Multi-terminal-oriented smart city digital twin map management system

Abstract

The invention relates to the technical field of display management, and discloses a multi-terminal-oriented smart city digital twin map management system, which is characterized in that a target map model is divided into K loading levels; s types of limiting factors are determined based on prior knowledge; determining a terminal device set of the target map model, obtaining performance index parameters of M terminal devices in the terminal device set, and clustering the M terminal devices based on the performance index parameters to obtain N types of feature terminal devices; allocating the nth type of feature terminal equipment to load the target map model according to a random loading level and a limiting factor, and performing fluency analysis on the loaded image set to obtain a fluency score corresponding to the nth type of feature terminal equipment; obtaining a target loading scheme of the nth type of feature terminal equipment based on an optimization algorithm; and loading the target map model on the target terminal equipment based on the target loading scheme.

Description

Technical Field

[0001] The present invention relates to the technical field of display management, and more specifically, it relates to a smart city digital twin map management system for multiple terminals. Background Art

[0002] With the continuous advancement of the construction of smart cities, the digital twin map management system constructs a virtual digital model of the city by integrating data in multiple aspects such as urban infrastructure, transportation, environment, and energy.

[0003] The existing smart city digital twin map management systems need to face various display terminals during application, such as large-screen consoles, desktop computers, tablets, and mobile devices. However, due to the large differences in processor computing power, rendering area, and graphics processing capabilities among various terminal devices, the standardized display processing will experience lag on some terminal devices, thus affecting the operation experience. Summary of the Invention

[0004] The present invention provides a smart city digital twin map management system for multiple terminals to solve the technical problems raised in the background art.

[0005] The present invention provides a smart city digital twin map management system for multiple terminals, including: A model grading module, configured to divide the target map model into K loading levels; wherein, from the first to the Kth loading level, the loading accuracy of the target map model increases in a gradient manner; A loading limit module, configured to determine S limiting factors based on prior knowledge; A device clustering module, configured to determine the set of terminal devices of the target map model, obtain the performance index parameters of M terminal devices in the set of terminal devices, and cluster the M terminal devices based on the performance index parameters to obtain N types of characteristic terminal devices; An intelligent loading module, configured to, within a preset time period, allocate the nth type of characteristic terminal device to load the target map model with a random loading level and limiting factor, and synchronously obtain the corresponding set of loading images at fixed time intervals; An intelligent recognition module, configured to perform a fluency analysis on the set of loading images to obtain the fluency score corresponding to the nth type of characteristic terminal device; 1 ≤ n ≤ N, and n is a positive integer; A loading optimization module, configured to, based on the fluency score, combine an optimization algorithm to obtain the target loading scheme for the nth type of characteristic terminal device; A multi-terminal loading management module, configured to execute the loading of the target map model on the target terminal device based on the target loading scheme of each type of characteristic terminal device.

[0006] Further, the loading accuracy includes: modeling the target city model through triangulation, where a number of sampling points are included in the unit area of the target map model; the difference in the number of sampling points in the unit area between the k-th loading level and the (k + 1)-th loading level of the target map model is ; where k and k + 1 are both indices of the K loading levels.

[0007] Further, based on prior knowledge, determine S impact indicators for the computing power requirements of the terminal device to execute the loading of the target city model, and use each impact indicator as a limiting factor; Among them, the method for determining the impact indicators is as follows: Determine the standard parameters of each impact indicator; Set a standard terminal device; For the standard terminal device to execute the loading of the target map model, it includes: adjusting the standard parameters of the s-th impact indicator with a fixed proportionality coefficient and keeping the standard parameters of the remaining impact indicators unchanged; Obtain the CPU occupancy ratio amplitude, memory occupancy ratio amplitude, disk occupancy ratio amplitude, and GPU occupancy ratio amplitude of the standard terminal device, and obtain the computing power requirements of the s-th impact indicator through weighted fusion; if the computing power requirements of the s-th impact indicator are greater than or equal to the preset computing power requirements threshold, retain the s-th impact indicator, otherwise discard the s-th impact indicator; where 1 ≤ s ≤ S and s is a positive integer.

[0008] Further, cluster M types of terminal devices based on performance metric parameters, including: The performance metric parameters include: CPU computing power indicator, GPU computing power indicator, memory performance indicator, storage performance indicator, and rendering requirement indicator; The CPU computing power indicator includes: CPU main frequency, number of CPU cores, CPU cache size, and CPU floating-point operation parameters; The GPU computing power indicator includes: GPU floating-point operation parameters, video memory bandwidth, and video memory capacity; The memory performance indicator includes: memory capacity and memory bandwidth; The storage performance indicator includes: number of input / output operations per second and data read and load speed; The rendering requirement indicator includes: monitor resolution, monitor size, monitor refresh rate, and monitor color depth; Fuse the performance metric parameters of the m-th terminal device to obtain the loading score of the m-th terminal device; perform K clustering on the M types of terminal devices based on the loading score to obtain N types of characteristic terminal devices; where 1 ≤ m ≤ M and m is a positive integer.

[0009] Further, allocating a feature terminal device of the nth type to load the target map model with a random loading level and a limiting factor includes: for the feature terminal device of the nth type, performing the loading of the target map model with a random loading level and random parameters of S limiting factors.

[0010] Further, performing a smoothness analysis on the loaded image set, including: The loaded image set includes U loaded images, where U = T / t; here, T represents a preset time period, and t represents a fixed time interval. Performing edge detection on the u-th loaded image to obtain the feature edges in the u-th loaded image; where 1 ≤ u ≤ U and u is a positive integer. Determining the intersection points of any number of feature edges in the u-th loaded image, and obtaining the color components of the intersection points and the color components of the pixel points in the eight-neighborhood to construct a matching matrix corresponding to the intersection points. Based on the matching matrix, calculating the Euclidean distance between any one intersection point in the u-th loaded image and the matching matrix of all intersection points in the (u + 1)-th loaded image to obtain a matching score. If the matching score is greater than or equal to a preset threshold, the matching of the corresponding intersection points in the u-th loaded image and the (u + 1)-th loaded image is successful. Respectively obtaining the row and column numbers of the successfully matched intersection points in the u-th loaded image and the row and column numbers in the (u + 1)-th loaded image, and calculating the ratio of the Euclidean distance to the number of successfully matched intersection points to obtain the local fluidity parameter of the u-th loaded image and the (u + 1)-th loaded image. Obtaining the local fluidity parameters of each group of adjacent loaded images, and calculating the variance of U - 1 local fluidity parameters to obtain the smoothness score of the loaded image set.

[0011] Further, obtaining the target loading scheme of the feature terminal device of the nth type by combining an optimization algorithm, including: Step 71, for the feature terminal device of the nth type, initializing and generating R loading schemes, and each loading scheme includes a random loading level and random parameters of S limiting factors. The fitness function includes: using the smoothness score of the loaded image set obtained by the feature terminal device of the nth type executing the r-th loading scheme as the fitness value of the r-th loading scheme; where 1 ≤ r ≤ R and r is a positive integer. Step 72, sorting the R loading schemes from small to large based on the fitness values of the loading schemes to obtain a feature sorting. Step 73, retaining a preset number of loading schemes from front to back for the feature sorting, and updating the remaining loading schemes as parents through crossover or mutation. Step 74: Repeat Step 72 and Step 73 for a preset number of times, and output the loading scheme ranked first in the corresponding feature ranking as the target loading scheme for the nth type of feature terminal device.

[0012] Furthermore, performing the loading of the target map model on the target terminal device includes: Obtaining the performance index parameters of the target terminal device and calculating the loading score of the target terminal device; Clustering the target terminal device based on the loading score of the target terminal device to obtain the type of the feature terminal device to which the target terminal device belongs; Based on the type of the feature terminal device of the target terminal device, obtain the corresponding target loading scheme, and perform the loading of the target map model based on the target loading scheme.

[0013] The beneficial effects of the present invention are as follows: Through the coordinated action of hierarchical loading, terminal device performance clustering, and intelligent fluency detection and optimization algorithms, it is possible to dynamically customize the map loading scheme according to the performance of each terminal, effectively alleviating the carding problem of the traditional unified loading method in a multi-terminal environment, thereby significantly improving the display fluency of the smart city digital twin map in a multi-terminal environment. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 is a module diagram of a smart city digital twin map management system for multiple terminals according to the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0015] Now, the subject matter described herein will be discussed with reference to exemplary embodiments. It should be understood that discussing these embodiments is only to enable those skilled in the art to better understand and thus implement the subject matter described herein, and changes can be made to the functions and arrangements of the elements discussed without departing from the protection scope of the content of this specification. Each example can omit, substitute, or add various processes or components as needed. Additionally, the features described relative to some examples can also be combined in other examples.

[0016] As Figure 1 shown, a smart city digital twin map management system for multiple terminals includes: A model grading module for dividing the target map model into K loading levels; wherein, from the first to the Kth loading level, the loading accuracy of the target map model increases in a gradient manner; A loading limit module for determining S limiting factors based on prior knowledge; The device clustering module is used to determine the set of terminal devices of the target map model, obtain the performance index parameters of M types of terminal devices in the set of terminal devices, and cluster the M types of terminal devices based on the performance index parameters to obtain N types of characteristic terminal devices; The intelligent loading module is used to, within a preset time period, allocate the nth type of characteristic terminal device to load the target map model with a random loading level and a limiting factor, and synchronously obtain the corresponding set of loaded images at fixed time intervals; The intelligent recognition module is used to perform a fluency analysis on the set of loaded images to obtain the fluency score corresponding to the nth type of characteristic terminal device; 1 ≤ n ≤ N, and n is a positive integer; The loading optimization module is used to, based on the fluency score, combine an optimization algorithm to obtain the target loading scheme for the nth type of characteristic terminal device; The multi-terminal loading management module is used to execute the loading of the target map model on the target terminal device based on the target loading scheme of each type of characteristic terminal device.

[0017] In an embodiment of the present invention, the loading accuracy includes: modeling the target urban model through triangulation, and a unit area of the target map model includes a number of sampling points; the difference in the number of sampling points per unit area between the kth loading level and the (k + 1)th loading level of the target map model is ; where k and k + 1 are both indices of K loading levels.

[0018] In an embodiment of the present invention, the target urban model is a three-dimensional model. First, sampling points at various locations in the city are collected, and a complete three-dimensional urban model is constructed by forming a polygon with every three sampling points. Subsequently, to meet the requirements of different loading accuracies, the model is divided into K loading levels. For example, if 10,000 sampling points are collected per unit area on the facade of a certain house, it constitutes the highest loading level (the Kth level); while at a lower loading level (such as the (K - 1)th level), by uniformly removing some sampling points, only 9,000 sampling points are retained to reconstruct the model. In this way, the model accuracy at different loading levels shows a progressive change, providing a flexible and refined selection basis for multi-terminal loading management.

[0019] Specifically, a target map model contains a number of sampling points per unit area, and the number of these sampling points determines the loading accuracy of the map model in this area. To achieve a smooth transition from low loading accuracy to high loading accuracy, the present invention divides the target map model into K loading levels, and each loading level corresponds to a different sampling point density. Specifically, for the k-th and (k + 1)-th loading levels (where both k and k + 1 are indices of the K loading levels), the increase in the number of sampling points per unit area, that is, the difference in the number of sampling points, is used as a key parameter to measure the improvement in loading accuracy.

[0020] For example, in an embodiment, it can be preset that the number of sampling points per unit area of the k-th loading level is N(k), and the number of sampling points per unit area of the (k + 1)-th loading level is N(k + 1), and the following relationship is satisfied between the two: N(k + 1) - N(k) = ΔN; where ΔN is a preset sampling point difference, used to control the amplitude of detail increase between adjacent loading levels. In this way, a lower loading level uses fewer sampling points to reduce the data volume and calculation burden, and is suitable for terminal devices with lower performance or lower requirements for display accuracy; while a higher loading level improves the detail performance of the map model by increasing the number of sampling points, and is suitable for high-performance devices or occasions with higher requirements for display accuracy.

[0021] In an embodiment of the present invention, S influencing indicators of the computing power requirement for the terminal device to execute the loading of the target city model are determined based on prior knowledge, and each influencing indicator is used as a limiting factor; Among them, the method for determining the influencing indicators is as follows: Determine the standard parameters of each influencing indicator; Set a standard terminal device; For the standard terminal device to execute the loading of the target map model, it includes: adjusting the standard parameters of the s-th influencing indicator with a fixed proportional coefficient, and keeping the standard parameters of the remaining influencing indicators unchanged; Obtain the CPU occupancy ratio amplitude H1, memory occupancy ratio amplitude H2, disk occupancy ratio amplitude H3, and GPU occupancy ratio amplitude H4 of the standard terminal device, and obtain the computing power requirement of the s-th influencing indicator through weighted fusion; if the computing power requirement of the s-th influencing indicator is greater than or equal to the preset computing power requirement threshold, retain the s-th influencing indicator, otherwise discard the s-th influencing indicator; where 1 ≤ s ≤ S, and s is a positive integer.

[0022] Specifically, in order to meet the computing power requirements of different terminal devices when loading the target city model, the present invention proposes a method for determining influence indicators based on prior knowledge. By comprehensively considering various performance indicators of the terminal device, several indicators (denoted as S items) that have a significant impact on the computing power demand during the loading process are determined and used as limiting factors for formulating subsequent loading strategies. The specific description is as follows: Basis for determining influence indicators: When the terminal device executes the target city model loading task, the occupancy of hardware resources such as its CPU, memory, disk, and GPU directly affects the loading efficiency and display smoothness. In order to quantify the importance of each resource during the loading process, the present invention pre-determines the standard parameters of each influence indicator based on prior knowledge (such as industry standards, historical data, experimental verification, etc.). The standard parameters reflect the reference values that each indicator should reach under typical working conditions, thus providing basic data for subsequent adjustments and comparisons.

[0023] Setting a standard terminal device: To ensure the comparability of the determination of influence indicators, the present invention selects a standard terminal device as the reference platform. This standard terminal device represents a common or representative device in actual applications, and its performance parameters serve as the benchmark for the adjustment and testing of all influence indicators. By executing the loading process of the target map model on the standard terminal device, the amplitude data of each indicator (such as CPU occupancy ratio, memory occupancy ratio, disk occupancy ratio, GPU occupancy ratio) can be obtained.

[0024] Application of a fixed proportional coefficient: For each influence indicator (denoted as the s-th item, 1 ≤ s ≤ S), the present invention uses a fixed proportional coefficient to adjust its standard parameters. That is, on the premise of keeping the standard parameters of other influence indicators unchanged, only the s-th influence indicator is corrected using a fixed proportional coefficient to make its value closer to the demand changes that may be encountered during the actual loading process. This process can be described as follows: Set the standard parameter of the s-th influence indicator as ; Through the fixed proportional coefficient Adjustment is performed to obtain the corrected parameter × ; The remaining influence indicators still use their respective standard parameters for loading tests.

[0025] Calculation and screening of computing power demand: After executing the loading of the target city model on the standard terminal device, the occupancy ratio amplitudes of the CPU, memory, disk, and GPU are obtained respectively. Subsequently, using the weighted fusion method, the above-mentioned occupancy data of each item is combined with the corresponding adjustment parameters to calculate the computing power demand value corresponding to the s-th influence indicator. Here, the weighted fusion method can use preset weight factors to reflect the contribution ratio of each indicator to the overall computing power demand.

[0026] Compare the computing power requirement of the s-th impact index obtained by calculation with a preset computing power requirement threshold: If the computing power requirement value is greater than or equal to the threshold, it is considered that the index has a significant impact during the loading process, so it is retained as a limiting factor; otherwise, the index is discarded and not used as a limiting factor in the subsequent loading strategy formulation.

[0027] In one embodiment of the present invention, the calculation formula for the computing power requirement of the s-th impact index is as follows: ; Wherein, represents the computing power requirement of the s-th impact index, , , and respectively represent the CPU occupancy ratio amplitude, the memory occupancy ratio amplitude, the disk occupancy ratio amplitude, and the GPU occupancy ratio amplitude.

[0028] In one embodiment of the present invention, the limiting factors include but are not limited to: the scaling, rotation, or translation of the target map model on the terminal device, and the corresponding change speed of the target map model; on the terminal device, the visualization angle and visualization depth of the target map model.

[0029] In one embodiment of the present invention, when the user quickly zooms in on the three-dimensional map model through a two-finger pinch gesture, the standard scaling change speed of the map model is used as the standard parameter of the corresponding limiting factor.

[0030] In one embodiment of the present invention, when the user quickly rotates the three-dimensional map model through a finger sliding operation, the standard rotation change speed of the map model is used as the standard parameter of the corresponding limiting factor.

[0031] In one embodiment of the present invention, when the user quickly moves the three-dimensional map model by dragging, the standard translation change speed of the map model is used as the standard parameter of the corresponding limiting factor.

[0032] In one embodiment of the present invention, the visualization angle of the three-dimensional map model represents the allowable rotation angle range of the three-dimensional map model on the horizontal plane.

[0033] In one embodiment of the present invention, the visualization depth of the three-dimensional map model represents the allowable rotation angle range of the three-dimensional map model on the vertical plane.

[0034] In one embodiment of the present invention, clustering M types of terminal devices based on performance index parameters includes: The performance index parameters include: CPU computing power index , GPU computing power index , memory performance index , storage performance metrics and rendering requirement metrics ; CPU computing power metrics include: CPU main frequency , number of CPU cores , CPU cache size and CPU floating-point operation parameters ; GPU computing power metrics include: GPU floating-point operation parameters , video memory bandwidth and video memory capacity ; Memory performance metrics include: memory capacity and memory bandwidth ; Storage performance metrics include: number of input / output operations per second and data read and load speed ; Rendering requirement metrics include: monitor resolution , monitor size , monitor refresh rate and monitor color depth ; Based on the performance metric parameters of the m-th type of terminal device, fusion is performed to obtain the loading score of the m-th type of terminal device; based on the loading score, K-clustering is performed on M types of terminal devices to obtain N types of characteristic terminal devices; where 1 ≤ m ≤ M and m is a positive integer.

[0035] Specifically, to accurately reflect the computing power and display capabilities required by the terminal device during the loading of the target city model, the present invention defines five types of performance metric parameters, namely: CPU computing power metrics: including CPU main frequency, number of CPU cores, CPU cache size, and CPU floating-point operation parameters. Among them, the CPU main frequency reflects the running speed of the processor, the number of CPU cores determines the parallel processing ability, and the cache size and floating-point operation parameters affect the numerical calculation and data processing efficiency.

[0036] GPU computing power metrics: including GPU floating-point operation parameters, video memory bandwidth, and video memory capacity. These metrics reflect the capabilities of the graphics processing unit in image rendering, graphics calculation, and data transmission, and are important bases for judging the graphics display performance of the device.

[0037] Memory performance metrics: including memory capacity and memory bandwidth. The memory capacity directly affects data storage and access, and the memory bandwidth determines the data transmission rate, which plays an important role in real-time loading and image processing.

[0038] Storage performance metrics: including the number of input / output operations per second and the data reading and loading speed. The response speed and data transmission capacity of storage devices have a direct impact on the loading of large amounts of map model data.

[0039] Rendering requirement metrics: including monitor resolution, monitor size, monitor refresh rate, and monitor color depth. These metrics are directly related to the display capabilities of the terminal device and determine the visual performance of the loaded model.

[0040] Fusion of performance metric parameters and calculation of loading scores: For each type of terminal device (denoted as the m-th device, where 1 ≤ m ≤ M), by obtaining its various performance metric parameters, a comprehensive evaluation of the device can be performed. In specific implementation, the present invention uses a weighted fusion method to integrate the above-mentioned metric parameters, thereby calculating a comprehensive loading score. During the fusion process, weight factors for each metric can be set so that the contribution of each metric to the loading score can reflect its impact on performance during actual loading. For example, for graphics-intensive tasks, the weights of GPU computing power and rendering requirement metrics can be appropriately increased; while for data processing-intensive tasks, it may focus on CPU and memory performance metrics. The result after weighted fusion is the loading score of the m-th terminal device, which can quantify the comprehensive performance of the terminal device when loading the target city model, thereby providing a basis for subsequent clustering.

[0041] Device clustering based on loading scores: By comparing the loading scores of all M types of terminal devices, the present invention uses a clustering algorithm (such as the K-means clustering algorithm or other appropriate clustering methods) to group these devices. Through the clustering operation, terminal devices with similar performance can be grouped into one category, forming N types of representative characteristic terminal device categories. Each type of characteristic terminal device reflects the common characteristics of a group of devices when loading the model, which helps to, in the subsequent formulation of loading strategies, according to the performance differences of each type of device, allocate different loading levels and limiting factors, thereby achieving the goal of multi-terminal adaptive loading. By refining multiple metrics such as CPU, GPU, memory, storage, and rendering requirements, it fully reflects the actual computing power and display performance of each terminal device when executing the loading task.

[0042] In an embodiment of the present invention, based on the performance metric parameters of the m-th terminal device for fusion, the loading score of the m-th terminal device is obtained, and the calculation formula is as follows: ; ; ; ; ; ; wherein, represents the loading score of the m-th type of terminal device, represents the CPU computing power index score, represents the GPU computing power index score, represents the memory performance index score, represents the storage performance index score, represents the rendering requirement index score, respectively represent the first to fifth weights, and their sum value is 1, and all are not 0; and respectively represent the sixth to seventh weights, and and their sum value is 1, and all are not 0; and respectively represent the eighth to ninth weights, and and their sum value is 1, and all are not 0; and respectively represent the tenth to eleventh weights, and and their sum value is 1, and all are not 0; respectively represent the twelfth to fifteenth weights, and their sum value is 1, and all are not 0; respectively represent the influence weights of the CPU computing power index score, the GPU computing power index score, the memory performance index score, the storage performance index score, and the rendering requirement index score.

[0043] In an embodiment of the present invention, based on the loading score, K clustering is performed on M types of terminal devices to obtain N types of characteristic terminal devices, including: Step 100, randomly select K terminal devices from the loading scores of all terminal devices as the initial clustering centers; Step 200, for the m-th terminal device, calculate the difference in the loading score from each clustering center, and assign the m-th terminal device to the clustering center with the smallest distance; Step 300, for each clustering Gn, update the clustering center based on the mean value of the loading scores of all terminal devices within the clustering Gn; Step 400, repeat Step 200 and Step 300 until a preset number of times is reached to obtain N clusters.

[0044] In an embodiment of the present invention, a feature terminal device of the nth type loads a target map model with a random loading level and a limiting factor, including: for the feature terminal device of the nth type, performing loading of the target map model with a random loading level and random parameters of S limiting factors.

[0045] For example, the feature terminal device of the nth type loads the target map model based on the 3rd loading level, 40% of the standard scaling change speed, 50% of the standard rotation change speed, 80% of the standard translation change speed, a visualization angle from 45 degrees to 225 degrees, and a visualization angle from 45 degrees to 225 degrees.

[0046] In an embodiment of the present invention, for the smoothness analysis of a set of loaded images, it includes: The set of loaded images includes U loaded images, where U = T / t; wherein, T represents a preset time period, and t represents a fixed time interval; Performing edge detection on the u-th loaded image to obtain the feature edges in the u-th loaded image; where 1 ≤ u ≤ U and u is a positive integer; Determine the intersection points of any multiple feature edges in the u-th loaded image, and obtain the color components of the intersection points and the color components of the pixel points in the eight-neighborhood to construct a corresponding matching matrix for the intersection points; Based on the matching matrix, calculate the Euclidean distance between the matching matrix of any one intersection point in the u-th loaded image and the matching matrices of all intersection points in the (u + 1)-th loaded image to obtain a matching score; If the matching score is greater than or equal to a preset threshold, the matching of the corresponding intersection points in the u-th loaded image and the (u + 1)-th loaded image is successful; Respectively obtain the row and column numbers of the successfully matched intersection points in the u-th loaded image and the row and column numbers in the (u + 1)-th loaded image, and calculate the ratio of the Euclidean distance to the number of successfully matched intersection points to obtain the local fluidity parameter of the u-th loaded image and the (u + 1)-th loaded image; Obtain the local fluidity parameters of each group of adjacent loaded images, and calculate the variance of the U - 1 local fluidity parameters to obtain the smoothness score of the set of loaded images.

[0047] In an embodiment of the present invention, the matching matrix includes: Confirm the color space of the loaded image, which is set to the RGB color space in this embodiment; the matching matrix is a 3×3 matrix; respectively obtain the RGB color components of the intersection points to construct an RGB color component vector; use it as the central element of the matching matrix; similarly obtain the color components of the pixel points in the eight-neighborhood of the intersection points to construct corresponding vectors; then fill them into the matching matrix to obtain the matching matrix.

[0048] Specifically, it is a method for analyzing the smoothness of a continuously collected set of loading images during the process of loading a target map model. Its purpose is to objectively evaluate the smoothness of the screen switching and display effect presented by the terminal device during the loading process, providing a quantitative basis for subsequent loading optimization, including: Composition of the loading image set: The loading image set contains U loading images, where U is determined by the formula U = T / t. Here, T is a preset monitoring time period, and t is a fixed image acquisition time interval. The determined value of U discretely samples the loading process in the time dimension, thereby forming a series of continuous loading images to reflect the dynamic changes during the loading process.

[0049] Edge detection to extract feature edges: For each image in the loading image set (denoted as the u-th image, 1 ≤ u ≤ U), an edge detection algorithm is used to extract the feature edges in the image. Edge detection techniques can choose common Canny algorithms or other appropriate detection methods to obtain clear feature edge information.

[0050] Determination of intersection points and construction of a matching matrix: In each loading image, for the multiple detected feature edges, the intersection points of these edges are further determined. For each intersection point, its own color components and the color components of the pixels in its eight-neighborhood are respectively collected. A corresponding matching matrix is constructed using these color information, which can reflect the local color features of the intersection point in the image and its environmental information, providing a basis for subsequent cross-image matching.

[0051] Cross-image matching and calculation of matching scores: To compare the smoothness between consecutive loading images, for an arbitrarily selected intersection point in the u-th image, its matching matrix is compared with the matching matrices formed by all intersection points in the (u + 1)-th image, and the Euclidean distance between the two is calculated to obtain the matching score. When the matching score is greater than or equal to the preset threshold, the intersection points in the u-th and (u + 1)-th images corresponding to the intersection point are considered the same points.

[0052] Calculation of local fluidity parameters: For the successfully matched intersection points, their row and column coordinates in the u-th image and the (u + 1)-th image are respectively recorded. Using these coordinate data, the corresponding Euclidean distance is calculated, and the ratio of this distance to the number of successfully matched intersection points is defined as the local fluidity parameter between the two images.

[0053] Determination of the overall smoothness score: Statistical analysis is performed on the local fluidity parameters (a total of U - 1) of adjacent image pairs in the loaded image set, and their variance is calculated. This variance serves as the smoothness score for the entire loaded image set: A smaller variance indicates that the local fluidity parameters are relatively consistent, indicating a smooth loading process and smooth switching; a larger variance reflects significant fluctuations during the loading process, which may lead to display stuttering or an inconsistent visual effect.

[0054] Through the above steps, the continuity and smoothness of the loaded image set can be effectively reflected. The analysis results can be used as input parameters for subsequent loading optimization algorithms to help the system dynamically adjust the loading strategy and improve the overall smoothness of multi - terminal display.

[0055] In an embodiment of the present invention, for example, the first loaded image and the second loaded image include 3 successfully matched intersection points, and the calculated Euclidean distances are 0.2, 0.3, and 0.4 respectively. Then the corresponding local fluidity parameter is the local fluidity parameter 0.3.

[0056] In an embodiment of the present invention, obtaining the target loading scheme for the nth type of characteristic terminal device in combination with the optimization algorithm includes: Step 71, for the nth type of characteristic terminal device, R loading schemes are initialized and generated. Each loading scheme includes a random loading level and random parameters of S limiting factors; The fitness function includes: The smoothness score of the loaded image set obtained by the nth type of characteristic terminal device executing the rth loading scheme is used as the fitness value of the rth loading scheme; where 1 ≤ r ≤ R, and r is a positive integer; Step 72, based on the fitness values of the loading schemes, the R loading schemes are sorted from smallest to largest to obtain a characteristic sorting; Step 73, a preset number of loading schemes are retained from the front to the back according to the characteristic sorting, and the remaining loading schemes are updated as parents through crossover or mutation; Step 74, repeat Step 72 and Step 73 for a preset number of times, and output the loading scheme ranked 1 in the corresponding characteristic sorting as the target loading scheme for the nth type of characteristic terminal device.

[0057] Specifically, the loading scheme optimization method based on the optimization algorithm aims to provide the best loading strategy for the nth type of characteristic terminal device during the execution of the target map model loading process. This method uses randomly generated candidate loading schemes and gradually optimizes the loading scheme through steps such as fitness evaluation, sorting, elite selection, and crossover mutation, and finally outputs the optimal target loading scheme.

[0058] Initialization of candidate loading schemes: For the nth type of feature terminal device, first randomly initialize and generate R loading schemes. Each loading scheme consists of two parts: a loading level and a restriction factor parameter. Loading level: It refers to the accuracy level used when loading the target map model, and this level is randomly selected from the preset K loading levels; Restriction factor parameter: Corresponding to S items of restriction factors, the parameters of each item are determined randomly. This random generation method can construct diverse candidate schemes within a large parameter space, providing rich search starting points for subsequent optimization.

[0059] Construction of the fitness function: After each candidate loading scheme is executed, based on the set of loading images generated after the nth type of feature terminal device loads the target map model, calculate its smoothness score. This smoothness score serves as the fitness value of this loading scheme. The fitness function uses the smoothness score of the set of loading images (such as the statistical results of the aforementioned edge matching and local mobility parameters) as an evaluation index to reflect the pros and cons of this loading scheme in achieving a smooth loading effect.

[0060] In this embodiment, a sorting method from small to large is adopted, that is, the loading scheme with a lower score (or the one that better meets the preset evaluation criteria) is considered more ideal.

[0061] Sorting and selection of loading schemes: For the R generated loading schemes, sort them according to their respective fitness values to form a feature sorting from excellent to inferior. The sorted result is used to distinguish the pros and cons of the loading schemes, providing a basis for retaining some excellent schemes (elite schemes) subsequently. Preset to retain a certain number of loading schemes as the elite set to ensure that excellent features are retained during the iteration process.

[0062] Crossing and mutation update: For the loading schemes that are not retained, use them as parents and generate new loading schemes through crossing and mutation operations. Crossing operation: By combining the loading levels and restriction factor parameters in two or more candidate loading schemes to form a new loading scheme, thus inheriting the excellent features of the parent schemes; Mutation operation: Randomly fine-tune some parameters in a loading scheme to increase the diversity of the search space and prevent the algorithm from falling into a local optimum. This step ensures that new schemes are generated in each generation, while retaining some proven excellent schemes to improve the overall optimization effect.

[0063] Iterative optimization: Repeat step 72 and step 73. After a preset number of iterative updates, the candidate loading scheme population gradually converges to the optimal direction. After each iteration, re-sort according to the newly calculated fitness value, update the elite set and perform a new round of crossing and mutation. After several iterations, the overall fitness value of the loading schemes tends to be stable, thereby screening out the optimal loading scheme.

[0064] Determination of the target loading scheme: When the iteration reaches the preset number of times, select the loading scheme ranked first (with the best fitness value) from the final feature ranking. This scheme is used as the target loading scheme for the feature terminal device of the nth type. This target loading scheme determines the specific loading level and the parameters of each constraint factor, and is the optimal strategy obtained through automatic search and adaptive adjustment during multiple optimization processes.

[0065] The present invention achieves the following effects: It can automatically screen out the most suitable scheme for the current terminal device performance and loading scenario from a large number of candidate loading schemes; it expands the search space through crossover and mutation operations, effectively avoiding the local optimum problem and improving the overall fluency of the loading process.

[0066] In an embodiment of the present invention, performing a loading target map model on a target terminal device includes: Obtain the performance index parameters of the target terminal device and calculate the loading score of the target terminal device; Cluster the target terminal device based on the loading score of the target terminal device to obtain the type of the feature terminal device to which the target terminal device belongs; Based on the type of the feature terminal device of the target terminal device, obtain the corresponding target loading scheme, and perform the loading target map model based on the target loading scheme.

[0067] Specifically, for the dynamic loading method of loading a target map model for a target terminal device, the core idea is to calculate its loading score according to the performance index parameters of the target terminal device, and determine the feature category to which the device belongs through clustering, and then perform the loading operation according to the pre-optimized loading scheme corresponding to the category. Acquisition of target terminal device performance parameters and calculation of loading score In this embodiment, in order to evaluate the actual performance of the target terminal device when loading the target map model, the system first obtains various performance index parameters of the terminal device. According to the above performance indexes, the system uses the previously set weighted fusion method to comprehensively calculate each index, so as to obtain the "loading score" of the target terminal device. The loading score can quantitatively represent the comprehensive computing power and display ability of the device when executing the target map model loading task, providing a quantitative basis for subsequent clustering. After obtaining the loading score of the target terminal device, the system calculates the difference between the loading score and the average loading scores of N clusters, and assigns it to the cluster corresponding to the smallest difference. Each type of characteristic terminal device corresponds to a target loading scheme obtained through the foregoing optimization algorithm. The target loading scheme includes: the loading level, that is, the accuracy or level of detail adopted when loading the target map model; the specific parameter settings of each limiting factor, reflecting the control of the device computing power requirements (such as the change speed limits of zooming, rotation, and translation, etc.). Once the target terminal device is classified, the target loading scheme corresponding to its category is selected as the loading scheme of the device. The advantage of doing so is that for devices with different performance levels, the most suitable loading strategy for their performance characteristics can be pre-designed to achieve the optimization of the loading effect and fluency. After obtaining the corresponding target loading scheme, the system then executes the operation of loading the target map model on the target terminal device according to the scheme.

[0068] The above describes the embodiments of this embodiment, but this embodiment is not limited to the above specific implementation manners. The above specific implementation manners are only illustrative and not restrictive. Under the inspiration of this embodiment, those of ordinary skill in the art can also make many forms, all of which fall within the protection scope of this embodiment.

Claims

1. A smart city digital twin map management system for multiple terminals, characterized in that, Including: A model grading module, configured to divide a target map model into K loading levels; wherein, from the first to the Kth loading level, the loading accuracy of the target map model increases in a gradient manner; A loading limit module, configured to determine S limiting factors based on prior knowledge; A device clustering module, configured to determine a set of terminal devices for the target map model, obtain performance index parameters of M types of terminal devices in the set of terminal devices, and cluster the M types of terminal devices based on the performance index parameters to obtain N types of characteristic terminal devices; An intelligent loading module, configured to, within a preset time period, allocate the nth type of characteristic terminal device to load the target map model with a random loading level and a limiting factor, and synchronously obtain a corresponding set of loaded images at fixed time intervals; An intelligent recognition module, configured to perform a fluency analysis on the set of loaded images to obtain a fluency score corresponding to the nth type of characteristic terminal device; 1 ≤ n ≤ N, and n is a positive integer; A loading optimization module, configured to, based on the fluency score, combine an optimization algorithm to obtain a target loading scheme for the nth type of characteristic terminal device; A multi-terminal loading management module, configured to execute loading the target map model on the target terminal device based on the target loading scheme of each type of characteristic terminal device.

2. The digital twin map management system for a smart city oriented to multiple terminals according to claim 1, wherein Loading accuracy, including: modeling the target city model through the triangulation method, where a number of sampling points are included in the unit area of the target map model; the difference in the number of sampling points in the unit area between the k-th loading level and the (k + 1)-th loading level of the target map model is ; where both k and k + 1 are indexes of K loading levels.

3. The smart city digital twin map management system for multiple terminals according to claim 2, characterized in that, Determine S influencing indicators for the computing power requirements of the terminal device to execute loading the target city model based on prior knowledge, and use each influencing indicator as a limiting factor; Wherein, the method for determining the influencing indicators is as follows: Determine the standard parameters of each influencing indicator; Set a standard terminal device; Execute loading the target map model for the standard terminal device, including: adjusting the standard parameters of the sth influencing indicator with a fixed proportionality coefficient and keeping the standard parameters of the remaining influencing indicators unchanged; Obtain the CPU occupancy ratio amplitude, memory occupancy ratio amplitude, disk occupancy ratio amplitude, and GPU occupancy ratio amplitude of the standard terminal device, and obtain the computing power requirement of the sth influencing indicator through weighted fusion; if the computing power requirement of the sth influencing indicator is greater than or equal to a preset computing power requirement threshold, retain the sth influencing indicator, otherwise discard the sth influencing indicator; wherein, 1 ≤ s ≤ S, and s is a positive integer.

4. The management system of the digital twin map of a smart city for multiple terminals according to claim 3, characterized in that, Clustering the M types of terminal devices based on the performance index parameters, including: The performance index parameters include: CPU computing power index, GPU computing power index, memory performance index, storage performance index, and rendering requirement index; The CPU computing power index includes: CPU main frequency, CPU core number, CPU cache size, and CPU floating-point operation parameters; The GPU computing power index includes: GPU floating-point operation parameters, video memory bandwidth, and video memory capacity; The memory performance index includes: memory capacity and memory bandwidth; The storage performance index includes: number of input / output operations per second and data read and load speed; The rendering requirement index includes: monitor resolution, monitor size, monitor refresh rate, and monitor color depth; Fuse based on the performance index parameters of the m-th type of terminal device to obtain the loading score of the m-th type of terminal device; perform K-clustering on the M types of terminal devices based on the loading score to obtain N types of characteristic terminal devices; where 1 ≤ m ≤ M and m is a positive integer.

5. The digital twin map management system for a smart city facing multiple terminals according to claim 4, wherein, Allocate the n-th type of characteristic terminal device to load the target map model with a random loading level and a limiting factor, including: for the n-th type of characteristic terminal device, execute loading the target map model with a random loading level and random parameters of S limiting factors.

6. The management system of a smart city digital twin map for multiple terminals according to claim 5, characterized in that, Perform fluency analysis on the loaded image set, including: The loaded image set includes U loaded images, where U = T / t; T represents a preset time period and t represents a fixed time interval. Perform edge detection on the u-th loaded image to obtain the characteristic edges in the u-th loaded image; where 1 ≤ u ≤ U and u is a positive integer. Determine the intersection points of any number of characteristic edges in the u-th loaded image, and obtain the color components of the intersection points and the color components of the pixel points in the eight-neighborhood to construct the matching matrix corresponding to the intersection points. Based on the matching matrix, calculate the Euclidean distance between the matching matrix of any one intersection point in the u-th loaded image and the matching matrix of all intersection points in the (u + 1)-th loaded image to obtain the matching score. If the matching score is greater than or equal to the preset threshold, the matching of the corresponding intersection points in the u-th loaded image and the (u + 1)-th loaded image is successful. Respectively obtain the row and column numbers of the successfully matched intersection points in the u-th loaded image and the row and column numbers in the (u + 1)-th loaded image, and calculate the ratio of the Euclidean distance to the number of successfully matched intersection points to obtain the local fluidity parameter of the u-th loaded image and the (u + 1)-th loaded image. Obtain the local fluidity parameters of each group of adjacent loaded images, and calculate the variance of U - 1 local fluidity parameters to obtain the fluency score of the loaded image set.

7. A smart city digital twin map management system for multiple terminals according to claim 6, characterized in that, Combine the optimization algorithm to obtain the target loading scheme of the n-th type of characteristic terminal device, including: Step 71, for the n-th type of characteristic terminal device, initialize and generate R loading schemes, and each loading scheme includes random loading levels and random parameters of S limiting factors. The fitness function includes: using the fluency score of the loaded image set obtained by the n-th type of characteristic terminal device executing the r-th loading scheme as the fitness value of the r-th loading scheme; where 1 ≤ r ≤ R and r is a positive integer. Step 72, sort the R loading schemes from small to large based on the fitness values of the loading schemes to obtain the characteristic sorting. Step 73, retain a preset number of loading schemes from front to back for the characteristic sorting, and update the remaining loading schemes as parents through crossover or mutation. Step 74, repeat Step 72 and Step 73 until the preset number of times, and output the loading scheme ranked 1 in the corresponding characteristic sorting as the target loading scheme of the n-th type of characteristic terminal device.

8. A multi-terminal-oriented smart city digital twin map management system according to claim 7, characterized in that, Execute loading the target map model on the target terminal device, including: Obtain the performance index parameters of the target terminal device, and calculate the loading score of the target terminal device. Cluster the target terminal device based on the loading score of the target terminal device to obtain the type of the characteristic terminal device to which the target terminal device belongs; Based on the type of the characteristic terminal device of the target terminal device, obtain the corresponding target loading scheme, and execute the loading of the target map model based on the target loading scheme.

Images