Monitoring information full-process management and control method and system based on artificial intelligence

Abstract

The invention discloses a monitoring information full-process management and control method and system based on artificial intelligence, and belongs to the technical field of computers, and the system comprises a process management and control middle table, a self-adaptive acquisition module, a transmission flow prediction module, a data transmission module, a data analysis module and a visual presentation module. Determining image acquisition parameters, acquiring image information of a target area, outputting a future transmission flow prediction result, determining a transmission bandwidth size, transmitting the image information according to the transmission bandwidth size, and obtaining an image feature, an image mode and an image analysis result output by the multi-dimensional analysis model; and generating an information presentation interface from the image features, the image modes and the image analysis results according to the user portrait and the current use scene, and displaying the information presentation interface. According to the invention, the quality of image information and the real-time performance and reliability of transmission are improved, and the analysis capability of monitoring information and the decision-making efficiency of the whole management and control process are improved.

Description

Technical Field

[0001] The present invention belongs to the field of computer technology, and specifically relates to a method and system for full-process control of monitoring information based on artificial intelligence. Background Art

[0002] In today's era of rapid development of digitalization and informatization, full-process control of surveillance information plays a vital role in many fields, such as security monitoring, industrial production monitoring, and network information monitoring.

[0003] Methods for managing and controlling the entire surveillance process typically encompass multiple stages, including information collection, transmission, and analysis. Existing methods rely heavily on hardware devices such as various sensors and cameras during the information collection phase. However, the accuracy of collected data is often significantly affected by environmental factors, such as changes in light that interfere with the clarity of video surveillance images, and data deviations due to aging or harsh environments. Conventional wired or wireless communication methods are often used during transmission. However, when faced with large amounts of surveillance information, bandwidth bottlenecks can easily occur, leading to information delays and packet loss, impacting the acquisition and processing of real-time information in subsequent stages. Analysis methods primarily rely on manually set rules or simple statistical analysis algorithms, which have limited ability to identify complex patterns and anomalies. This makes it difficult to extract deep information from massive amounts of data, and potentially important information can be easily missed. Summary of the Invention

[0004] In view of the above-mentioned existing problems, the present invention aims to improve the quality of image information as well as the real-time and reliability of transmission, and enhance the analysis capability of monitoring information and the decision-making efficiency of the entire management and control process.

[0005] In order to solve the above technical problems, a full-process control method of monitoring information based on artificial intelligence is proposed, including: Image acquisition parameters are determined based on the environmental feature information and ambient light information of the target area at the current time, and image information of the target area at the current time is acquired based on the image acquisition parameters; the historical data transmission size of the target area at each historical moment is input into the transmission traffic prediction model to obtain a future transmission traffic prediction result output by the transmission traffic prediction model; the transmission bandwidth is determined based on the future transmission traffic prediction result, the current time and the total amount of data of the image information, and the image information is transmitted with the transmission bandwidth; the image information is input into a multidimensional analysis model to obtain the image features, image patterns and image analysis results output by the multidimensional analysis model, and encrypted storage is performed based on the image analysis results; the image features, the image patterns and the image analysis results are generated into an information presentation interface and displayed according to the user portrait and the current usage scenario.

[0006] As a preferred solution of the artificial intelligence-based full-process control method for surveillance information described in the present invention, the method of determining image acquisition parameters based on the environmental feature information and ambient light information of the target area at the current time includes: Obtaining distance information between the center area of ​​the target area and the imaging device, the horizontal and vertical edge spans of the image in the imaging device, the overall light intensity value of the target area, the regional light intensity value of each sub-area in the target area, and light intensity fluctuation information of the target area; Determining a focal length parameter for image acquisition based on a focal length range of an imaging device, the distance information, the edge transverse span, and the edge longitudinal span; Determining aperture parameters for image acquisition based on the aperture range of the imaging device, the overall light intensity value, and the light intensity values ​​of each area; determining a sensitivity parameter for image acquisition based on a sensitivity range of the imaging device, the overall light intensity value, and the fluctuation information; The focal length parameter used for image acquisition, the aperture range used for image acquisition, and the sensitivity parameter used for image acquisition are determined as the image acquisition parameters.

[0007] As a preferred solution of the artificial intelligence-based full-process monitoring information control method described in the present invention, the future transmission flow prediction result includes the transmission flow at each moment in the future and the flow status at each moment in the future; Determining the transmission bandwidth includes determining a target future time corresponding to the current time in the future transmission traffic prediction result, and determining a traffic state at the target future time; determining a future transmission flow rate of the target at a future time and a transmission level of the image information; When the traffic state is a low-peak state, determining the transmission bandwidth based on the future transmission traffic, the transmission level, and the total amount of data; When the traffic state is at a peak state, the bandwidth congestion level at the target future time is obtained, and the transmission bandwidth is determined based on the future transmission traffic, the transmission level, the bandwidth congestion level, and the total amount of data.

[0008] As a preferred solution of the artificial intelligence-based monitoring information full-process control method of the present invention, the generating and displaying of the information presentation interface includes: Based on the user portrait and the current usage scenario, generating layout parameters of the image features, the image mode, and the image analysis results; the layout parameters include region range, region location, and region display style; Creating an original feature display area based on the area range and area position of the image feature, filling the original feature display area with the image feature according to the feature value, and rendering the filled original feature display area in the area display style of the image feature to obtain a target feature display area; Creating an original mode description area based on the area range and area position of the image mode, filling the original mode description area with mode description text that matches the image mode in a preset mode description text library, and rendering the filled original mode description area in the area display style of the image mode to obtain a target mode description area; Creating an original analysis result interpretation area based on the regional range and regional position of the image analysis result, filling the original analysis result interpretation area with a visualization result generated according to the data type of the image analysis result, and rendering the filled original analysis result interpretation area in the regional display style of the image analysis result to obtain a target analysis result interpretation area; The target feature display area, the target pattern description area and the target analysis result interpretation area are integrated to obtain the information presentation interface.

[0009] As a preferred solution of the artificial intelligence-based full-process monitoring information control method described in the present invention, the transmission flow prediction model includes data transmission at sample moments and corresponding transmission flow result labels trained, and the training process is as follows: Input each sample time into the initially trained initial prediction model to obtain the transmission flow prediction result of each sample time output by the initial prediction model; Inputting the data transmission and transmission flow prediction results at each sample moment into the first objective optimization function of the initial prediction model to obtain a first output result of the first objective optimization function; If the first output result is greater than a first preset threshold, adjusting the parameters of the first objective optimization function and the hidden layer in the initial prediction model based on the data transmission and transmission flow prediction results at each sample moment, and outputting the transmission flow prediction result at each sample moment based on the adjusted initial prediction model; If it is determined based on the data transmission and transmission flow prediction results at each sample moment that the first training constraint condition is not satisfied, adjusting the parameters of the first objective optimization function and the hidden layer in the initial prediction model until the output result of the adjusted first objective optimization function satisfies the first training constraint condition, thereby obtaining a transmission flow prediction model; The first training constraint condition is that the output result of the optimization function is less than or equal to a first preset threshold.

[0010] As a preferred solution of the artificial intelligence-based full-process monitoring information control method of the present invention, the multi-dimensional analysis model includes label training based on sample image information and its image features, image patterns and image analysis results. The training process is as follows: Inputting each sample image information into the initial analysis model that has completed initial training, and obtaining a multi-dimensional prediction result of each sample image information output by the initial analysis model; the multi-dimensional prediction result includes image features, image patterns, and image analysis results; Inputting the multi-dimensional information and the multi-dimensional prediction result of each sample image information into the second objective optimization function of the initial analysis model to obtain a second output result of the second objective optimization function; If it is determined that the second output result based on each sample image information does not satisfy the second training constraint, adjusting the parameters of the second objective optimization function and the output layer in the initial prediction model until the output result of the adjusted second objective optimization function satisfies the second training constraint, thereby obtaining the multidimensional analysis model; The second training constraint condition is that the output result of the optimization function is less than or equal to a second preset threshold, and the difference between the output results of two adjacent optimization functions is less than a preset difference.

[0011] As a preferred solution of the artificial intelligence-based full-process control method for surveillance information described in the present invention, the encrypted storage includes determining an encryption mode based on the image features, and encrypting the image analysis result based on the encryption mode to obtain an encrypted result; determining index information based on the image pattern; A target storage unit is traversed based on the index information, and the encrypted result is stored in the target storage unit.

[0012] Another object of the present invention is to provide an artificial intelligence-based full-process monitoring information management and control system, which aims to improve the quality of image information, the real-time transmission and reliability, and enhance the ability to analyze monitoring information and the decision-making efficiency of the entire management and control process.

[0013] As a preferred solution of the artificial intelligence-based monitoring information full-process control system described in the present invention, it is characterized by including a process control center, an adaptive acquisition module, a transmission flow prediction module, a data transmission module, a data analysis module, a data storage module and a visualization presentation module; The process control center is respectively connected to the adaptive acquisition module, the transmission traffic prediction module, the data transmission module, the data analysis module and the visualization presentation module to manage each module; the adaptive acquisition module includes a camera unit and a recognition unit based on a convolutional neural network architecture; The adaptive acquisition module is configured to call the recognition unit to determine image acquisition parameters based on the environmental feature information and ambient light information of the target area at the current time, and call the camera unit to capture image information of the target area at the current time using the image acquisition parameters; The transmission flow prediction module is used to input the historical data transmission size of the target area at each historical moment into the transmission flow prediction model to obtain the future transmission flow prediction result output by the transmission flow prediction model; The data transmission module is configured to determine a transmission bandwidth based on the future transmission flow prediction result, the current time, and the total amount of image information data, and transmit the image information to the data analysis module using the transmission bandwidth; The data analysis module is used to input the image information into the multi-dimensional analysis model to obtain the image features, image patterns and image analysis results output by the multi-dimensional analysis model; The data storage module is configured to determine an encryption mode based on the image features, and encrypt the image analysis result based on the encryption mode to obtain an encrypted result; determine index information based on the image mode; traverse a target storage unit based on the index information, and store the encrypted result in the target storage unit; The visualization presentation module is used to generate and display the information presentation interface based on the image features, the image mode and the image analysis results according to the user portrait and the current usage scenario.

[0014] A computer device includes a memory and a processor, wherein the memory stores a computer program, and is characterized in that when the processor executes the computer program, it implements the steps of the full-process management and control method of monitoring information based on artificial intelligence.

[0015] A computer-readable storage medium having a computer program stored thereon, characterized in that when the computer program is executed by a processor, the steps of the method for full-process control of monitoring information based on artificial intelligence are implemented.

[0016] Beneficial effects of the present invention: The present invention uses artificial intelligence recognition technology to adjust image acquisition parameters through an adaptive adjustment module, thereby avoiding the problem of decreased accuracy of image data collected when the light changes suddenly, ensuring that clear and usable image information is collected, and improving the quality of image information. Furthermore, the transmission traffic prediction model based on artificial intelligence determines the transmission bandwidth size for transmitting image information, optimizes the bandwidth allocation mechanism, effectively alleviates the bandwidth bottleneck problem during peak transmission, reduces information delay and packet loss, ensures that image information can be transmitted to subsequent links in real time and completely, and enhances the real-time and reliability of image information transmission. On the other hand, by analyzing image information using a multi-dimensional analysis model of artificial intelligence, it is possible to extract image features, image patterns and image analysis results from the image information, avoid missing details in the image information, and enhance the ability to analyze monitoring information. Finally, by generating a diversified and personalized information presentation interface based on user portraits and usage scenarios, the needs of different user roles are met, allowing users to obtain key information more conveniently and intuitively, and improving the decision-making efficiency of the entire management and control process. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0018] Figure 1 An overall flow chart of an artificial intelligence-based full-process monitoring information management method provided for one embodiment of the present invention.

[0019] Figure 2 An embodiment diagram of an electronic device for an artificial intelligence-based full-process monitoring information management method provided by an embodiment of the present invention.

[0020] Figure 3 An embodiment diagram of a computer-readable storage medium for an artificial intelligence-based full-process management and control method for monitoring information provided by an embodiment of the present invention.

[0021] Figure 4 A system solution module diagram of an artificial intelligence-based full-process monitoring information management and control system provided by one embodiment of the present invention.

[0022] In the figure, 300 is an electronic device; 310 is a memory; 320 is a processor; 311 is a computer program; 400 is a computer-readable storage medium; and 411 is a computer program. DETAILED DESCRIPTION

[0023] To make the above-mentioned objects, features, and advantages of the present invention more clearly understood, the following detailed description of the specific embodiments of the present invention is given in conjunction with the accompanying drawings. It is obvious that the described embodiments are only part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in this field without creative work should fall within the scope of protection of the present invention.

[0024] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0025] Secondly, the term "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in various places throughout this specification does not necessarily refer to the same embodiment, nor does it individually or selectively refer to an embodiment that is mutually exclusive of other embodiments.

[0026] The present invention is described in detail with reference to schematic diagrams. For ease of illustration, cross-sectional views of device structures may be partially enlarged and not to scale when describing embodiments of the present invention. Furthermore, the schematic diagrams are merely illustrative and should not limit the scope of the present invention. Furthermore, in actual production, the three-dimensional dimensions of length, width, and depth should be included.

[0027] Furthermore, in the description of the present invention, it should be noted that the terms "upper, lower, inner, and outer" and other references to orientations or positional relationships are based on the orientations or positional relationships shown in the accompanying drawings and are intended solely to facilitate and simplify the description of the present invention. They are not intended to indicate or imply that the devices or components referred to must have, be constructed, or operate in a specific orientation, and therefore should not be construed as limiting the present invention. Furthermore, the terms "first, second, or third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0028] In this disclosure, unless otherwise specified or limited, the terms "mounted," "connected," and "connected" should be interpreted broadly. For example, they may refer to fixed, removable, or integral connections. They may also refer to mechanical, electrical, or direct connections, indirect connections through an intermediary, or internal communication between two components. Those skilled in the art will understand the specific meanings of these terms in this disclosure.

[0029] Example 1, with reference to Figure 1 , which is the first embodiment of the present invention, provides a method for full-process management and control of monitoring information based on artificial intelligence, including: Step 10: determining image acquisition parameters based on the environmental feature information and ambient light information of the target area at the current time, and acquiring image information of the target area at the current time based on the image acquisition parameters.

[0030] Specifically, the full-process control system includes a camera unit and a recognition unit based on a convolutional neural network architecture. The recognition unit also includes an environmental feature recognition unit and a light intensity sensor. The camera unit can be understood as an imaging device. Therefore, the full-process control system uses the convolutional neural network-based recognition unit to identify the target area's current environment and obtain the target area's current environmental feature information. It also uses the light intensity sensor to identify the target area's current environment and obtain the target area's current ambient light information.

[0031] Furthermore, the full-process control system determines image acquisition parameters based on the environmental feature information and ambient light information of the target area at the current time, and calls the imaging device to acquire image information of the target area at the current time using the image acquisition parameters. Specifically, steps 101 to 105 are described as follows: Step 101: Obtain the distance information between the center area of ​​the target area and the imaging device, the horizontal and vertical edge spans of the image in the imaging device, the overall light intensity value of the target area, the regional light intensity value of each sub-area in the target area, and the fluctuation information of the light intensity of the target area.

[0032] Optionally, the environmental feature information in the embodiment of the present invention includes the distance information between the center area of ​​the target area and the imaging device and the openness of the scene where the target area is located, and the ambient light information includes the overall intensity of the ambient light and the uniformity of each sub-area.

[0033] Therefore, the full-process control system obtains the distance information between the center area of ​​the target area and the imaging device, wherein the center area can be determined based on the center pixel point of the target area in the image of the imaging device, such as determining the preset range area of ​​the center pixel point as the center area, and the preset range area is such as 3*3, 4*4, etc.

[0034] Furthermore, the full-process management and control system obtains the horizontal edge span and the vertical edge span of the picture in the imaging device, where the span is the total number of pixels. Therefore, the embodiment of the present invention uses an edge detection algorithm to count the total horizontal edge pixels and the total vertical edge pixels of the picture in the imaging device.

[0035] Furthermore, the full-process management and control system obtains data from the light intensity sensor, wherein the data from the light intensity sensor includes the overall light intensity value of the target area, the regional light intensity value of each sub-area in the target area, and the fluctuation information of the light intensity of the target area.

[0036] Among them, for the regional light intensity value of each sub-area, the target area is divided into multiple small areas, such as 9, 16, 25 sub-areas of the same area, and the light intensity value of each sub-area is measured respectively.

[0037] For the fluctuation information, the number of frames of the target area captured by the imaging device at the current time is obtained, and the light intensity of each frame is determined. Furthermore, the fluctuation information of the light intensity of the target area is determined based on the light intensity of each frame. The fluctuation information is a sequence of the rate of change of the light intensity.

[0038] Step 102 : determining focal length parameters for image acquisition based on the focal length range, distance information, edge transverse span, and edge longitudinal span of the imaging device.

[0039] Furthermore, the full-process control system determines the distance assessment factor of the target area based on the distance information between the central area and the imaging device, as well as the maximum distance information and minimum distance information of the imaging device. The calculation formula of the distance assessment factor is as follows: , in, represents the distance evaluation factor, Indicates the distance information between the center area and the imaging device. Indicates the minimum distance information of the imaging device, Indicates the maximum distance information of the imaging device.

[0040] Furthermore, the full-process control system calculates the scene openness factor based on the total horizontal pixels, total vertical pixels, total horizontal edge pixels, and total vertical edge pixels of the image in the imaging device. The calculation formula of the scene openness factor is as follows: , in, represents the scene openness factor, Indicates the total horizontal pixels of the edge, Indicates the total vertical pixels of the edge, Indicates the total horizontal pixels, Indicates the total vertical pixels.

[0041] Furthermore, the full-process control system obtains the focal length range of the imaging device, that is, the minimum focal length and maximum focal length of the imaging device, and determines the focal length parameter used for image acquisition based on the minimum focal length and maximum focal length of the imaging device, as well as the distance assessment factor and the scene openness factor. The specific formula is as follows: , in, represents the focal length parameter used for image acquisition, Indicates the minimum focal length of the imaging device, Indicates the maximum focal length of the imaging device; and Represents the preset weight coefficient for TDF and SOF, where .

[0042] Step 103 : determining aperture parameters for image acquisition based on the aperture range of the imaging device, the overall light intensity value, and the light intensity values ​​of each region.

[0043] Furthermore, the full-process control system calculates the light intensity assessment factor based on the preset reference light intensity value and the overall light intensity value. The specific calculation formula of the light intensity assessment factor is as follows: , in, represents the light intensity evaluation factor, Represents the overall light intensity value, Indicates the reference light intensity value.

[0044] Furthermore, the full-process control system calculates the light uniformity factor based on the light intensity value of each area. The calculation formula of the light uniformity factor is as follows: , in, represents the light uniformity factor, represents the number of sub-regions, Indicates the The regional light intensity value of each sub-region, Represents the mean of the regional light intensity values ​​of all sub-regions, is the variable index.

[0045] Furthermore, the full-process control system obtains the aperture range of the imaging device, where the aperture range of the imaging device is the minimum aperture and maximum aperture of the imaging device. Based on the minimum aperture and maximum aperture of the imaging device, as well as the light intensity evaluation factor and the light uniformity factor, the aperture parameter used for image acquisition is determined. The specific formula is as follows: , in, represents the aperture parameter used for image acquisition, Indicates the minimum aperture of the imaging device, Indicates the maximum aperture of an imaging device.

[0046] Step 104 : determining a sensitivity parameter for image acquisition based on the sensitivity range, the overall light intensity value, and the fluctuation information of the imaging device.

[0047] Furthermore, the whole process control system obtains the light intensity evaluation factor determined according to the overall light intensity value , and supplement the factor based on the light sufficiency , where the calculation formula is: , Furthermore, the full-process control system determines the light stability factor based on the fluctuation information, wherein the calculation formula of the light stability factor is as follows: , in, represents the light stability factor, Indicates the number of frames of the target area in the imaging device at the current time. Represents fluctuation information (the sequence of the rate of change of light intensity), Indicates the current time No. The light intensity value of the frame, Indicates the current time No. The light intensity value of the frame.

[0048] Furthermore, the full-process control system obtains the sensitivity range of the imaging device, wherein the sensitivity range of the imaging device is the maximum sensitivity and minimum sensitivity of the imaging device.

[0049] Furthermore, the full-process control system determines the sensitivity parameters used for image acquisition based on the maximum and minimum sensitivity of the imaging device, as well as the light sufficiency supplement factor and the light stability factor. The specific formula for the sensitivity parameters is as follows: , in, Represents the sensitivity parameter used for image acquisition, Indicates the minimum sensitivity of the imaging device. Indicates the maximum sensitivity of the imaging device. and Indicates the preset weight coefficients for LSF and SSF.

[0050] Step 105 : Determine the focal length parameter for image acquisition, the aperture range for image acquisition, and the sensitivity parameter for image acquisition as image acquisition parameters.

[0051] Furthermore, the full-process control system determines the focal length parameters, aperture range and sensitivity parameters used for image acquisition as image acquisition parameters.

[0052] The embodiment of the present invention adjusts image acquisition parameters through artificial intelligence recognition technology, avoiding the problem of decreased accuracy of image data collected when light changes suddenly, ensuring that clear and usable image information is collected, and improving the quality of image information.

[0053] Step 20: Input the historical data transmission size of the target area at each historical moment into the transmission traffic prediction model to obtain a future transmission traffic prediction result output by the transmission traffic prediction model.

[0054] Optionally, a transmission traffic prediction model is embedded in the full-process control system of an embodiment of the present invention, wherein the transmission traffic prediction model is trained based on the data transmission size at the sample moment and its corresponding transmission traffic result label, and the transmission traffic result label includes a transmission traffic size label and a traffic status label, wherein the training process of the transmission traffic prediction model is as described in steps 201 to 204.

[0055] Therefore, the full-process management and control system obtains the historical data transmission size of the target area at each historical moment, where the historical moment can be obtained based on actual conditions, such as the past 1 month, the past 15 days, etc., and outputs the historical data transmission size of the target area at each historical moment into the transmission traffic prediction model to obtain the future transmission traffic prediction results output by the transmission traffic prediction model, where the future transmission traffic prediction results include the transmission traffic at each future moment and the traffic status at each future moment.

[0056] In one embodiment, the first training constraint condition is that the output result of the optimization function is less than or equal to a first preset threshold. The first preset threshold is set according to actual conditions. Specifically, steps 201 to 204 are described as follows: In step 201 , each sample moment is input into an initially trained initial prediction model to obtain a transmission flow prediction result at each sample moment output by the initial prediction model.

[0057] Optionally, the initial prediction model of an embodiment of the present invention is obtained in advance through training samples. Therefore, the full-process management and control system inputs each sample moment into the initially trained initial prediction model to obtain the transmission flow prediction result of each sample moment output by the initial prediction model.

[0058] Step 202: Input the data transmission size and transmission flow prediction results at each sample moment into the first objective optimization function of the initial prediction model to obtain a first output result of the first objective optimization function.

[0059] Furthermore, the full-process control system inputs the data transmission size and transmission flow prediction results at each sample moment into the first objective optimization function of the initial prediction model to obtain a first output result of the first objective optimization function, wherein the first objective optimization function is: , in, Represents the first output result, represents the first objective optimization function, represents the number of samples, Indicates the The data transmission size at each sample moment, Indicates the The transmission flow prediction results at the sample time are: Represents the parameters of the first objective optimization function.

[0060] Step 203: If the first output result is greater than the first preset threshold, the parameters of the hidden layer in the first objective optimization function and the initial prediction model are adjusted based on the data transmission size and the transmission flow prediction result at each sample moment, and the transmission flow prediction result at each sample moment is output based on the adjusted initial prediction model.

[0061] Furthermore, if it is determined that the first output result is less than or equal to the first preset threshold, it means that the parameters of the first objective optimization function and the hidden layer in the initial prediction model are already optimal parameters. Therefore, the full-process management and control system determines the initial prediction model at this time as the transmission traffic prediction model.

[0062] Furthermore, if it is determined that the first output result is greater than the first preset threshold, it means that the parameters of the first objective optimization function and the hidden layer in the initial prediction model are not yet optimal parameters. Therefore, the full-process management and control system determines the first parameter optimization coefficient based on the data transmission size and transmission flow prediction results at each sample moment. The specific formula of the first parameter optimization coefficient is: , in, represents the first parameter optimization coefficient, and Indicates preset parameters.

[0063] Furthermore, the full-process control system adjusts the parameters of the first objective optimization function and the hidden layer in the initial prediction model through the first parameter optimization coefficient, and outputs the transmission flow prediction result at each sample time based on the adjusted initial prediction model. In the embodiment of the present invention, the parameters of the hidden layer include the weight matrix and bias vector of the hidden layer, wherein the parameter adjustment process is as follows: For the parameters of the first objective optimization function, the optimized parameters are : .

[0064] For the parameters of the hidden layer, the optimized weight matrix is : , the optimized bias vector is : ,in, represents the original weight matrix of the hidden layer, Represents the original bias vector of the hidden layer.

[0065] In step 204, if it is determined that the first training constraint is not satisfied based on the data transmission size and transmission flow prediction results at each sample moment, the parameters of the hidden layer in the first objective optimization function and the initial prediction model are adjusted until the output result of the adjusted first objective optimization function satisfies the first training constraint, thereby obtaining a transmission flow prediction model.

[0066] Furthermore, if the first training constraint is determined to be met based on the data transmission size and transmission flow prediction results at each sample moment, that is, the output result of the adjusted first objective optimization function is less than or equal to the first preset threshold, it means that the parameters of the first objective optimization function and the hidden layer in the initial prediction model are already optimal parameters, and the full-process management and control system determines the initial prediction model at this time as the transmission flow prediction model.

[0067] Furthermore, if it is determined that the first training constraint condition is not met based on the data transmission size and transmission flow prediction results at each sample moment, that is, the output result of the adjusted first objective optimization function is greater than the first preset threshold, the full-process management and control system continues to adjust the parameters of the first objective optimization function and the hidden layer in the initial prediction model according to the data transmission size and transmission flow prediction results at each sample moment until the output result of the first objective optimization function of the adjusted initial prediction model meets the first training constraint condition, thereby obtaining a transmission flow prediction model.

[0068] The embodiment of the present invention trains a transmission traffic prediction model, which can subsequently determine the transmission bandwidth size for transmitting image information based on the artificial intelligence transmission traffic prediction model, optimize the bandwidth allocation mechanism, effectively alleviate the bandwidth bottleneck problem during transmission peaks, reduce information delays and packet loss, ensure that image information can be transmitted to subsequent links in real time and completely, and enhance the real-time and reliability of image information transmission.

[0069] Step 30 , determining the transmission bandwidth size according to the future transmission flow prediction result, the current time and the total amount of image information data, and transmitting the image information with the transmission bandwidth size.

[0070] Furthermore, the full-process management and control system matches the future transmission flow prediction results with the current time to obtain the future transmission flow and flow status at the current time.

[0071] Furthermore, the full-process management and control system determines the transmission bandwidth size based on the future transmission traffic and traffic status at the current time, and the total amount of image information data at the current time, as described in steps 301 to 304, and transmits the image information according to the transmission bandwidth size.

[0072] Specifically, steps 301 to 304 are described as follows: Step 301: determine a target future time corresponding to the current time in the future transmission traffic prediction result, and determine the traffic state at the target future time.

[0073] Optionally, the future transmission flow prediction result in the embodiment of the present invention includes the following: Transmission traffic And every moment in the future Traffic status In one embodiment, the future transmission flow prediction result is as follows: ,in, Indicates the total amount and traffic status at future times Including peak state and low peak state, when the flow state is peak state, ; When the traffic state is low peak, .

[0074] The peak state indicates that the flow index and load index carried by the system are greater than or equal to the preset level during a specific time period. The off-peak state indicates that the flow index and load index carried by the system are less than the preset level during a specific time period.

[0075] Furthermore, the full-process management and control system matches the future transmission traffic prediction results with the current time, determines the target future moment corresponding to the current time in the future transmission traffic prediction results, and determines the traffic state at the target future moment, which is the traffic state when transmitting image information.

[0076] In one embodiment, the future transmission flow prediction result , the current time corresponds to the third future moment in the future transmission flow prediction result ,Therefore, the traffic state when transmitting image information is a low-peak state.

[0077] Step 302: Determine the future transmission flow rate and the transmission level of the image information at the target future time.

[0078] Furthermore, the full-process management and control system determines the traffic status and transmission level of the image information at the target future moment, and matches them in the preset mapping table according to the transmission level to obtain the expected time for completing the transmission of the image information, wherein the preset mapping table is constructed in advance based on the mapping relationship between each transmission level and its corresponding expected time.

[0079] Step 303: If the traffic state is a low-peak state, the transmission bandwidth size is determined based on the future transmission traffic, transmission level and total data volume.

[0080] Furthermore, if the traffic state is determined to be off-peak, the full-process control system determines the transmission bandwidth size based on the future transmission traffic, expected time, and total data volume. The specific calculation formula is as follows: , in, Indicates the transmission bandwidth size. Indicates the future transmission flow, Indicates the transmission flow weight coefficient in the off-peak state, Indicates the total amount of data, Indicates the expected time.

[0081] Step 304: If the traffic state is at a peak state, the bandwidth congestion level at the target future time is obtained, and the transmission bandwidth size is determined based on the future transmission traffic, transmission level, bandwidth congestion level and total data volume.

[0082] Furthermore, if the traffic state is determined to be a peak state, the full-process control system obtains the bandwidth congestion level at the target future time, and matches it with the preset transmission restriction table according to the bandwidth congestion level to obtain a transmission restriction factor that matches the bandwidth congestion level. The preset transmission restriction table is constructed in advance based on the mapping relationship between each bandwidth congestion level and its corresponding transmission restriction factor.

[0083] Furthermore, the full-process control system determines the transmission bandwidth size based on future transmission traffic, expected time, transmission limiting factor, and total data volume. The specific calculation formula is as follows: , in, Indicates the transmission flow weight coefficient in peak state, Represents the transmission limiting factor, where the greater the bandwidth congestion, the greater the transmission limiting factor.

[0084] The embodiments of the present invention can adjust the transmission bandwidth size of image information during transmission according to peak and off-peak conditions, optimize the bandwidth allocation mechanism, effectively alleviate the bandwidth bottleneck problem during peak transmission, reduce information delay and packet loss, ensure that image information can be transmitted to subsequent links in real time and completely, and enhance the real-time performance and reliability of image information transmission.

[0085] Step 40: input the image information into the multi-dimensional analysis model to obtain the image features, image patterns and image analysis results output by the multi-dimensional analysis model.

[0086] Optionally, a multi-dimensional analysis model is embedded in the full-process management and control system of an embodiment of the present invention, wherein the multi-dimensional analysis model is obtained based on label training of sample image information and its image features, image patterns and image analysis results. The specific training process of the multi-dimensional analysis model is as described in steps 401 to 403.

[0087] Therefore, the full-process control system inputs image information into the multi-dimensional analysis model and obtains the image features, image patterns, and image analysis results output by the multi-dimensional analysis model. The specific analysis process is as follows: The image information in the embodiment of the present invention is a grayscale image represented by a two-dimensional matrix, and the image size of the image information is Pixels. The multi-dimensional analysis model includes an image feature extraction layer, an image pattern recognition layer, and an image analysis layer. Therefore, the full-process control system inputs image information into the multi-dimensional analysis model. The image feature extraction layer extracts the image information and obtains image features. The specific formula is: , in, represents image features, O represents image information, represents the weight coefficient matrix, Indicates that the image is at coordinates The gray value at represents the adjustment parameter, Represents the variable used to traverse the coordinates of all pixels in the image information; It represents the maximum value that can be obtained within the entire image pixel coordinate range. , i and j are variable indices.

[0088] Furthermore, the image pattern recognition layer clusters the image features to obtain multiple clustering results, and the pattern identifier corresponding to the clustering result with the largest number of image features is determined as the image pattern. .

[0089] In one embodiment, the setting identifies Different image modes, initialization Cluster center vectors , for each image feature, calculate the distance metric to each cluster center , the specific calculation formula is as follows: , in, Indicates the image features, Indicates the cluster centers, The dimension representing the cluster centers.

[0090] Furthermore, each image feature is assigned to the category corresponding to the cluster center with the smallest distance. After multiple iterations of updating the cluster centers, the final multiple clustering results are obtained.

[0091] Furthermore, the image analysis layer analyzes the image features and image patterns to generate image analysis results. , the specific formula is as follows: , in, Represents the coefficients of each dimension of image features under different modes, represents the adjustment parameter, Represents the exponential function, i is the variable index, which is a positive constant.

[0092] Specifically, in one embodiment, the second training constraint condition is: the output result of the optimization function is less than or equal to a second preset threshold, and the difference between the output results of two adjacent optimization functions is less than a preset difference. The second preset threshold and the preset difference are set according to actual conditions. Steps 401 to 403 are described as follows: Step 401: input each sample image information into the initially trained initial analysis model to obtain a multi-dimensional prediction result of each sample image information output by the initial analysis model.

[0093] Optionally, the initial analysis model in the embodiment of the present invention is pre-trained through training samples. Therefore, the full-process management and control system inputs each sample image information into the initially trained initial analysis model to obtain a multi-dimensional prediction result of each sample image information output by the initial analysis model, wherein the multi-dimensional prediction result includes image features, image patterns and image analysis results.

[0094] Step 402: Input the multi-dimensional information and the multi-dimensional prediction result of each sample image information into the second objective optimization function of the initial analysis model to obtain a second output result of the second objective optimization function.

[0095] Furthermore, the full-process control system inputs the multi-dimensional information and multi-dimensional prediction results of each sample image information into the second objective optimization function of the initial analysis model to obtain a second output result of the second objective optimization function, wherein the second objective optimization function is: , in, Represents the second output result, represents the second objective optimization function, represents the number of samples, Indicates the Multi-dimensional information of sample image information, Indicates the The multi-dimensional prediction results of sample image information, Represents the parameters of the second objective optimization function.

[0096] Furthermore, if it is determined that the second output result is less than or equal to the second preset threshold, and the difference between the output results of two adjacent optimization functions is less than the preset difference, it means that the parameters of the second target optimization function and the output layer in the initial analysis model are already optimal parameters. Therefore, the full-process management and control system determines the initial analysis model at this time as a multi-dimensional analysis model.

[0097] Furthermore, if it is determined that at least one second output result is greater than a second preset threshold, or / and, the difference between the output results of at least two adjacent optimization functions is greater than or equal to the preset difference, it indicates that the parameters of the second target optimization function and the output layer in the initial analysis model are not yet optimal parameters. Therefore, the full-process control system determines the second parameter optimization coefficient based on the multi-dimensional information and multi-dimensional prediction results of each sample image information. , where the second parameter optimization coefficient The specific formula is: , Furthermore, the full-process control system adjusts the parameters of the second objective optimization function and the output layer in the initial analysis model through the second parameter optimization coefficient, and outputs the multi-dimensional information and multi-dimensional prediction results of each sample image information based on the adjusted initial analysis model. The parameter adjustment process is as follows: For the parameters of the second objective optimization function, the optimized parameters are : .

[0098] For the parameters of the output layer, the optimized weight matrix is : , the optimized bias vector is : ,in, represents the original weight matrix of the output layer, Represents the original bias vector of the output layer.

[0099] Furthermore, if it is determined based on the multi-dimensional information and multi-dimensional prediction results of each sample image information that the second training constraint is met, that is, the output result of the adjusted second objective optimization function is less than or equal to the second preset threshold, and the difference between the output results of the two adjacent optimization functions is less than the preset difference, it means that the parameters of the second objective optimization function and the output layer in the initial analysis model are already optimal parameters, and the full-process management and control system will determine the initial analysis model at this time as a multi-dimensional analysis model.

[0100] Step 403: If the second output result based on each sample image information is determined to not satisfy the second training constraint, the parameters of the second objective optimization function and the output layer in the starting prediction model are adjusted until the output result of the adjusted second objective optimization function satisfies the second training constraint, thereby obtaining a multidimensional analysis model.

[0101] Furthermore, if it is determined based on the multi-dimensional information and multi-dimensional prediction results of each sample image information that the second training constraint is not met, that is, at least one second output result of the output result of the adjusted second objective optimization function is greater than the second preset threshold, or / and, the difference between the output results of at least two adjacent optimization functions is greater than or equal to the preset difference, the full-process management and control system continues to adjust the parameters of the second objective optimization function and the output layer in the initial prediction model according to the multi-dimensional information and multi-dimensional prediction results of each sample image information until the output result of the second objective optimization function of the adjusted initial prediction model meets the second training constraint, thereby obtaining a multi-dimensional analysis model.

[0102] The embodiment of the present invention trains a multi-dimensional analysis model, which can subsequently analyze image information through the multi-dimensional analysis model of artificial intelligence. It can extract image features, image patterns and image analysis results from the image information, avoid missing details in the image information, and improve the analysis ability of surveillance information.

[0103] Step 50: Generate and display an information presentation interface based on the image features, image patterns, and image analysis results according to the user portrait and the current usage scenario.

[0104] Furthermore, the full-process management and control system generates a feature display area for image features, a pattern description area for image patterns, and an analysis result interpretation area for image analysis results based on the user portrait and the current usage scenario, and integrates the feature display area, pattern description area, and analysis result interpretation area to obtain an information presentation interface, and displays the information presentation interface on the display interface of the user terminal, as specifically described in steps 501 to 505.

[0105] In one embodiment, steps 501 to 505 are described as follows: Step 501: Generate layout parameters of image features, image patterns, and image analysis results based on the user portrait and the current usage scenario.

[0106] Optionally, the user portrait in the embodiment of the present invention is obtained in advance based on user information, and the user portrait is represented by a vector In the form of, including user age, interest preferences and usage habits, the current usage scenario is represented by vector Indicates, for example, whether it is indoors, outdoors, daytime, nighttime, and browsing purpose.

[0107] Furthermore, the whole process control system is based on user portraits and current usage scenarios , generating layout parameters of image features, image patterns, and image analysis results. The layout parameters include region size, region position, and region display style. The specific formula is as follows: , in, represents the vector dot product operation, express The modulus of the vector, The weight vectors representing the area size, area position, and area display style in the layout respectively.

[0108] For region size, an array representing the width and height ratios of each region, for example, They correspond to the proportion of the three areas in the entire interface width.

[0109] For the regional position, it can be expressed in the form of coordinates, such as taking the upper left corner of the interface as the origin , the coordinate value of the upper left corner vertex of each area, such as Represents the horizontal and vertical coordinates of the upper left corner vertices of the three regions in turn (the unit can be pixels).

[0110] For the regional display style, it includes a set of parameters such as color theme (such as ["light", "dark"] for light or dark theme), font style (font size, font family, etc.), and border style.

[0111] Step 502: Create an original feature display area based on the area size and area position of the image feature, fill the image feature into the original feature display area according to the feature value, render the filled original feature display area with the area display style of the image feature, and obtain the target feature display area.

[0112] Furthermore, the full-process control system creates an original feature display area based on the regional size and regional position of the image feature. It then fills each image feature into the original feature display area based on its characteristic value, in ascending order, from top to bottom, and from left to right, to obtain a filled original feature display area. Furthermore, the full-process control system renders the filled original feature display area based on the regional display style of the image feature to obtain a target feature display area.

[0113] Step 503: Create an original mode description area based on the area size and area position of the image mode, fill the mode description text matching the image mode in the preset mode description text library into the original mode description area, render the filled original mode description area with the area display style of the image mode, and obtain the target mode description area.

[0114] Furthermore, the full-process management and control system creates an original pattern description area based on the area size and area position of the image pattern, and matches the image pattern in the preset pattern description text library to obtain the pattern description text that matches the image pattern in the preset pattern description text library, wherein the preset pattern description text library is constructed based on each image pattern and its corresponding pattern description text.

[0115] Furthermore, the full-process management and control system fills the mode description text into the original mode description area to obtain the filled original mode description area, and renders the filled original mode description area according to the regional display style of the image mode to obtain the target mode description area.

[0116] Step 504: Create an original analysis result interpretation area based on the area size and area position of the image analysis result, fill the original analysis result interpretation area with the visualization result generated according to the data type of the image analysis result, and render the filled original analysis result interpretation area with the area display style of the image analysis result to obtain the target analysis result interpretation area.

[0117] Furthermore, the full-process management and control system creates interpretation areas for the original analysis results based on the size and location of the image analysis results, and generates visualizations based on the data type of the image analysis results. In one embodiment, the image analysis results are numerical values, such as a comprehensive image score, and the visualizations can be in text form. The image analysis results are multi-dimensional vectors (e.g., containing information such as the proportions of elements of different categories), and the visualizations can be bar charts, pie charts, and other forms drawn using a chart library.

[0118] Furthermore, the full-process management and control system fills the visualization results into the original analysis result interpretation area to obtain the filled original analysis result interpretation area, and renders the filled original analysis result interpretation area according to the regional display style of the image analysis result to obtain the target analysis result interpretation area.

[0119] Step 505 , the target feature display area, the target pattern description area, and the target analysis result interpretation area are integrated to obtain an information presentation interface.

[0120] Furthermore, the full-process management and control system integrates the target feature display area, the target pattern description area and the target analysis result interpretation area to obtain an information presentation interface.

[0121] The embodiments of the present invention meet the needs of different user roles by generating diversified and personalized information presentation interfaces based on user portraits and usage scenarios, allowing users to obtain key information more conveniently and intuitively, thereby improving the decision-making efficiency of the entire management and control process.

[0122] The embodiments of the present invention adjust image acquisition parameters through artificial intelligence recognition technology, avoiding the problem of decreased accuracy of collected image data when light levels suddenly change, ensuring the acquisition of clear and usable image information, and improving the quality of image information. Furthermore, an artificial intelligence-based transmission traffic prediction model determines the transmission bandwidth for image information, optimizing the bandwidth allocation mechanism and effectively alleviating bandwidth bottlenecks during peak transmission times. This reduces information delays and packet loss, ensuring that image information can be transmitted to subsequent links in real time and completely, and enhancing the real-time and reliability of image information transmission. Furthermore, by analyzing image information using an artificial intelligence multi-dimensional analysis model, image features, image patterns, and image analysis results can be extracted from the image information, avoiding the omission of details in the image information and improving the analysis capabilities of surveillance information. Finally, by generating diversified and personalized information presentation interfaces based on user profiles and usage scenarios, the needs of different user roles are met, allowing users to more conveniently and intuitively obtain key information, thereby improving the decision-making efficiency of the entire management and control process.

[0123] In one embodiment, the artificial intelligence-based full-process monitoring information management method also includes encrypting and storing the image processing results.

[0124] It should be noted that the encryption mode is determined based on the image features, and the image analysis result is encrypted based on the encryption mode to obtain the encrypted result.

[0125] Optionally, in the embodiment of the present invention, the dimension of the image feature is z, so the whole process control system controls the image feature Quantization processing is performed, where Represents the zth image feature vector in the image feature. The specific quantization formula is as follows: , in, represents the quantization result of the zth image feature vector, Represents the smallest image feature vector in the image features, Represents the largest image feature vector among image features; Indicates the number of preset quantization levels, such as , which means quantizing the image feature vector in the image feature to 10 levels from 0 to 9.

[0126] Therefore, the quantized image feature vector obtained after the image features are quantized is ; Furthermore, the whole process control system maps the quantized image feature vector to different encryption modes. In one embodiment, the mode center vectors corresponding to the preset q encryption modes are: , j is the variable index, which is a positive number. The similarity measure between the quantized image feature vector and the pattern center vector of each encryption pattern is calculated, and the encryption pattern corresponding to the pattern center vector with the largest similarity measure is taken as the final encryption pattern. , where the similarity measure The calculation formula is as follows: , Among them, G represents the dimension of the pattern center vector, and the encryption pattern Including encryption parameter set, encryption mode It can be expressed as .

[0127] in, , , , Furthermore, if the image analysis result is a numerical value, the whole process control system will be encrypted. Encryption parameter set in Encrypt the image analysis results to obtain the encrypted results , the specific encryption formula is as follows: , Furthermore, if the image analysis result is a vector, Then encrypt each element in the image analysis result separately, namely: , in, The core of the encryption algorithm is to achieve data obfuscation through a combination of linear transformation and nonlinear sine function; i is the variable index, which is a positive number; Furthermore, index information is determined based on the image mode.

[0128] Furthermore, the whole process control system determines the index information based on the image mode to generate index information, wherein the index information The generation formula is as follows: , in, and Represents the preset weight parameter.

[0129] It should also be noted that the target storage unit is traversed based on the index information, and the encrypted result is stored in the target storage unit.

[0130] Optionally, the storage unit set in the embodiment of the present invention (u is the total number of storage units), each storage unit has its corresponding index range. Therefore, the full process control system traverses the storage unit corresponding to the storage unit index range with the smallest distance from the index information according to the index information, and obtains the target storage unit , where the traversal formula is as follows: , in, Represents a storage unit (belonging to the storage unit set) corresponding index range.

[0131] Furthermore, the whole-process control system stores the encrypted results in the target storage unit.

[0132] The embodiment of the present invention encrypts the image analysis results through image features, thereby improving the security of the image analysis results. At the same time, index information of the storage unit where the encrypted results are stored is generated according to the image pattern, thereby improving the efficient management of the encrypted results.

[0133] Example 2, reference Figure 2 and Figure 3 The second embodiment of the present invention is different from the previous embodiment in that: like Figure 2 As shown, an embodiment of the present invention provides an electronic device 300, including a memory 310, a processor 320, and a computer program 311 stored in the memory 310 and executable on the processor 320. When the processor 320 executes the computer program 311, the following steps are implemented: Determining image acquisition parameters based on environmental feature information and ambient light information of the target area at the current time, and acquiring image information of the target area at the current time based on the image acquisition parameters; Input the historical data transmission size of the target area at each historical moment into the transmission traffic prediction model to obtain the future transmission traffic prediction result output by the transmission traffic prediction model; Determine the transmission bandwidth size based on the future transmission traffic prediction result, the current time and the total amount of image information data, and transmit the image information based on the transmission bandwidth size; Inputting image information into a multi-dimensional analysis model to obtain image features, image patterns and image analysis results output by the multi-dimensional analysis model; Based on the user portrait and current usage scenario, the image features, image patterns and image analysis results are generated into an information presentation interface and displayed.

[0134] See also Figure 3 , Figure 3 Detailed description of an embodiment of a computer-readable storage medium provided by an embodiment of the present invention. Figure 3 As shown, this embodiment provides a computer-readable storage medium 400 on which a computer program 411 is stored. When the computer program 411 is executed by a processor, the following steps are implemented: Determining image acquisition parameters based on environmental feature information and ambient light information of the target area at the current time, and acquiring image information of the target area at the current time based on the image acquisition parameters; Input the historical data transmission size of the target area at each historical moment into the transmission traffic prediction model to obtain the future transmission traffic prediction result output by the transmission traffic prediction model; Determine the transmission bandwidth size based on the future transmission traffic prediction result, the current time and the total amount of image information data, and transmit the image information based on the transmission bandwidth size; Inputting image information into a multi-dimensional analysis model to obtain image features, image patterns and image analysis results output by the multi-dimensional analysis model; Based on the user portrait and current usage scenario, the image features, image patterns and image analysis results are generated into an information presentation interface and displayed.

[0135] If the functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the portion that contributes to the prior art, or the portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage media include various media that can store program code, such as USB flash drives, mobile hard drives, read-only memories (ROMs), random access memories (RAMs), magnetic disks, or optical disks.

[0136] The logic and / or steps represented in the flowcharts or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing the logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (e.g., a computer-based system, a system including a processor, or other system that can fetch and execute instructions from an instruction execution system, apparatus, or device). For purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transport a program for use by, or in conjunction with, an instruction execution system, apparatus, or device.

[0137] More specific examples (a non-exhaustive list) of computer-readable media include the following: an electrical connection with one or more wires (electronic devices), a portable computer disk cartridge (magnetic devices), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disc read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, and then editing, interpreting, or processing in another suitable manner as necessary, and then storing it in a computer memory.

[0138] It should be understood that various components of the present invention may be implemented using hardware, software, firmware, or a combination thereof. In the above-described embodiments, multiple steps or methods may be implemented using software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented using hardware, as in another embodiment, any one of the following technologies known in the art or a combination thereof may be used: a discrete logic circuit having logic gate circuits for implementing logic functions on data signals, an application-specific integrated circuit having suitable combinational logic gate circuits, a programmable gate array (PGA), a field-programmable gate array (FPGA), etc.

[0139] Example 3, reference Figure 4 , which is the third embodiment of the present invention, provides an artificial intelligence-based monitoring information full-process management and control system, including a process management and control center, an adaptive collection module, a transmission flow prediction module, a data transmission module, a data analysis module, and a visualization presentation module; The process control center connects to the adaptive acquisition module, transmission flow prediction module, data transmission module, data analysis module, and visualization presentation module to manage each module. The adaptive acquisition module includes a camera unit and a recognition unit based on a convolutional neural network architecture.

[0140] Optionally, the recognition unit in the adaptive acquisition module includes an environmental feature recognition unit and a light intensity sensor. Therefore, the recognition unit is called to determine image acquisition parameters based on the environmental feature information and ambient light information of the target area at the current time, and the camera unit is called to capture image information of the target area at the current time with the image acquisition parameters.

[0141] Optionally, a transmission traffic prediction model is embedded in the transmission traffic prediction module, wherein the transmission traffic prediction model is trained based on the data transmission size at the sample moment and its corresponding transmission traffic result label. Therefore, the historical data transmission size of the target area at each historical moment is input into the transmission traffic prediction model to obtain the future transmission traffic prediction result output by the transmission traffic prediction model.

[0142] Optionally, the data transmission module matches the future transmission traffic prediction result with the current time to obtain the future transmission traffic and traffic status at the current time, and determines the transmission bandwidth size based on the future transmission traffic and traffic status at the current time, as well as the total amount of image information data at the current time, and transmits the image information to the data analysis module at the transmission bandwidth size.

[0143] Optionally, a multidimensional analysis model is embedded in the data analysis module, wherein the multidimensional analysis model is trained based on labels of sample image information and its image features, image patterns and image analysis results. Therefore, the image information is input into the multidimensional analysis model to obtain image features, image patterns and image analysis results output by the multidimensional analysis model.

[0144] Optionally, the visualization presentation module generates a feature display area for image features, a pattern description area for image patterns, and an analysis result interpretation area for image analysis results based on the user portrait and the current usage scenario, and integrates the feature display area, pattern description area, and analysis result interpretation area to obtain an information presentation interface, and displays the information presentation interface on the display interface of the user terminal.

[0145] The embodiments of the present invention utilize artificial intelligence (AI) recognition technology to adjust image acquisition parameters through an adaptive adjustment module, thus avoiding the problem of decreased image data accuracy during sudden changes in lighting conditions, ensuring the capture of clear and usable image information, and improving image quality. Furthermore, an AI-based transmission traffic prediction model determines the transmission bandwidth required for image information transmission, optimizing the bandwidth allocation mechanism and effectively alleviating bandwidth bottlenecks during peak transmission times. This reduces information delays and packet loss, ensuring that image information can be transmitted to subsequent links in real time and completely, and enhancing the real-time and reliability of image information transmission. Furthermore, the use of AI's multi-dimensional analysis model to analyze image information can extract image features, image patterns, and image analysis results from image information, avoiding the omission of details in image information and improving the ability to analyze surveillance information. Finally, by generating diverse and personalized information presentation interfaces based on user profiles and usage scenarios, the needs of different user roles are met, allowing users to more conveniently and intuitively access key information, thereby improving the decision-making efficiency of the entire management and control process.

[0146] In this embodiment, the surveillance information full-process management and control system also includes a data storage module. Therefore, the data storage module determines an encryption mode based on image features and encrypts the image analysis results based on the encryption mode to obtain an encrypted result. Furthermore, the data storage module determines index information based on the image mode, traverses the target storage unit based on the index information, and stores the encrypted result in the target storage unit.

[0147] The embodiment of the present invention encrypts the image analysis results through image features, thereby improving the security of the image analysis results. At the same time, index information of the storage unit where the encrypted results are stored is generated according to the image pattern, thereby improving the efficient management of the encrypted results.

[0148] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A method for full-process control of surveillance information based on artificial intelligence, characterized by: include, Determining image acquisition parameters based on environmental feature information and ambient light information of the target area at the current time, and acquiring image information of the target area at the current time based on the image acquisition parameters; Inputting the historical data transmission size of the target area at each historical moment into the transmission traffic prediction model to obtain a future transmission traffic prediction result output by the transmission traffic prediction model; determining a transmission bandwidth according to the future transmission traffic prediction result, the current time and the total amount of image information data, and transmitting the image information with the transmission bandwidth; Inputting the image information into a multidimensional analysis model, obtaining image features, image patterns, and image analysis results output by the multidimensional analysis model, and encrypting and storing the image analysis results; The image features, the image mode and the image analysis results are generated into an information presentation interface and displayed according to the user portrait and the current usage scenario.

2. The method for full-process control of surveillance information based on artificial intelligence according to claim 1, characterized in that: Determining the image acquisition parameters based on the environmental feature information and ambient light information of the target area at the current time includes: Obtaining distance information between the center area of ​​the target area and the imaging device, the horizontal and vertical edge spans of the image in the imaging device, the overall light intensity value of the target area, the regional light intensity value of each sub-area in the target area, and light intensity fluctuation information of the target area; Determining a focal length parameter for image acquisition based on a focal length range of an imaging device, the distance information, the edge transverse span, and the edge longitudinal span; Determining aperture parameters for image acquisition based on the aperture range of the imaging device, the overall light intensity value, and the light intensity values ​​of each area; determining a sensitivity parameter for image acquisition based on a sensitivity range of the imaging device, the overall light intensity value, and the fluctuation information; The focal length parameter used for image acquisition, the aperture range used for image acquisition, and the sensitivity parameter used for image acquisition are determined as the image acquisition parameters.

3. The method for full-process control of surveillance information based on artificial intelligence according to claim 2, characterized in that: The future transmission flow prediction result includes the transmission flow at each moment in the future and the flow status at each moment in the future; Determining the transmission bandwidth includes determining a target future time corresponding to the current time in the future transmission traffic prediction result, and determining a traffic state at the target future time; determining a future transmission flow rate of the target at a future time and a transmission level of the image information; When the traffic state is a low-peak state, determining the transmission bandwidth based on the future transmission traffic, the transmission level, and the total amount of data; When the traffic state is at a peak state, the bandwidth congestion level at the target future time is obtained, and the transmission bandwidth is determined based on the future transmission traffic, the transmission level, the bandwidth congestion level, and the total amount of data.

4. The method for full-process control of surveillance information based on artificial intelligence according to claim 3, characterized in that: The generating and displaying of the information presentation interface includes: Generate layout parameters of image features, image patterns, and image analysis results based on the user profile and current usage scenario; the layout parameters include region range, region location, and region display style; Creating an original feature display area based on the area range and area position of the image feature, filling the original feature display area with the image feature according to the feature value, and rendering the filled original feature display area in the area display style of the image feature to obtain a target feature display area; Creating an original mode description area based on the area range and area position of the image mode, filling the original mode description area with mode description text that matches the image mode in a preset mode description text library, and rendering the filled original mode description area in the area display style of the image mode to obtain a target mode description area; Creating an original analysis result interpretation area based on the regional range and regional position of the image analysis result, filling the original analysis result interpretation area with a visualization result generated according to the data type of the image analysis result, and rendering the filled original analysis result interpretation area in the regional display style of the image analysis result to obtain a target analysis result interpretation area; The target feature display area, the target pattern description area and the target analysis result interpretation area are integrated to obtain the information presentation interface.

5. The method for controlling the entire process of monitoring information based on artificial intelligence according to claim 4, characterized in that: The transmission flow prediction model includes data transmission at the sample time and the corresponding transmission flow result label training, and the training process is as follows: Input each sample time into the initially trained initial prediction model to obtain the transmission flow prediction result of each sample time output by the initial prediction model; Inputting the data transmission and transmission flow prediction results at each sample moment into the first objective optimization function of the initial prediction model to obtain a first output result of the first objective optimization function; If the first output result is greater than a first preset threshold, adjusting the parameters of the first objective optimization function and the hidden layer in the initial prediction model based on the data transmission and transmission flow prediction results at each sample moment, and outputting the transmission flow prediction result at each sample moment based on the adjusted initial prediction model; If it is determined based on the data transmission and transmission flow prediction results at each sample moment that the first training constraint condition is not satisfied, adjusting the parameters of the first objective optimization function and the hidden layer in the initial prediction model until the output result of the adjusted first objective optimization function satisfies the first training constraint condition, thereby obtaining a transmission flow prediction model; The first training constraint condition is that the output result of the optimization function is less than or equal to a first preset threshold.

6. The method for full-process control of surveillance information based on artificial intelligence according to claim 5, characterized in that: The multi-dimensional analysis model includes label training based on sample image information and its image features, image patterns and image analysis results. The training process is as follows: Inputting each sample image information into the initial analysis model that has completed initial training, and obtaining a multi-dimensional prediction result of each sample image information output by the initial analysis model; the multi-dimensional prediction result includes image features, image patterns, and image analysis results; Inputting the multi-dimensional information and the multi-dimensional prediction result of each sample image information into the second objective optimization function of the initial analysis model to obtain a second output result of the second objective optimization function; If it is determined that the second output result based on each sample image information does not satisfy the second training constraint, adjusting the parameters of the second objective optimization function and the output layer in the initial prediction model until the output result of the adjusted second objective optimization function satisfies the second training constraint, thereby obtaining the multidimensional analysis model; The second training constraint condition is that the output result of the optimization function is less than or equal to a second preset threshold, and the difference between the output results of two adjacent optimization functions is less than a preset difference.

7. The method for controlling the entire process of monitoring information based on artificial intelligence according to claim 6, characterized in that: The encrypted storage includes determining an encryption mode based on the image features, and encrypting the image analysis result based on the encryption mode to obtain an encrypted result; determining index information based on the image pattern; A target storage unit is traversed based on the index information, and the encrypted result is stored in the target storage unit.

8. A system using the artificial intelligence-based full-process monitoring information management method according to any one of claims 1 to 7, characterized in that: It includes process control center, adaptive collection module, transmission flow prediction module, data transmission module, data analysis module, data storage module and visualization presentation module; The process control center is respectively connected to the adaptive acquisition module, the transmission traffic prediction module, the data transmission module, the data analysis module and the visualization presentation module to manage each module; the adaptive acquisition module includes a camera unit and a recognition unit based on a convolutional neural network architecture; The adaptive acquisition module is configured to call the recognition unit to determine image acquisition parameters based on the environmental feature information and ambient light information of the target area at the current time, and call the camera unit to capture image information of the target area at the current time using the image acquisition parameters; The transmission flow prediction module is used to input the historical data transmission size of the target area at each historical moment into the transmission flow prediction model to obtain the future transmission flow prediction result output by the transmission flow prediction model; The data transmission module is configured to determine a transmission bandwidth based on the future transmission flow prediction result, the current time, and the total amount of image information data, and transmit the image information to the data analysis module using the transmission bandwidth; The data analysis module is used to input image information into the multi-dimensional analysis model to obtain image features, image patterns and image analysis results output by the multi-dimensional analysis model; The data storage module is configured to determine an encryption mode based on the image features, and encrypt the image analysis result based on the encryption mode to obtain an encrypted result; determining index information based on the image pattern; Traversing a target storage unit based on the index information, and storing the encrypted result in the target storage unit; The visualization presentation module is used to generate and display the information presentation interface based on the image features, the image mode and the image analysis results according to the user portrait and the current usage scenario.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the full-process management and control method of monitoring information based on artificial intelligence described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of a full-process management and control method of monitoring information based on artificial intelligence as described in any one of claims 1 to 7 are implemented.

Images