collect the core camera data of the working palm-mounted camera to complete the lifespan prediction logic chain.

[0156] Specifically, after determining the weight of each core element of the camera, image quality, working data, and core component usage data can be continuously collected from sample cameras and online devices. Then, based on the collected core camera data, a lifespan prediction logic chain can be completed.

[0157] Specifically, the lifespan prediction logic chain can include data collection and preprocessing, feature extraction engineering, lifespan prediction model training, model evaluation and validation, real-time prediction and maintenance plan development, and model updating and optimization.

[0158] Specifically, the data collection and preprocessing process can include data collection, data cleaning, and data standardization. Data collection refers to continuously collecting image quality, operational data, and core component usage data from sample cameras and online devices. Data cleaning involves removing outliers and noise to ensure data accuracy and consistency. Data standardization involves standardizing different types of data to the same scale to facilitate subsequent analysis.

[0159] Specifically, the feature extraction process includes extracting image quality features, extracting working data features, extracting core component usage features, and calculating scores. Extracting image quality features refers to extracting image quality scores (including palmprint clarity and vein clarity). Extracting working data features refers to extracting working data scores (CPU temperature, laser temperature, brightness, frame rate, etc.). Extracting core component usage features refers to extracting core component usage data scores (laser usage frequency and duration). Scoring calculation involves weighting the scores of each camera's core elements according to their respective weights, and then summing the results.

[0160] Specifically, life expectancy prediction model training includes training set preparation, model selection, and model training. Training set preparation refers to preparing training and test sets using collected data. Model selection refers to choosing a suitable machine learning model (e.g., XGBoost) for training. Model training refers to using the training set to train the life expectancy prediction model and optimize its parameters.

[0161] Specifically, model evaluation and validation includes model evaluation and model validation. Model evaluation refers to evaluating the model's performance using a test set and calculating error metrics (such as MSE, MAE, etc.). Model validation refers to verifying the model's stability and generalization ability through methods such as cross-validation.

[0162] Specifically, real-time prediction and maintenance planning includes real-time data input and maintenance plan development. Real-time data input refers to inputting real-time data from online equipment into the lifespan prediction model for real-time lifespan prediction. Maintenance plan development involves creating a maintenance schedule based on the prediction results, proactively arranging equipment maintenance and replacement, and ensuring the normal operation of the equipment.

[0163] Specifically, model updates and optimizations include continuous learning and optimization strategies. Continuous learning refers to periodically updating the model and retraining it with the latest data to improve model performance. Optimization strategies refer to continuously optimizing weight allocation and model parameters based on feedback from real-world applications to improve prediction accuracy.

[0164] Step S34: Perform lifespan prediction on the working palm-mounted camera and formulate a maintenance plan based on the lifespan prediction results.

[0165] Specifically, lifespan prediction refers to inputting the core data of the working palm-swiping camera into the lifespan prediction model trained in step S33 to calculate the estimated remaining lifespan of the working palm-swiping camera. For example, if the lifespan prediction model outputs a score for the usability of the working palm-swiping camera, this score can be converted into specific time or usage counts to predict the remaining lifespan of the device. Based on the lifespan prediction results, the health status of the camera is assessed to determine whether it needs maintenance or replacement.

[0166] Specifically, maintenance plans can be triggered by setting trigger thresholds. For example, based on the predicted remaining lifespan of the equipment, a maintenance trigger threshold can be set (e.g., if the remaining lifespan is less than a certain time or number of uses). When the predicted result is below the threshold, the maintenance plan is triggered. Maintenance plans can include preventative maintenance plans and corrective maintenance plans. Preventative maintenance plans involve scheduling regular inspections and maintenance of the equipment in advance based on the predicted results to avoid sudden malfunctions during use. Corrective maintenance plans involve developing detailed repair or replacement plans for equipment that is partially or completely damaged.

[0167] Optionally, all maintenance operations can be recorded, including maintenance time, maintenance content, and maintenance results, serving as the data basis for subsequent analysis and model optimization. Based on maintenance records and actual usage, the life prediction model and maintenance plan can be continuously optimized to improve prediction accuracy and maintenance efficiency.

[0168] The method for predicting the availability of a palm-scanning camera provided in this application can improve maintenance efficiency. By predicting the camera's lifespan in advance, preventative maintenance can be performed before the camera completely fails, avoiding the impact of business interruption on user experience and reducing system downtime to ensure business continuity. It can also enhance user experience by ensuring the camera is always in optimal working condition through timely maintenance and replacement, reducing payment failures or delays caused by camera malfunctions, thereby increasing the success rate and satisfaction of users when using palm-scanning and facial recognition payments. Furthermore, it can optimize inventory management by rationally allocating spare parts inventory based on the camera's lifespan prediction results, avoiding excessive or insufficient inventory, optimizing resource allocation, and reducing operating costs.

[0169] Please see Figure 7 , Figure 7 This is a schematic diagram of the structure of a camera state prediction device provided in an embodiment of this application. The camera state prediction device can be a computer program (including program code) running on a computer device; for example, the camera state prediction device is an application software. The camera state prediction device 1 can be used to execute corresponding steps in the camera state prediction method provided in the embodiments of this application. Figure 7 As shown, the camera state prediction device 1 may include: a data acquisition module 110, an element evaluation module 120, and a state prediction module 130.

[0170] The data acquisition module 110 is used to acquire the core data of the permission verification camera; the permission verification camera refers to the camera used to collect images of the permission verification area of ​​the first object; the permission verification area images are used to verify the business permissions of the first object; the core data of the camera includes the permission verification area images, camera working data, and core component usage data;

[0171] The element evaluation module 120 is used to evaluate the camera element of the camera core data to obtain the image quality parameters of the permission verification area, the working performance parameters of the camera working data, and the core device performance parameters of the core device usage data.

[0172] The status prediction module 130 is used to perform availability status prediction processing on the permission verification camera based on the image quality parameters of the location, the working performance parameters, the core component performance parameters, and the camera element parameter weights, to obtain the predicted availability status of the permission verification camera; the predicted availability status is used to evaluate the availability of the permission verification camera.

[0173] In one possible implementation, the element evaluation module 120 is used to perform camera element evaluation on the camera core data. When obtaining the image quality parameters corresponding to the permission verification area image, the working performance parameters corresponding to the camera working data, and the core device performance parameters corresponding to the core device usage data, it is specifically used to perform the following operations:

[0174] Image quality assessment is performed on the images of the authorization verification area to obtain the corresponding image quality parameters of the authorization verification area.

[0175] The camera's working data is used to evaluate its performance, and the corresponding performance parameters are obtained.

[0176] The performance of core components is evaluated based on the usage data, and the corresponding performance parameters of the core components are obtained.

[0177] In one possible implementation, the part image quality parameters include conventional image quality parameters;

[0178] The element evaluation module 120 is used to evaluate the image quality of the authorization verification area image. When obtaining the image quality parameters of the area corresponding to the authorization verification area image, it is specifically used to perform the following operations:

[0179] Perform edge detection on the image of the permission verification area to obtain the edge of the area image;

[0180] Perform image contrast calculation on the image of the authorization verification area to obtain the image contrast of the area;

[0181] Brightness distribution detection is performed on the image of the permission verification area to obtain the brightness distribution of the area image;

[0182] The image sharpness of the permission verification area is calculated to obtain the image sharpness of the area.

[0183] The traditional image quality parameters corresponding to the permission verification area image are determined based on the area image edge, area image contrast, area image brightness distribution, and area image clarity.

[0184] In one possible implementation, the permission verification area image includes a color area image and an infrared area image; the area image quality parameters include intelligent image quality parameters.

[0185] The element evaluation module 120 is used to evaluate the image quality of the authorization verification area image. When obtaining the image quality parameters of the area corresponding to the authorization verification area image, it is specifically used to perform the following operations:

[0186] Image preprocessing is performed on the color region image to obtain a standardized color region image;

[0187] Image preprocessing is performed on the infrared region image to obtain a standardized infrared region image;

[0188] The texture sharpness of standardized color area images is evaluated to obtain texture sharpness parameters;

[0189] The vein clarity was evaluated using standardized infrared images of specific areas to obtain vein clarity parameters.

[0190] The texture clarity parameter and the vein clarity parameter are weighted and summed to obtain the intelligent image quality parameter corresponding to the image of the authorization verification area.

[0191] In one possible implementation, when the feature evaluation module 120 performs image preprocessing on the color region image to obtain a standardized color region image, it specifically performs the following operations:

[0192] Image normalization is performed on the color part image to obtain a normalized color part image;

[0193] Denoising is performed on the normalized color region image to obtain a denoised color region image;

[0194] The image size of the denoised color region image is adjusted to a standardized image size to obtain a standardized color region image.

[0195] In one possible implementation, the camera's operating data includes device temperature data; the operating performance parameters include device temperature performance parameters.

[0196] The element evaluation module 120 is used to evaluate the working performance of the camera's working data. When obtaining the working performance parameters corresponding to the camera's working data, it is specifically used to perform the following operations:

[0197] If the device temperature data is less than or equal to the standard temperature data, then the standard temperature performance parameter shall be determined as the device temperature performance parameter.

[0198] If the device temperature data is greater than the standard temperature data, then determine the temperature data difference between the device temperature data and the standard temperature data, and determine the loss temperature performance parameters based on the temperature data difference;

[0199] The difference between the standard temperature performance parameter and the loss temperature performance parameter is determined as the device temperature performance parameter.

[0200] In one possible implementation, the camera's operating data includes image brightness data; the operating performance parameters include image brightness performance parameters.

[0201] The element evaluation module 120 is used to evaluate the working performance of the camera's working data. When obtaining the working performance parameters corresponding to the camera's working data, it is specifically used to perform the following operations:

[0202] Determine the magnitude of the deviation between the image brightness data and the standard brightness data;

[0203] If the deviation is less than or equal to the deviation threshold, the standard brightness performance parameter is determined as the image brightness performance parameter.

[0204] If the deviation amplitude is greater than the deviation amplitude threshold, then the amplitude difference between the deviation amplitude and the deviation amplitude threshold is determined, and the loss brightness performance parameter is determined based on the amplitude difference.

[0205] The difference between the standard brightness performance parameter and the lost brightness performance parameter is determined as the image brightness performance parameter.

[0206] In one possible implementation, the camera operating data includes device frame rate data; the operating performance parameters include device frame rate performance parameters.

[0207] The element evaluation module 120 is used to evaluate the working performance of the camera's working data. When obtaining the working performance parameters corresponding to the camera's working data, it is specifically used to perform the following operations:

[0208] If the device frame rate data is greater than or equal to the standard frame rate data, then the standard frame rate performance parameter is determined as the device frame rate performance parameter.

[0209] If the device frame rate data is less than the standard frame rate data, then the frame rate data difference between the standard frame rate data and the device frame rate data is determined, and the loss frame rate performance parameter is determined based on the frame rate data difference.

[0210] The difference between the standard frame rate performance parameter and the lossy frame rate performance parameter is determined as the device frame rate performance parameter.

[0211] In one possible implementation, the core device usage data includes the number of times the core device is used;

[0212] The element evaluation module 120 is used to evaluate the performance of core components based on the usage data of the core components. When obtaining the performance parameters of the core components corresponding to the usage data, it is specifically used to perform the following operations:

[0213] The usage time of the core component is determined based on the number of times the core component is used and the duration of a single use of the core component.

[0214] Obtain the total available time of core components, determine the ratio between the used time of core components and the total available time of core components, and determine the core component performance parameters corresponding to the core component usage data based on the ratio of the used time.

[0215] In one possible implementation, predicting the available state includes predicting the available score;

[0216] The status prediction module 130 is used to perform availability status prediction processing on the permission verification camera based on the image quality parameters, working performance parameters, core component performance parameters, and camera element parameter weights. When obtaining the predicted availability status of the permission verification camera, it is specifically used to perform the following operations:

[0217] A camera element fusion feature vector is generated based on the image quality parameters of the location, the working performance parameters, the performance parameters of the core components, and the weights of the camera element parameters.

[0218] Obtain a camera availability state prediction model; the camera availability state prediction model contains N camera availability state decision trees; different camera availability state decision trees are used to represent different camera availability state decision strategies; N is a positive integer;

[0219] Traverse the available state decision trees of N cameras to obtain the available state decision tree of the i-th camera; i is a positive integer less than N;

[0220] The camera availability status decision score is obtained by processing the camera element fusion feature vector through the availability status decision tree of the i-th camera.

[0221] When the decision tree of available states of N cameras is completed, the decision scores of available states of N cameras are weighted and summed to obtain the predicted available score of the permission verification camera.

[0222] In one possible implementation, predicting the available state includes predicting the available score;

[0223] The state prediction module 130 is also used to perform the following operations:

[0224] Based on the predicted available score and the total available time of the cameras, determine the predicted remaining usage time of the permission verification camera;

[0225] If the predicted remaining usage time is less than the remaining usage time threshold, a maintenance notification is sent to the camera management terminal; the maintenance notification is used to instruct the permission verification camera to undergo maintenance.

[0226] The device provided in this application embodiment can dynamically evaluate the core camera data generated by the permission verification camera during actual use, obtain multiple camera element parameters that reflect the actual use of the permission verification camera, and then predict the availability status of the permission verification camera based on the multiple camera element parameters and the influence weight of each camera element parameter on the availability status of the permission verification camera, thereby improving the accuracy of predicting the availability status of the permission verification camera.

[0227] Please see Figure 8 , Figure 8 This is a schematic diagram of the structure of a computer device provided in an embodiment of this application. Figure 8 As shown above, Figure 7 The camera state prediction device 1 in the corresponding embodiment can be applied to a computer device 1000, which may include a processor 1001, a network interface 1004, and a memory 1005. Furthermore, the computer device 1000 may also include a user interface 1003 and at least one communication bus 1002. The communication bus 1002 is used to enable communication between these components. The user interface 1003 may include a display screen and a keyboard; optionally, the user interface 1003 may also include a standard wired interface or a wireless interface. The network interface 1004 may optionally include a standard wired interface or a wireless interface (such as a Wi-Fi interface). The memory 1005 may be high-speed RAM or non-volatile memory, such as at least one disk storage device. Optionally, the memory 1005 may also be at least one storage device located remotely from the processor 1001. Figure 8 As shown, the memory 1005, which is a computer-readable storage medium, may include an operating system, a network communication module, a user interface module, and a device control application.

[0228] In such Figure 8 In the computer device 1000 shown, the network interface 1004 provides network communication elements; the user interface 1003 is mainly used to provide an input interface for the user; and the processor 1001 can be used to call the device control application stored in the memory 1005 to achieve:

[0229] Obtain the core data of the permission verification camera; the permission verification camera refers to the camera used to capture images of the permission verification area of ​​the first object; the permission verification area images are used to verify the business permissions of the first object; the core data of the camera includes the permission verification area images, camera working data, and core component usage data;

[0230] The camera's core data is evaluated to obtain the image quality parameters of the authorized verification area, the working performance parameters of the camera's working data, and the core device performance parameters of the core device usage data.

[0231] Based on the image quality parameters of the location, the operating performance parameters, the performance parameters of the core components, and the weighted parameters of the camera element, the access control camera is processed to predict its availability status, thus obtaining the predicted availability status of the access control camera. The predicted availability status is used to evaluate the availability of the access control camera.

[0232] It should be understood that the computer device 1000 described in the embodiments of this application can execute the foregoing text. Figure 3 , Figure 5 The description of the camera state prediction method in any corresponding embodiment will not be repeated here. Furthermore, the beneficial effects of using the same method will also not be repeated.

[0233] Furthermore, it should be noted that this application embodiment also provides a computer-readable storage medium, which stores a computer program executed by the aforementioned camera state prediction device 1. The computer program includes program instructions, and when the processor executes the program instructions, it can execute the aforementioned... Figure 3 , Figure 5 The description of the camera state prediction method in any corresponding embodiment is already provided, and therefore will not be repeated here. Furthermore, the beneficial effects of using the same method will also not be repeated. For technical details not disclosed in the computer-readable storage medium embodiments related to this application, please refer to the description of the method embodiments of this application.

[0234] The aforementioned computer-readable storage medium can be the camera status prediction device provided in any of the foregoing embodiments or the internal storage unit of the aforementioned computer device, such as the hard drive or memory of the computer device. The computer-readable storage medium can also be an external storage device of the computer device, such as a plug-in hard drive, smart media card (SMC), secure digital (SD) card, flash card, etc., provided on the computer device. Furthermore, the computer-readable storage medium can include both internal storage units and external storage devices of the computer device. The computer-readable storage medium is used to store the computer program and other programs and data required by the computer device. The computer-readable storage medium can also be used to temporarily store data that has been output or will be output.

[0235] Furthermore, it should be noted that this application also provides a computer program product or computer program, which includes computer instructions stored in a computer-readable storage medium. The processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the aforementioned... Figure 3 , Figure 5 The method provided in any of the corresponding embodiments.

[0236] The terms "first," "second," etc., in the specification, claims, and drawings of this application are used to distinguish different objects, not to describe a specific order. Furthermore, the term "comprising," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, apparatus, product, or device that includes a series of steps or units is not limited to the listed steps or modules, but may optionally include steps or modules not listed, or may optionally include other step units inherent to these processes, methods, apparatuses, products, or devices.

[0237] In this application embodiment, the terms "module" or "unit" refer to a computer program or part of a computer program that has a predetermined function and works with other related parts to achieve a predetermined goal, and can be implemented wholly or partially using software, hardware (such as processing circuitry or memory), or a combination thereof. Similarly, a processor (or multiple processors or memory) can be used to implement one or more modules or units. Furthermore, each module or unit can be part of an overall module or unit that includes the functionality of that module or unit.

[0238] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in the foregoing description as a network element. Whether these network elements are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can implement the described network elements using different methods for each specific application, but such implementation should not be considered beyond the scope of this application.

[0239] The above-disclosed embodiments are merely preferred embodiments of this application and should not be construed as limiting the scope of this application. Therefore, any equivalent variations made in accordance with the claims of this application shall still fall within the scope of this application.

Claims

1. A camera state prediction method, characterized by, include: Obtain permission to verify the camera's core data; The permission verification camera refers to a camera used to capture images of the permission verification area of ​​the first object; the permission verification area image is used to verify the business permissions of the first object; the core data of the camera includes the permission verification area image, camera working data, and core device usage data; The camera element evaluation is performed on the core data of the camera to obtain the image quality parameters of the area corresponding to the permission verification area image, the working performance parameters of the camera working data, and the core device performance parameters of the core device usage data. Based on the image quality parameters of the specified location, the working performance parameters, the performance parameters of the core components, and the weights of the camera element parameters, the usability status prediction processing of the access verification camera is performed to obtain the predicted usability status of the access verification camera. The predicted availability status is used to assess the availability of the permission verification camera.

2. The method according to claim 1, characterized in that, The step of evaluating camera element data to obtain image quality parameters corresponding to the permission verification area, performance parameters corresponding to the camera working data, and performance parameters corresponding to the core device usage data includes: The image quality of the permission verification area image is evaluated to obtain the area image quality parameters corresponding to the permission verification area image. The working performance of the camera is evaluated to obtain the working performance parameters corresponding to the camera working data. The core device performance is evaluated based on the usage data of the core device to obtain the core device performance parameters corresponding to the usage data.

3. The method according to claim 2, characterized in that, The image quality parameters for the affected area include traditional image quality parameters; The step of performing image quality evaluation on the image of the permission verification area to obtain the image quality parameters of the corresponding area includes: Perform image edge detection on the image of the permission verification area to obtain the edge of the area image; The image contrast of the permission verification area is calculated to obtain the area image contrast. Brightness distribution detection is performed on the image of the permission verification area to obtain the brightness distribution of the area image; The image sharpness of the permission verification area is calculated to obtain the image sharpness of the area. The traditional image quality parameters corresponding to the permission verification area image are determined based on the edge of the area image, the contrast of the area image, the brightness distribution of the area image, and the clarity of the area image.

4. The method according to claim 2, characterized in that, The permission verification area image includes a color area image and an infrared area image; the area image quality parameters include intelligent image quality parameters. The step of performing image quality evaluation on the image of the permission verification area to obtain the image quality parameters of the corresponding area includes: The color region image is preprocessed to obtain a standardized color region image; The infrared region image is preprocessed to obtain a standardized infrared region image; The texture sharpness of the standardized color area image is evaluated to obtain texture sharpness parameters; The vein clarity of the standardized infrared region image is evaluated to obtain vein clarity parameters; The texture clarity parameter and the vein clarity parameter are weighted and summed to obtain the intelligent image quality parameter corresponding to the image of the permission verification area.

5. The method according to claim 4, characterized in that, The step of preprocessing the colored region image to obtain a standardized colored region image includes: The image of the colored part is subjected to image normalization processing to obtain a normalized image of the colored part; The normalized color region image is denoised to obtain a denoised color region image; The image size of the denoised color region image is adjusted to a standardized image size to obtain a standardized color region image.

6. The method according to claim 2, characterized in that, The camera's operating data includes device temperature data; the operating performance parameters include device temperature performance parameters. The step of evaluating the working performance of the camera's working data to obtain the corresponding working performance parameters includes: If the device temperature data is less than or equal to the standard temperature data, then the standard temperature performance parameter is determined as the device temperature performance parameter. If the device temperature data is greater than the standard temperature data, then the temperature data difference between the device temperature data and the standard temperature data is determined, and the loss temperature performance parameter is determined based on the temperature data difference. The difference between the standard temperature performance parameter and the loss temperature performance parameter is determined as the device temperature performance parameter.

7. The method according to claim 2, characterized in that, The camera's operating data includes image brightness data; the operating performance parameters include image brightness performance parameters. The step of evaluating the working performance of the camera's working data to obtain the corresponding working performance parameters includes: Determine the deviation range between the image brightness data and the standard brightness data; If the deviation amplitude is less than or equal to the deviation amplitude threshold, then the standard brightness performance parameter is determined as the image brightness performance parameter; If the deviation amplitude is greater than the deviation amplitude threshold, then the amplitude difference between the deviation amplitude and the deviation amplitude threshold is determined, and the loss brightness performance parameter is determined based on the amplitude difference. The difference between the standard brightness performance parameter and the lost brightness performance parameter is determined as the image brightness performance parameter.

8. The method according to claim 2, characterized in that, The camera's operating data includes device frame rate data; the operating performance parameters include device frame rate performance parameters. The step of evaluating the working performance of the camera's working data to obtain the corresponding working performance parameters includes: If the device frame rate data is greater than or equal to the standard frame rate data, then the standard frame rate performance parameter is determined as the device frame rate performance parameter. If the device frame rate data is less than the standard frame rate data, then the frame rate data difference between the standard frame rate data and the device frame rate data is determined, and the loss frame rate performance parameter is determined based on the frame rate data difference. The difference between the standard frame rate performance parameter and the lost frame rate performance parameter is determined as the device frame rate performance parameter.

9. The method according to claim 2, characterized in that, The core device usage data includes the number of times the core device has been used. The process of evaluating the performance of the core components based on their usage data to obtain the core component performance parameters corresponding to the usage data includes: The usage time of the core device is determined based on the number of times the core device is used and the duration of a single use of the core device. Obtain the total available time of the core device, determine the time ratio between the time the core device has been used and the total available time of the core device, and determine the core device performance parameters corresponding to the core device usage data based on the time ratio.

10. The method according to claim 1, characterized in that, The predicted availability status includes a predicted availability score; the process of predicting the availability status of the access verification camera based on the image quality parameters of the location, the operational performance parameters, the core component performance parameters, and the camera element parameter weights, to obtain the predicted availability status of the access verification camera, includes: A camera element fusion feature vector is generated based on the image quality parameters of the described area, the operating performance parameters, the core device performance parameters, and the camera element parameter weights. Obtain a camera availability status prediction model; the camera availability status prediction model contains N camera availability status decision trees; different camera availability status decision trees are used to represent different camera availability status decision strategies; N is a positive integer; Traverse the available state decision trees of the N cameras to obtain the available state decision tree of the i-th camera; i is a positive integer less than N; The camera availability status decision score is obtained by processing the camera element fusion feature vector through the i-th camera availability status decision tree to obtain the i-th camera availability status decision score. When the decision tree for the available states of the N cameras is traversed, the available state decision scores of the N cameras are weighted and summed to obtain the predicted available score for the permission verification camera.

11. The method according to claim 1, characterized in that, The predicted availability status includes a predicted availability score; the method further includes: Based on the predicted available score and the total available time of the cameras, determine the predicted remaining usage time corresponding to the permission verification camera; If the predicted remaining usage time is less than the remaining usage time threshold, a maintenance notification is sent to the camera management terminal; the maintenance notification is used to instruct the permission verification camera to undergo maintenance processing.

12. A camera state prediction device, characterized in that, include: The data acquisition module is used to acquire the core camera data of the permission verification camera; The permission verification camera refers to a camera used to capture images of the permission verification area of ​​the first object; the permission verification area image is used to verify the business permissions of the first object; the core data of the camera includes the permission verification area image, camera working data, and core device usage data; The element evaluation module is used to evaluate the camera element of the camera core data to obtain the image quality parameters of the area corresponding to the permission verification area image, the working performance parameters of the camera working data, and the core device performance parameters of the core device usage data. The status prediction module is used to perform availability status prediction processing on the permission verification camera based on the image quality parameters of the location, the working performance parameters, the core component performance parameters, and the camera element parameter weights, to obtain the predicted availability status of the permission verification camera. The predicted availability status is used to assess the availability of the permission verification camera.

13. A computer device, characterized in that, include: Processor, memory, and network interface; The processor is connected to the memory and the network interface, wherein the network interface is used to provide data communication functions, the memory is used to store program code, and the processor is used to call the program code to execute the method according to any one of claims 1-11.

14. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program adapted to be loaded by a processor and to execute the method according to any one of claims 1-11.

15. A computer program product comprising a computer program / instructions, characterized in that, When the computer program / instructions are executed by the processor, they can perform the method described in any one of claims 1-11.

Images