A method and system for visualizing construction management

Abstract

The application provides a kind of visual construction management method and system, belong to construction management field.Therein, the method includes: the standardization processing of the multi-source laminated data in construction display area is carried out, and the standardization data structure is constructed;The laminated feature vector of standardization data structure is extracted;According to laminated feature vector, the conflict field calculation of any two laminated in construction display area is carried out, and the laminated conflict intensity distribution diagram is generated;The conflict field feature of laminated conflict intensity distribution diagram is extracted, and the conflict clustering output multiclass conflict level area is carried out;The differential fusion strategy analysis of multiclass conflict level area is carried out, and the multiclass visual fusion expression is obtained, according to multiclass visual fusion expression, visual output management is carried out in construction display area.The application carries out laminated conflict quantitative analysis and differential adaptive fusion, so that multi-source laminated data is clearly and accurately expressed in the same construction display area, and information conflict and expression distortion are eliminated.

Description

Technical Field

[0001] This invention relates to the field of construction management, and in particular to a visual construction management method and system. Background Technology

[0002] In construction management, visualization technology has been widely used in project progress monitoring, resource scheduling, and on-site situation presentation. With the deep integration of multi-source heterogeneous data such as Building Information Modeling (BIM), Geographic Information Systems (GIS), and IoT sensor data in construction management, the types of data and data layers that construction visualization systems need to support are becoming increasingly complex. Multiple layers, such as topography, building structure models, construction progress markings, equipment location information, and safety monitoring data, often need to be overlaid and presented simultaneously within the same construction display area.

[0003] However, existing construction visualization methods typically employ simple layer overlay or fixed priority sorting for rendering output when displaying multi-source, multi-layered data. When multiple data layers overlap within the same spatial area, their spatial occupancy overlaps, their information density varies significantly, and their semantic importance differs. Simple overlay methods struggle to effectively coordinate the display relationships between layers, easily leading to problems such as key information being obscured, chaotic information stacking in high-density areas, and the inability to distinguish the primary and secondary information of data at different semantic levels. Ultimately, this results in distorted visualization, severely impacting construction managers' accurate assessment of the site situation and their decision-making efficiency. Summary of the Invention

[0004] This invention addresses the technical problem in existing construction visualization scenarios where spatial overlap and information conflict in multi-source overlay data within the same construction display area lead to distorted visualization representations. It provides a visualization construction management method and system to solve this problem.

[0005] The technical solution of the present invention to solve the above-mentioned technical problems is as follows:

[0006] In a first aspect, the present invention provides a visual construction management method, comprising: standardizing multi-source overlay data within a construction display area to construct a standardized data structure; extracting overlay feature vectors from the standardized data structure, wherein the overlay feature vectors include space occupancy feature vectors, information density feature vectors, semantic importance feature vectors, and time sensitivity feature vectors; constructing an overlay conflict determination function, wherein the overlay conflict determination function calculates a conflict field for any two overlays within the construction display area based on the overlay feature vectors to generate an overlay conflict intensity distribution map; extracting conflict field features from the overlay conflict intensity distribution map, performing conflict clustering according to the conflict field features to output multiple conflict level regions; performing differentiated fusion strategy analysis on the multiple conflict level regions according to a visual adaptive fusion module to obtain multiple visual fusion expressions, and performing visual output management in the construction display area according to the multiple visual fusion expressions.

[0007] Secondly, the present invention provides a visual construction management system, comprising: a data standardization processing unit for standardizing multi-source overlay data within a construction display area to construct a standardized data structure; an overlay feature extraction unit for extracting overlay feature vectors from the standardized data structure, wherein the overlay feature vectors include space occupancy feature vectors, information density feature vectors, semantic importance feature vectors, and time sensitivity feature vectors; a conflict field calculation unit for constructing an overlay conflict determination function, wherein the overlay conflict determination function performs conflict field calculations on any two overlays within the construction display area based on the overlay feature vectors to generate an overlay conflict intensity distribution map; a conflict clustering and grading unit for extracting conflict field features from the overlay conflict intensity distribution map, performing conflict clustering according to the conflict field features to output multiple conflict level regions; and a visual fusion output unit for performing differentiated fusion strategy analysis on the multiple conflict level regions according to a visual adaptive fusion module to obtain multiple visual fusion expressions, and performing visual output management in the construction display area according to the multiple visual fusion expressions.

[0008] The beneficial effects of this invention are:

[0009] First, the multi-source overlay data within the construction display area is standardized to construct a standardized data structure. This unifies the diverse sources and formats of the overlay data into a comparable and computable data form, laying the data foundation for subsequent feature extraction and conflict analysis. Based on this, overlay feature vectors are extracted from the standardized data structure. These vectors include spatial occupancy, information density, semantic importance, and time sensitivity features. This multi-dimensional feature characterization comprehensively represents the attributes of each overlay in terms of spatial distribution, information carrying capacity, semantic weight, and timeliness, providing a characteristic basis for quantitatively assessing the conflict relationships between overlays. Next, an overlay conflict determination function is constructed. This function calculates the conflict field for any two overlays within the construction display area based on the overlay feature vectors, generating an overlay conflict intensity distribution map. This transforms the qualitative description of the information conflict relationship between overlays into a quantifiable spatial distribution expression, giving the degree of conflict a clear intensity measure within the display area. Then, conflict field features are extracted from the layered conflict intensity distribution map. Conflict clustering is performed based on these features to output multiple conflict level regions. The continuously distributed conflict intensity is divided into discrete regions with clear level differences, providing a basis for subsequent implementation of targeted fusion strategies. Next, a differentiated fusion strategy analysis is conducted on the multiple conflict level regions using the visualization adaptive fusion module to obtain multiple visualization fusion representations. These representations are then managed for visualization output in the construction display area. Appropriate fusion processing methods are applied to regions with different conflict levels, achieving adaptive coordination and optimized representation of the layered data at the display level.

[0010] Through the above technical solution, the multi-source overlay data in the construction display area is sequentially standardized, multi-dimensional feature extracted, conflict intensity quantified, conflict level clustered, and differentiated adaptive fusion is performed, forming a processing link from data standardization to conflict identification and then to fusion output. This enables the multi-source overlay data to be clearly and accurately co-expressed in the same construction display area, effectively eliminating the problem of visualization distortion caused by spatial overlap and information conflict. Attached Figure Description

[0011] Figure 1 A flowchart illustrating a visual construction management method provided by the present invention;

[0012] Figure 2 This is a schematic diagram of the structure of a visual construction management system provided by the present invention.

[0013] In the attached diagram, the components represented by each number are as follows:

[0014] The system includes a data standardization processing unit 11, a layered feature extraction unit 12, a conflict field calculation unit 13, a conflict clustering and hierarchical unit 14, and a visualization fusion output unit 15. Detailed Implementation

[0015] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0016] In the description of this invention, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the stated features. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0017] In the description of this invention, the term "for example" is used to mean "used as an example, illustration, or description." Any embodiment described as "for example" in this invention is not necessarily to be construed as being more preferred or advantageous than other embodiments. The following description is provided to enable any person skilled in the art to make and use the invention. Details are set forth in the following description for purposes of explanation. It should be understood that those skilled in the art will recognize that the invention can be made without using these specific details. In other instances, well-known structures and processes will not be described in detail to avoid obscuring the description of the invention with unnecessary detail. Therefore, the invention is not intended to be limited to the embodiments shown, but is consistent with the broadest scope of the principles and features disclosed herein.

[0018] Example 1, as Figure 1 As shown, this embodiment of the invention provides a visual construction management method, including:

[0019] S1. Standardize the multi-source overlay data within the construction display area to construct a standardized data structure.

[0020] Specifically, in construction visualization management, the same construction display area typically needs to simultaneously support multiple data layers from different sources and in different formats. These include 3D structural model data layers provided by Building Information Modeling (BIM), topographic data layers provided by Geographic Information Systems (GIS), safety monitoring data layers collected by IoT sensors, progress labeling data layers generated by the construction management platform, equipment location data layers output by the positioning system, environmental meteorological data layers provided by the meteorological system, and on-site image data layers generated by the video surveillance system. The data carried by these multiple data layers constitutes multi-source overlay data within the construction display area. These multi-source overlay data often exhibit significant differences in data format, coordinate system, unit scale, and time stamping specifications. If they are directly overlaid onto the same construction display area for rendering, spatial relationship comparison and information density measurement between different data layers cannot be performed within a unified reference framework. Subsequent overlay conflict analysis will also lose its quantitative basis due to the inconsistency in data representation. Therefore, the first step is to standardize the multi-source overlay data within the construction display area to construct a unified standardized data structure.

[0021] Specifically, standardization processing includes at least the spatial coordinate level, data dimension level, time stamp level, and data format level. At the spatial coordinate level, the different coordinate reference systems used by various data sources in multi-source overlay data are uniformly transformed to the target coordinate system used in the construction display area. For example, an affine transformation matrix can be established to achieve mapping transformation between different coordinate systems, unifying the spatial location expression of each overlay data into the same three-dimensional coordinate space and eliminating spatial misalignment caused by coordinate system differences. At the data dimension level, the attribute values ​​carried by different data sources in multi-source overlay data are normalized. For example, a minimum-maximum normalization method can be used to linearly map each attribute value to a numerical range of 0 to 1, making data attributes from different sources and with different physical meanings numerically comparable in subsequent feature extraction and conflict intensity calculation processes. At the time stamping level, the different time representation formats and time precisions used by various data sources in multi-source overlay data are uniformly aligned to the same time base and time granularity. For example, the timestamps of various data sources can be uniformly converted into the same time format, and time-series data with different sampling frequencies can be aligned to a uniform time resolution through interpolation or resampling. This allows for comparisons at a consistent time scale when evaluating the time sensitivity characteristics of each overlay data. At the data format level, the heterogeneous storage formats of data from various sources in multi-source overlay data are parsed and converted into a unified structured expression. Each standardized data record includes the three-dimensional spatial coordinates of the data point in the target coordinate system, the normalized attribute value, the aligned timestamp, and the unique identifier of the data layer to which it belongs, thus forming a standardized data structure.

[0022] Through the above standardization process, the multi-source overlay data in the construction display area is organized into a standardized data structure with unified format, consistent coordinates, comparable dimensions, and time alignment, providing a standardized data foundation for subsequent processing. It also provides a consistent input data level for the quantitative calculation of the overlay conflict determination function.

[0023] S2. Extract the stacked feature vector of the standardized data structure. The stacked feature vector includes space occupancy feature vector, information density feature vector, semantic importance feature vector, and time sensitivity feature vector.

[0024] Specifically, after standardization processing to obtain a standardized data structure, layer feature vectors that characterize the display characteristics of each layer of data within the construction display area are extracted. This allows the subsequent layer conflict determination function to quantify the conflict relationship between any two layers based on these feature vectors. Before extracting the layer feature vectors, the construction display area is first spatially meshed to obtain a set of three-dimensional mesh cells covering the entire area. Each three-dimensional mesh cell serves as a local spatial unit for subsequent feature extraction, and this set of three-dimensional mesh cells also serves as a unified spatial computation framework for subsequent conflict field calculations. The layer feature vectors include spatial occupancy feature vectors, information density feature vectors, semantic importance feature vectors, and time sensitivity feature vectors, which characterize the display characteristics of each layer from the spatial, informational, semantic, and temporal dimensions, respectively.

[0025] The space occupancy feature vector characterizes the spatial distribution and occupancy level of each layer of data within the construction display area. Specifically, for each local spatial unit, the number of data points falling into that local spatial unit is counted and divided by the volume of that local spatial unit to obtain the corresponding space occupancy density value. If no data points fall into that local spatial unit, the corresponding space occupancy density value is 0. The space occupancy density values ​​of each layer in each local spatial unit are arranged and combined in spatial order to form the space occupancy feature vector of that layer. This space occupancy feature vector, in a vectorized form, completely records the spatial occupancy intensity distribution of each local location of the layer within the construction display area.

[0026] The information density feature vector is used to characterize the spatial distribution characteristics of the amount of information carried by each layer of data within the construction display area. Specifically, for each local spatial unit, the set of normalized attribute values ​​of the data points falling within that local spatial unit is statistically analyzed, and the information entropy of this attribute value set is calculated as the information density value corresponding to that local spatial unit. When calculating the information entropy, since the normalized attribute values ​​range from 0 to 1, this range is divided into K equal intervals, each with a width of 1 / K. The normalized attribute values ​​of each data point within the local spatial unit are categorized according to the interval they fall into. The ratio of the number of data points falling into the i-th interval to the total number of data points within the local spatial unit is calculated as the probability value pi for the i-th interval. Based on the probability values ​​pi for each interval, the information entropy value of the local spatial unit is calculated using the Shannon entropy formula, which involves multiplying each probability value pi by its logarithm, summing the results, and taking the negative value. The base of the logarithm can be, for example, 2. A higher information entropy value indicates a more dispersed distribution of attribute values ​​and a greater information carrying capacity within the local spatial unit. The information density value of each layer in each local spatial unit is arranged and combined in spatial order to form the information density feature vector of that layer. This information density feature vector records the information carrying density distribution of the layer at each local location with the same spatial dimensional structure as the spatial occupancy feature vector.

[0027] Semantic importance feature vectors are used to characterize the semantic priority and importance of each layer of data in construction management decisions. Specifically, a data layer semantic importance mapping table is pre-established. This mapping table assigns a basic semantic importance level value to each data layer identifier based on its functional role in construction management. For example, the safety monitoring data layer, which is directly related to construction safety early warning, has its basic semantic importance level value set to the highest level; the construction progress annotation layer, which is related to schedule control, has its basic semantic importance level value set to the second highest level; and the basic semantic importance level values ​​of the equipment location layer and the terrain layer decrease sequentially. Based on the basic semantic importance level values, further dynamic adjustments are made according to the urgency of the data state in each local spatial unit for each layer. Specifically, for each local spatial unit, a normal threshold range for attribute values ​​is preset for that layer. The ratio of the number of data points in that local spatial unit whose normalized attribute values ​​exceed the upper limit or fall below the lower limit of the normal threshold to the total number of data points in that layer within that local spatial unit is used as the anomaly deviation degree. The anomaly deviation degree is multiplied by the basic semantic importance level value to obtain the semantic importance value corresponding to that local spatial unit. A higher anomaly deviation degree indicates a greater degree of deviation from the normal state of the data in that region, a higher urgency of the information carried in that region, and a correspondingly higher semantic importance value. The semantic importance values ​​of each layer in each local spatial unit are arranged and combined in spatial order to form the semantic importance feature vector of that layer. This semantic importance feature vector simultaneously integrates the global functional positioning of the layer and the semantic information of the local data state.

[0028] The time-sensitivity feature vector is used to characterize the sensitivity and update frequency of each layer's data to time changes. Specifically, based on the aligned timestamp sequence of data points in each layer of the standardized data structure, the data update frequency of that layer is calculated. The data update frequency is the ratio of the number of data updates within a preset statistical time window to the duration of the statistical time window. Based on this, for each local spatial unit, the time interval between the most recent data update timestamp of that layer within that local spatial unit and the current time is calculated. This time interval is input into a preset time-sensitivity decay function to obtain the time sensitivity value corresponding to that local spatial unit. The time-sensitivity decay function can be a negative exponential decay function, and its decay rate parameter is determined by the data update frequency of that layer. A higher update frequency results in a faster decay rate, indicating that the time sensitivity of the layer's data decays more rapidly and the real-time requirements are higher. The time sensitivity values ​​of each layer in each local spatial unit are arranged and combined in spatial order to form the time sensitivity feature vector of that layer. This time sensitivity feature vector records the distribution of the time sensitivity of that layer at each local location.

[0029] Through the extraction of the aforementioned layered feature vectors, the display characteristics of each layered data within the construction display area are quantitatively characterized from four dimensions: space occupancy, information density, semantic importance, and time sensitivity. Furthermore, all four types of feature vectors are extracted based on the same local spatial unit, possessing a unified spatial dimension alignment relationship. This enables the subsequent layered conflict determination function to compare the corresponding features of any two layers at the granularity of each three-dimensional grid unit, providing a complete feature foundation for accurately quantifying the conflict intensity distribution between layers.

[0030] S3. Construct a layer conflict determination function. The layer conflict determination function calculates the conflict field of any two layers in the construction display area based on the layer feature vector and generates a layer conflict intensity distribution map.

[0031] Specifically, after obtaining the feature vectors of each overlay, the conflict relationships between overlays within the construction display area are quantitatively analyzed. Since multiple overlays exist simultaneously within the construction display area, differences in dimensions such as space occupancy, information density, semantic importance, and temporal sensitivity can lead to varying degrees of conflict in display representation. For example, if two overlays both have high space occupancy density within the same spatial area, spatial occupancy conflict will occur; if two overlays both carry a large amount of information within the same area, information stacking conflict will occur; if the semantic importance levels of two overlays differ significantly, it is necessary to distinguish between primary and secondary displays; and if the temporal sensitivity of two overlays differs significantly, there is a need for coordination in update responses. To comprehensively capture these multi-dimensional overlay conflict relationships, an overlay conflict determination function is constructed to calculate the conflict field for any two overlays within the construction display area, generating an overlay conflict intensity distribution map.

[0032] The overlay conflict determination function is constructed as follows: It takes as input the spatial overlap conflict factor, information density conflict factor, semantic importance conflict factor, and time-sensitive conflict factor of any pair of overlays on the same 3D grid cell. The output is the local conflict intensity distribution of the overlay pair on that 3D grid cell, characterizing the overall conflict severity of the overlay pair at that spatial location. The four conflict factors are calculated based on the eigenvalues ​​of the corresponding overlay feature vectors on that 3D grid cell, quantifying the conflict contribution of the overlay pair in each dimension of the 3D grid cell. The overlay conflict determination function uses a weighted fusion method to integrate the four conflict factors and output a scalar value as the local conflict intensity distribution of the overlay pair on that 3D grid cell. Specifically, a weight coefficient is assigned to each conflict factor, and the sum of the products of each conflict factor and its corresponding weight coefficient yields the local conflict intensity distribution. The weight coefficients of each conflict factor can be preset and allocated according to the actual importance of each conflict dimension in the construction management scenario. For example, in a scenario where construction safety monitoring is prioritized, the weight coefficient of the spatial overlap conflict factor can be set to 0.3, the weight coefficient of the information density conflict factor can be set to 0.2, the weight coefficient of the semantic importance conflict factor can be set to 0.35, and the weight coefficient of the time sensitivity conflict factor can be set to 0.15, so that the conflict contribution of the semantic importance dimension and the spatial overlap dimension occupies a larger proportion in the fusion result. Alternatively, adaptive calculation can be performed according to the statistical distribution of the feature vectors of each layer in the current construction display area. For example, when the semantic importance difference of each layer in the construction display area is more prominent, the weight coefficient of the semantic importance conflict factor can be increased accordingly.

[0033] After the overlay conflict determination function is constructed, for each pair of overlays within the construction display area, all three-dimensional mesh cells within the construction display area are traversed. On each three-dimensional mesh cell, the local conflict intensity distribution of the overlay pair is calculated based on its overlay feature vector and the overlay conflict determination function. After traversing all overlay pairs, for each three-dimensional mesh cell, a comprehensive conflict intensity value is generated based on the local conflict intensity distribution of all overlay pairs on that three-dimensional mesh cell. The comprehensive conflict intensity values ​​of all three-dimensional mesh cells within the construction display area are organized according to spatial location to generate an overlay conflict intensity distribution map covering the entire construction display area. This map records the comprehensive conflict intensity value of overlays at each location within the construction display area, indexed by three-dimensional mesh cells. A higher conflict intensity value indicates a more severe display conflict between overlays at that location.

[0034] By using the layered conflict intensity distribution map, the layered conflict status at each location within the construction display area is quantified into measurable intensity values, providing a spatial basis for conflict quantification for subsequent conflict clustering based on conflict field characteristics, classifying conflict level areas, and formulating differentiated fusion strategies.

[0035] In one feasible implementation, the layer conflict determination function calculates the conflict field for any two layers within the construction display area based on the layer feature vector, generating a layer conflict intensity distribution map. The method further includes:

[0036] S31. The construction display area is spatially gridded to obtain a set of three-dimensional grid cells;

[0037] S32. For each three-dimensional mesh unit in the set of three-dimensional mesh units, calculate the local conflict factor for the stacked feature vectors of any two stacks to obtain the spatial overlap conflict factor, information density conflict factor, semantic importance conflict factor and time sensitivity conflict factor.

[0038] S33. Based on the stacked conflict determination function, the input spatial overlap conflict factor, information density conflict factor, semantic importance conflict factor and time sensitivity conflict factor are fused into a multi-factor system, and the local conflict intensity distribution corresponding to each three-dimensional mesh unit is output.

[0039] S34. Normalize and average the local conflict intensity distribution of each three-dimensional mesh element to obtain the stacked conflict intensity distribution map.

[0040] Specifically, firstly, the construction display area is spatially meshed to obtain a set of three-dimensional mesh units. Specifically, the construction display area is uniformly divided along the three coordinate axes according to a preset meshing step size, resulting in a set of three-dimensional mesh units covering the entire construction display area. Each three-dimensional mesh unit corresponds to a fixed-size spatial volume element within the construction display area. This set of three-dimensional mesh units provides a spatial computational framework for subsequent unit-by-unit conflict factor calculations at a uniform spatial granularity. The preset meshing step size is set based on the spatial scale of the construction display area and the required computational precision. For example, the span of the construction display area along each coordinate axis can be divided into N equal parts, with the length of each part serving as the meshing step size along that coordinate axis. A larger value for N results in a smaller meshing step size, finer division of the three-dimensional mesh units, higher spatial resolution for conflict field calculation, and a correspondingly larger computational load.

[0041] Then, for each 3D mesh cell in the 3D mesh cell set, local conflict factors are calculated for the stacked feature vectors of any two stacks, yielding spatial overlap conflict factors, information density conflict factors, semantic importance conflict factors, and time sensitivity conflict factors. The stacked feature vectors of each stack are extracted based on the same local spatial cell, which is each 3D mesh cell in the 3D mesh cell set. Therefore, the conflict factors can be calculated directly by reading the corresponding feature values ​​of any two stacks on the same 3D mesh cell. Any two stacks within the construction display area form any pair of stacks, constituting one stack pair. For example, the safety monitoring data layer and the progress annotation data layer form one stack pair, the safety monitoring data layer and the equipment location data layer form another stack pair, and the 3D structural model data layer and the terrain data layer form yet another stack pair.

[0042] The spatial overlap conflict factor is used to quantify the degree of spatial overlap between any pair of stacked layers in the same 3D mesh cell. It is calculated as follows: for any pair of stacked layers, the spatial occupancy density values ​​of both layers in the 3D mesh cell are read, and the spatial overlap conflict factor is obtained by multiplying the two spatial occupancy density values. Spatial overlap conflict only occurs when both stacked layers occupy space within the same 3D mesh cell. When the spatial occupancy density value of either layer in the 3D mesh cell is 0, the product is 0, and the spatial overlap conflict factor is 0, indicating that there is no spatial overlap conflict in the 3D mesh cell. When both stacked layers have high spatial occupancy densities in the 3D mesh cell, the product increases accordingly, indicating a higher degree of spatial occlusion and more severe spatial overlap conflict between the two stacked layers in the 3D mesh cell.

[0043] The information density conflict factor is used to quantify the degree of information stacking in any pair of stacks within the same 3D mesh cell. It is calculated as follows: for any pair of stacks, the information entropy values ​​of both stacks within the 3D mesh cell are read, and the two entropy values ​​are multiplied to obtain the information density conflict factor. Information stacking conflict only occurs when both stacks carry a certain amount of information within the same 3D mesh cell. When the information entropy value of either stack in the 3D mesh cell is 0, the product is 0, and the information density conflict factor is 0, indicating that there is no information stacking conflict in that 3D mesh cell. When the information entropy values ​​of both stacks in the 3D mesh cell are high, the product increases accordingly, indicating that both stacks simultaneously carry a large amount of information within the 3D mesh cell, and the information stacking conflict is more severe.

[0044] The semantic importance conflict factor is used to quantify the degree of difference in semantic priority between any pair of overlays on the same 3D mesh cell. It is calculated as follows: for any pair of overlays, the semantic importance values ​​of each overlay on the 3D mesh cell are read, the difference between the two semantic importance values ​​is calculated, and the absolute value is taken to obtain the semantic importance conflict factor. Semantic importance conflict reflects the difference in semantic priority between two overlays. The greater the difference, the more disparate the primary and secondary relationships between the two overlays, and the more necessary it is to prioritize the high semantic importance overlay in the display representation to prevent it from being occluded or interfered with by the low semantic importance overlay.

[0045] The time-sensitivity conflict factor is used to quantify the degree of difference in time response requirements between any pair of overlays on the same 3D mesh cell. The calculation method is as follows: for any pair of overlays, the time sensitivity values ​​of both overlays on the 3D mesh cell are read, the difference between the two time sensitivity values ​​is calculated, and the absolute value is taken to obtain the time-sensitivity conflict factor. The time-sensitivity conflict reflects the difference in response requirements of two overlays to time changes. The greater the difference, the more likely one overlay requires high-frequency real-time updates while the other updates slowly. This necessitates differentiated coordination in display update strategies to avoid the display of the low-update-frequency overlay interfering with the real-time information presentation of the high-update-frequency overlay.

[0046] After obtaining the spatial overlap conflict factor, information density conflict factor, semantic importance conflict factor, and time sensitivity conflict factor for each 3D mesh cell, the numerical ranges of the four conflict factors differ due to the different calculation methods. Therefore, the four conflict factors are normalized by dividing each conflict factor by its maximum value across all 3D mesh cells, so that the value ranges of the four conflict factors are mapped to the interval between 0 and 1. This ensures that the weight coefficients can accurately reflect the actual importance of each conflict dimension when the subsequent input to the stacked conflict determination function is weighted and fused.

[0047] Next, the input spatial overlap conflict factor, information density conflict factor, semantic importance conflict factor, and time sensitivity conflict factor are fused using the overlay conflict determination function to output the local conflict intensity distribution for each 3D mesh cell. Specifically, the four types of conflict factors after normalization are input into the aforementioned overlay conflict determination function, which performs a weighted summation according to the weight coefficients corresponding to each conflict factor, outputting the local conflict intensity distribution of the overlay pair on the 3D mesh cell. For each 3D mesh cell, all overlay pairs within the construction display area are traversed, and the local conflict intensity distribution of each overlay pair on the 3D mesh cell is calculated in the above manner.

[0048] Next, the local conflict intensity distribution of each 3D mesh element is normalized and averaged to obtain the stacked conflict intensity distribution map. Specifically, the 3D mesh element is denoted as... Assume there are M layers in the construction display area, denoted as follows: , ,..., Any two stacked layers and Forming a stacked pair ,Will denoted as stacked pair In three-dimensional mesh cells The local conflict intensity distribution output by the stack conflict determination function shows that the total number of all non-repeating stack pairs is... For each 3D mesh element The local conflict intensity distribution of all non-repeating stack pairs on the 3D mesh element is normalized and averaged, which is to sum the local conflict intensity distributions of all non-repeating stack pairs and divide by the total number of stack pairs. The conflict intensity value of the three-dimensional mesh element is obtained. The specific calculation formula is as follows:

[0049] ;

[0050] in, This represents summing over all unique stack pairs, i.e., for each stack... and Only Calculations are performed only once per operation to avoid duplicate statistics for the same layer. The conflict intensity values ​​of all 3D mesh elements are organized according to their spatial location to obtain a layer conflict intensity distribution map covering the entire construction display area.

[0051] For example, there is a safety monitoring data layer in the construction display area. Progress label data layer and device location data layer There are a total of 3 stacks, so there are 3 unique stack pairs, namely: , , ,Right now For a given 3D mesh element, denoted as The distribution of local conflict intensity of the three stacked pairs on the three-dimensional mesh element. , , Performing a normalized average, i.e., summing and dividing by 3, yields the conflict intensity value of the 3D mesh element. .

[0052] Through the above processing, the conflict relationships between the layers in the construction display area in terms of spatial overlap, information density, semantic importance and time sensitivity were refined and quantified in three-dimensional grid units. The generated layer conflict intensity distribution map completely recorded the comprehensive conflict intensity at each location in the construction display area with a unified spatial granularity, providing a spatial conflict quantification basis for subsequent extraction of conflict field features and conflict clustering to classify conflict level areas.

[0053] In a preferred embodiment, the construction display area is spatially gridded, the method comprising:

[0054] S311. Obtain the ratio coefficient of the construction display area to the total visualization area;

[0055] S312. Obtain the key area enhancement coefficient of the area type to which the construction display area belongs;

[0056] S313. The nonlinear mapping function is used to analyze and configure the gridding step size for the scaling factor and the key area enhancement factor, and the construction display area is spatially gridded based on the gridding step size.

[0057] Specifically, firstly, the proportion of the construction display area to the total visualization area is obtained. The total visualization area refers to the complete spatial range available for display in the construction visualization system, and the construction display area is the target area within the total visualization area that currently needs to be used for overlay data display and conflict analysis. The proportion is the ratio of the spatial volume of the construction display area to the spatial volume of the total visualization area, with a value ranging from 0 to 1. The proportion reflects the relative observation scale of the current construction display area within the total visualization area. When the proportion is small, it indicates that the construction display area is only a small local area within the total visualization area, and the user is currently focused on local details, requiring a finer mesh step size to ensure the accuracy of local conflict analysis. When the proportion is large, it indicates that the construction display area covers most of the total visualization area, and the user is currently focused on the global situation. If the mesh step size is too small, the number of 3D mesh units will be too large, leading to excessive computation, requiring an appropriate increase in the mesh step size to control the computational load. For example, if the spatial volume of the total visualization area is 10,000 cubic meters and the spatial volume of the construction display area is 3,000 cubic meters, then the proportion is 0.3.

[0058] Next, the critical area enhancement coefficients for the area type to which the construction display area belongs are obtained. A pre-established correspondence between area types and critical area enhancement coefficients is created. Area types are divided according to the functional positioning of the construction display area in construction management; for example, they can be divided into key safety monitoring areas, main structure construction areas, auxiliary facility areas, and general operation areas. Different area types correspond to different critical area enhancement coefficients. For example, the key safety monitoring area, due to its highest requirement for visualization accuracy, has a critical area enhancement coefficient set to 2.0; the main structure construction area has a critical area enhancement coefficient set to 1.5; the auxiliary facility area has a critical area enhancement coefficient set to 1.2; and the general operation area has a critical area enhancement coefficient set to 1.0. A larger critical area enhancement coefficient indicates that the construction display area requires more refined meshing to ensure the accuracy of conflict analysis.

[0059] Subsequently, a nonlinear mapping function is used to analyze and configure the gridding step size for the scaling factor and the critical area enhancement factor. Based on the gridding step size, the construction display area is spatially gridded. The nonlinear mapping function takes the scaling factor and the critical area enhancement factor as input and outputs the gridding step size. In this nonlinear mapping function, the gridding step size increases with the scaling factor, that is, the larger the spatial scale of the construction display area, the larger the gridding step size to control the computational load; at the same time, the gridding step size decreases with the critical area enhancement factor, that is, the higher the criticality of the area, the smaller the gridding step size to ensure the accuracy of conflict analysis. Any nonlinear function that satisfies the above mapping relationship can be used as a nonlinear mapping function, such as a power function or an exponential function.

[0060] The grid step size configuration driven by the above-mentioned scaling factor and key area enhancement factor enables spatial gridding to adaptively adjust according to the actual spatial scale and importance of the construction display area, achieving a balance between computational efficiency and analysis accuracy.

[0061] In a preferred embodiment, a nonlinear mapping function is used to analyze and configure the gridding step size for the scaling factor and the key region enhancement factor. The method includes:

[0062] S3131, The nonlinear mapping function includes a direct proportional nonlinear mapping relationship between the scaling factor and the meshing step size, an inverse proportional nonlinear mapping relationship between the key region enhancement factor and the meshing step size, and a coupled mapping relationship between the key region enhancement factor and the scaling factor;

[0063] S3132. Based on the nonlinear mapping function, analyze and configure the scaling factor and the enhancement factor of the key area in real time, and output the gridded step size.

[0064] Specifically, the nonlinear mapping function includes the direct proportional nonlinear mapping relationship between the scaling factor and the gridding step size, the inverse proportional nonlinear mapping relationship between the critical region enhancement factor and the gridding step size, and the coupled mapping relationship between the critical region enhancement factor and the scaling factor.

[0065] The direct proportional nonlinear mapping relationship between the scaling factor and the meshing step size means that the meshing step size increases nonlinearly with the scaling factor; that is, the larger the scaling factor, the larger the meshing step size. However, the relationship is not a simple linear one, but rather exhibits a nonlinear growth trend. Through this direct proportional nonlinear mapping relationship, when the construction display area is large, the meshing step size is increased accordingly to avoid excessive computation due to an excessive number of 3D mesh elements. When the construction display area is a local detail area, the meshing step size is decreased accordingly to ensure the spatial resolution of the conflict analysis.

[0066] The inversely proportional nonlinear mapping relationship between the critical region enhancement coefficient and the meshing step size means that the meshing step size decreases nonlinearly as the critical region enhancement coefficient increases; that is, the larger the critical region enhancement coefficient, the smaller the meshing step size. However, the relationship is not a simple linear inverse proportion, but rather exhibits a nonlinear shrinkage trend. Through this inversely proportional nonlinear mapping relationship, for construction display areas with high criticality, such as key safety monitoring areas, the meshing step size is nonlinearly reduced, resulting in higher spatial accuracy in conflict field calculations within critical regions and ensuring that layered conflicts within these regions can be accurately captured.

[0067] The coupled mapping relationship between the critical area enhancement coefficient and the scaling factor means that the scaling factor and the critical area enhancement coefficient do not act independently on the gridding step size, but rather have an interactive coupling effect. When the scaling factor is small and the critical area enhancement coefficient is large, the coupling effect further reduces the gridding step size, enhancing the need for refined analysis of key local areas. When the scaling factor is large and the critical area enhancement coefficient is small, the coupling effect weakens, and the gridding step size is mainly determined by the direct proportional nonlinear mapping and the inverse proportional nonlinear mapping. This coupled mapping relationship avoids the unreasonable step size configuration that might occur when the direct proportional nonlinear mapping and the inverse proportional nonlinear mapping act independently. For example, when a user zooms in to view local details of a key security monitoring area, the combined effect of a small scaling factor and a large critical area enhancement coefficient works synergistically through the coupling effect, resulting in a more sufficient reduction in the gridding step size and ensuring the accuracy of conflict analysis in this scenario.

[0068] For example, the specific form of a nonlinear mapping function that simultaneously satisfies the above three mapping relationships can be:

[0069] ;

[0070] in, For gridded step size, As the reference step size, This is the proportionality coefficient. For the enhancement coefficient of the key area, It is a proportional nonlinear mapping exponent. It is an inverse proportional nonlinear mapping exponent. This is the coupling adjustment coefficient. In the above formula, This reflects a direct proportional nonlinear mapping relationship between the scaling factor and the gridding step size. Greater than 0, when When the value is greater than 1, the amplification effect of the scaling factor on the step size increases rapidly with the increase of the scaling factor. When the value is less than 1, the amplification effect tends to level off as the proportionality coefficient increases. This reflects the inverse proportional nonlinear mapping relationship between the enhancement coefficient in the key region and the gridding step size. Greater than 0, when When the value is greater than 1, the effect of the enhancement coefficient on the reduction of step size in the critical region increases more rapidly with the increase of the enhancement coefficient. When the value is less than 1, the reduction effect tends to level off as the enhancement coefficient increases. This reflects the coupling mapping relationship between the enhancement coefficient and the scaling factor in the key area. Greater than 0, coupling term Enhancement coefficient of key region With proportionality coefficient The ratio determines the strength of the coefficient; when the proportionality coefficient is small and the enhancement coefficient in the key area is large, When the value is large, the coupling term increases, and the meshing step size is further reduced; when the scaling factor is large and the enhancement factor in the critical region is small... When the value of is small, the coupling term approaches zero, and the effect of the coupling mapping weakens.

[0071] Among them, the reference step size The settings are based on the spatial scale of the construction visualization system; for example, they can be set as a preset proportion of the total visualization area spanning along the longest coordinate axis. (Direct proportional nonlinear mapping exponent) Inverse proportional nonlinear mapping exponent and coupling adjustment coefficient The optimal gridding step size can be determined by fitting historical data or expert experience data from actual construction management scenarios. For example, optimal gridding step size sample data under different scaling factors and key area enhancement factors in multiple construction projects can be collected, and the values ​​of each parameter can be determined by nonlinear curve fitting methods. Alternatively, it can be directly set empirically according to actual needs. Set to 0.5 Set to 0.8 Set to 0.3.

[0072] by , , , For example, for different combinations of scaling factors and key region enhancement factors, the gridding step size calculated by the above nonlinear mapping function is shown in Table 1 (example of nonlinear mapping function calculation):

[0073] Table 1 Examples of nonlinear mapping function calculations

[0074]

[0075] As shown in Table 1, when the scaling factor remains constant at 0.3, the enhancement coefficient of the critical region increases from 1.0 to 2.0, and the meshing step size decreases from 2.74 to 1.05, reflecting an inverse proportional nonlinear mapping relationship. When the enhancement coefficient of the critical region remains constant at 1.0, the scaling factor increases from 0.3 to 0.8, and the meshing step size increases from 2.74 to 6.50, reflecting a direct proportional nonlinear mapping relationship. At the same time, when the scaling factor is 0.3 and the enhancement coefficient of the critical region is 2.0, the coupling term is 2.0, which is much larger than the coupling term of 0.375 when the scaling factor is 0.8 and the enhancement coefficient of the critical region is 1.0. The meshing step size is further reduced to 1.05, which reflects the strengthening and reduction effect of the coupling mapping relationship on the meshing step size in the critical local area scenario.

[0076] Based on the aforementioned nonlinear mapping function, the input scaling factor and key area enhancement factor are analyzed and configured, and the meshing step size is output. When changes occur in the construction display area, causing updates to the scaling factor or key area enhancement factor, the nonlinear mapping function recalculates the meshing step size in real time based on the updated input, achieving dynamic adaptive adjustment of meshing accuracy.

[0077] Through the synergistic effect of the aforementioned direct proportional nonlinear mapping, inverse proportional nonlinear mapping, and coupled mapping, the nonlinear mapping function can comprehensively consider the interactive influence between the spatial scale and criticality of the construction display area. Compared with the previous method of determining the preset grid step size based on the spatial scale and computational accuracy requirements of the construction display area, this method can dynamically adjust the grid step size according to the actual observation scale and importance of the construction display area, achieving a better balance between computational efficiency and analysis accuracy. Based on the output grid step size, the construction display area is uniformly divided along the three coordinate axes, forming a set of three-dimensional grid units composed of multiple three-dimensional grid units, each of which is a spatial volume element with a side length equal to the grid step size.

[0078] In a preferred embodiment, the stacked conflict determination function further includes a nonlinear enhancement exponential factor; the local conflict intensity distribution output by the stacked conflict determination function is optimized using the nonlinear enhancement exponential factor to obtain an optimized local conflict intensity distribution.

[0079] Specifically, the local conflict intensity distribution output by the stacked conflict determination function through weighted fusion is a linearly weighted result, and the difference in conflict intensity between each 3D mesh unit is entirely determined by the weighted sum of four types of conflict factors. However, in actual construction visualization scenarios, the difference in conflict intensity between different 3D mesh units may be small, and the distinction between high-conflict and low-conflict areas is not obvious enough. Directly using this for subsequent conflict clustering may lead to inaccurate conflict level classification; or in some scenarios, there may be a few extreme high-value points in the conflict distribution, causing the clustering results to be overly concentrated in a few extreme areas and ignoring the overall conflict distribution. Therefore, a nonlinear enhancement index factor is introduced to nonlinearly optimize and adjust the local conflict intensity distribution, enhancing the prominence of high-conflict areas or smoothing the overall conflict distribution according to the needs of different scenarios.

[0080] For example, the optimization method uses exponentiation. After the stacked conflict determination function weights and fuses the four types of conflict factors to output the local conflict intensity distribution, the value of the local conflict intensity distribution is used as the base, and the nonlinear enhancement exponent factor is used as the exponent to perform exponentiation to obtain the optimized local conflict intensity distribution, i.e.:

[0081]

[0082] in, For stacked pairs In three-dimensional mesh cells The local conflict intensity distribution is obtained by weighted fusion of the superimposed conflict determination function output. It is a non-linear enhancement exponential factor. This represents the optimized local conflict intensity distribution.

[0083] Due to the distribution of local conflict intensity After normalization, the value ranges from 0 to 1. When performing exponentiation, the nonlinearity-enhancing exponential factor is... Different values ​​will produce different optimization effects.

[0084] When the nonlinear enhancement exponential factor When the value is greater than 1, for 3D mesh cells with high original local conflict intensity distribution values, the value changes little after exponentiation because their base is close to 1, and they still maintain a high conflict intensity value. However, for 3D mesh cells with low original local conflict intensity distribution values, the value is further reduced after exponentiation because their base is much less than 1, and the conflict intensity value decreases further. As a result, the intensity difference between high-conflict and low-conflict regions is amplified, and high-conflict regions become more prominent in the stacked conflict intensity distribution map, which is beneficial for more accurate identification and division of high-conflict level regions during subsequent conflict clustering.

[0085] When the nonlinear enhancement exponential factor When the value is less than 1 and greater than 0, for 3D mesh elements with low original local conflict intensity distribution values, the value is increased after exponentiation; while for 3D mesh elements with high original local conflict intensity distribution values, the increase in value is relatively small. As a result, the intensity difference between high-conflict and low-conflict regions is reduced, and the overall conflict distribution tends to be smoother. This is beneficial in construction scenarios with relatively uniform conflict distribution, avoiding clustering bias caused by individual extremely high-conflict points.

[0086] Therefore, the nonlinear enhancement exponential factor The value is set according to the needs of the actual construction visualization scenario. When it is necessary to enhance the prominence of high-contact areas, Set to a value greater than 1, for example, in scenarios where multiple layers overlap significantly within a key security monitoring area, and high-conflict areas need to be identified and addressed first. It can be set to 1.5 or 2.0 to amplify the intensity difference between high-conflict and low-conflict areas, making high-conflict areas more prominent in the distribution map and facilitating subsequent targeted fusion processing. When smoothing the overall conflict distribution is required, Set to a value less than 1 and greater than 0, for example, in scenarios with a large global overview and relatively uniform conflict distribution across layers. It can be set to 0.5 or 0.8 to reduce the intensity difference between high-conflict and low-conflict areas, avoiding the over-concentration of subsequent conflict clustering and fusion strategies on local areas due to individual extremely high-conflict points, thus neglecting the global conflict coordination needs. When the value is equal to 1, the optimized local conflict intensity distribution is the same as the original value, meaning no nonlinear optimization adjustment is performed.

[0087] The optimized local conflict intensity distribution replaces the original local conflict intensity distribution in the subsequent normalized average calculation, generating the final layered conflict intensity distribution map. By introducing a nonlinear enhancement exponential factor, the layered conflict determination function can flexibly adjust the contrast characteristics of the conflict intensity distribution according to the needs of different construction management scenarios, enhancing the adaptability of conflict field calculation to different conflict distribution characteristics.

[0088] S4. Extract the conflict field features of the layered conflict intensity distribution map, and perform conflict clustering according to the conflict field features to output multiple conflict level regions.

[0089] Specifically, after obtaining the overlay conflict intensity distribution map, further analysis is conducted to identify spatial regions with different conflict levels within the construction display area. This provides a basis for subsequent differentiated visualization fusion strategies targeting different conflict levels. The overlay conflict intensity distribution map records the conflict intensity values ​​of each three-dimensional grid cell within the construction display area. However, these conflict intensity values ​​are discrete values ​​calculated independently for each grid cell, and do not yet reveal the overall spatial distribution pattern and structural characteristics of the conflict. Therefore, conflict field features are first extracted from the overlay conflict intensity distribution map. These features characterize the spatial distribution of conflict intensity from multiple dimensions, reflecting information such as the absolute level of conflict intensity at each three-dimensional grid cell, the spatial trend of conflict intensity variation, and the stability of conflict intensity over time.

[0090] After extracting the conflict field features, conflict clustering is performed on each 3D mesh unit within the construction display area according to these features. 3D mesh units with similar conflict field features are grouped into the same category, outputting multiple conflict level regions. Different conflict level regions represent spatial partitions within the construction display area with varying degrees of conflict. For example, 3D mesh units in high conflict level regions generally have high conflict intensity and drastic conflict changes, while 3D mesh units in low conflict level regions have lower conflict intensity and more stable conflict distribution. By dividing the construction display area into multiple conflict level regions through conflict clustering, subsequent visualization fusion processing can adopt differentiated fusion strategies for different conflict level regions. Stronger fusion intervention is applied to high conflict regions, while lighter fusion processing is used for low conflict regions, achieving a reasonable allocation of fusion resources and overall optimization of the fusion effect.

[0091] In a preferred embodiment, the conflict field features of the stacked conflict intensity distribution map are extracted, and conflict clustering is performed according to the conflict field features to output multiple conflict level regions. The method includes:

[0092] The conflict field features include conflict intensity features, conflict gradient change features, and temporal stability mean features;

[0093] S41. Introduce spatial continuity constraints to perform density clustering on the conflict field features to obtain the category label corresponding to each three-dimensional mesh unit. The category label includes at least low conflict type, medium conflict type and high conflict type.

[0094] S42. Perform connectivity analysis on the same category of labels to obtain a set of conflict regions, and output multiple conflict level regions according to the set of conflict regions.

[0095] Specifically, the conflict field characteristics include conflict intensity characteristics, conflict gradient variation characteristics, and temporal stability mean characteristics. Among them, the conflict intensity characteristic is the conflict intensity value of each three-dimensional grid cell in the stacked conflict intensity distribution map, which directly reflects the severity of the stacked conflict at that three-dimensional grid cell.

[0096] The conflict gradient variation characteristic is used to reflect the drastic spatial variation of conflict intensity. For each 3D mesh cell, the difference between the conflict intensity value of that 3D mesh cell and the conflict intensity values ​​of its adjacent 3D mesh cells along the three coordinate axes is calculated, yielding the conflict intensity gradient components along each of the three coordinate axes. The square root of the sum of the squares of the gradient components in the three directions is used to obtain the conflict gradient variation characteristic value of that 3D mesh cell. A larger conflict gradient variation characteristic value indicates a more drastic spatial variation in conflict intensity at that 3D mesh cell, placing that location in a boundary zone where conflict intensity transitions sharply; a smaller conflict gradient variation characteristic value indicates a more gradual spatial variation in conflict intensity at that 3D mesh cell, placing that location within a region of relatively uniform conflict intensity.

[0097] The temporal stability mean characteristic is used to reflect the stability of conflict intensity over time. Since the multi-source overlay data within the construction display area is continuously updated over time, the overlay conflict intensity distribution map also changes dynamically. Therefore, it is necessary to evaluate the fluctuation of conflict intensity in each 3D mesh unit over time. For each 3D mesh unit, conflict intensity values ​​are collected at multiple times within a preset time observation window. The mean of these conflict intensity values ​​is calculated as the time mean of the 3D mesh unit. Then, the mean of the sum of squares of the deviations between the conflict intensity values ​​at each time point and the time mean is calculated, and the square root is taken to obtain the temporal stability mean characteristic value of the 3D mesh unit, i.e., the standard deviation of conflict intensity over time. A larger temporal stability mean characteristic value indicates that the conflict intensity at the 3D mesh unit fluctuates more drastically over time and the conflict state is more unstable; a smaller temporal stability mean characteristic value indicates that the conflict intensity at the 3D mesh unit remains relatively stable over time.

[0098] The conflict field feature vector of each three-dimensional mesh cell is formed by combining the conflict intensity feature, conflict gradient change feature value, and temporal stability mean feature value.

[0099] After extracting the conflict field feature vectors, spatial continuity constraints are introduced to perform density clustering on the conflict field features, obtaining the category label corresponding to each 3D mesh cell. The density clustering can employ the DBSCAN algorithm, which clusters data points based on their density distribution, does not require pre-specifying the number of categories, and can automatically identify clusters of arbitrary shapes. In this embodiment, the conflict field feature vectors of each 3D mesh cell are used as the input data points for density clustering, and the Euclidean distance between the conflict field feature vectors is used as a similarity measure. The spatial continuity constraint is introduced as follows: when calculating the distance between two 3D mesh elements, not only the Euclidean distance between their conflict field feature vectors is considered, but the spatial distance between the two 3D mesh elements in the construction display area is also included as a constraint in the distance calculation. The Euclidean distance of the conflict field feature vectors and the spatial distance are weighted and summed to obtain a comprehensive distance metric. The sum of the weights of the Euclidean distance and the spatial distance is 1. For example, the weight of the Euclidean distance is set to 0.7, and the weight of the spatial distance is set to 0.3. This ensures that the similarity of conflict field features plays a dominant role in clustering while also taking into account the spatial continuity constraint. The larger the weight of the spatial distance, the stronger the spatial continuity constraint. Even if 3D mesh elements that are spatially far apart have similar conflict field features, their comprehensive distance is still large, making them less likely to be clustered into the same category, thus ensuring the spatial continuity of the clustering results. After density clustering is completed, the clusters are labeled with conflict level based on the average conflict intensity feature of the three-dimensional mesh units within each cluster. The cluster with the highest average conflict intensity feature is labeled as high conflict category, the cluster with the lowest average conflict intensity feature is labeled as low conflict category, and the clusters in between are labeled as medium conflict category. Thus, the category label corresponding to each three-dimensional mesh unit is obtained. The category label includes at least low conflict type, medium conflict category, and high conflict category.

[0100] Next, connectivity analysis is performed on labels of the same category to obtain a set of conflict regions. Multiple conflict level regions are then output based on this set. Specifically, for each category label, all 3D mesh cells with that category label within the construction display area are extracted, and spatial connectivity analysis is performed on these 3D mesh cells. Since 3D mesh cells with the same category label may be distributed in multiple spatially unconnected locations within the construction display area, the purpose of spatial connectivity analysis is to identify and merge spatially adjacent 3D mesh cells into independent regions. Connectivity is determined as follows: if two 3D mesh cells with the same category label share a face in any of the three coordinate axes, they are considered connected. Starting from any 3D mesh cell, the process extends along the shared face to all connected 3D mesh cells with the same category label, merging these interconnected 3D mesh cells into a single connected region. If multiple sets of spatially unconnected 3D mesh cells exist under the same category label, multiple independent connected regions are formed. The above connectivity analysis is performed on all category labels. Each connected region constitutes an independent conflict level region, and its conflict level is determined by the category label of the 3D mesh cells within that connected region. All connected regions form a set of conflict regions. The conflict area set outputs multiple conflict level regions. Each conflict level region consists of a set of spatially continuous three-dimensional grid units, with a clear spatial range and corresponding category label. This provides a spatial partitioning basis for subsequent development of differentiated visualization fusion strategies for different conflict level regions.

[0101] S5. Based on the visualization adaptive fusion module, perform differentiated fusion strategy analysis on the multi-type conflict level areas, obtain multi-type visualization fusion expressions, and perform visualization output management on the construction display area according to the multi-type visualization fusion expressions.

[0102] Specifically, the adaptive visualization fusion module is a software module in the construction visualization system used to perform visualization rendering output of overlay data. It possesses the ability to configure visualization function parameters such as color mapping, transparency control, boundary rendering, and dynamic refresh. It can adaptively adjust its visualization function parameters according to the conflict characteristics of regions with different conflict levels, and then perform fusion rendering output of multi-source overlay data according to the adjusted parameter configuration to generate a visualized fusion expression. Because the degree of conflict between overlays in regions with different conflict levels varies significantly, if a uniform visualization processing method is used for all regions, the information occlusion and stacking problems in high-conflict categories cannot be effectively solved, the hierarchy of overlays in medium-conflict categories cannot be reasonably coordinated, and low-conflict categories may be subjected to unnecessary overprocessing, resulting in the destruction of the original display effect. Therefore, the adaptive visualization fusion module differentiates its visualization function parameters according to the conflict characteristics of regions with different conflict levels, and performs targeted rendering and fusion processing of multi-source overlay data in each conflict level region according to the configured parameters.

[0103] Specifically, the process begins by analyzing each conflict level region within the multi-conflict level region to determine the applicable strategy type and corresponding visualization strategy parameters. These visualization strategy parameters are configured using the visualization function parameters of the adaptive fusion module itself. Then, based on the determined strategy type and visualization strategy parameters, multi-source overlay data within each conflict level region is fused to obtain multi-class visualization fusion expressions. Each visualization fusion expression corresponds to the fusion processing result of a conflict level region, including the visualization display method of each overlay data within that region after fusion processing, such as color mapping schemes, transparency settings, boundary enhancement effects, and dynamic display methods for each overlay. Visualization output management is then implemented in the construction display area according to these multi-class visualization fusion expressions. This involves applying the visualization fusion expressions of each conflict level region to their respective spatial ranges for rendering output. This ensures that overlay information in high-conflict regions is clearly expressed through strong fusion processing, maintaining its original display effect through lightweight fusion, and achieving a balance between information integrity and display clarity through moderate fusion of overlay information in medium-conflict regions.

[0104] Through the above-mentioned differentiated fusion strategy analysis and visualization output management, the visualization distortion caused by spatial overlap and information conflict of multi-source overlay data in the construction display area is eliminated, enabling construction managers to accurately judge the on-site situation and make efficient decisions.

[0105] In a preferred embodiment, the method involves performing differentiated fusion strategy analysis on the multiple conflict level regions using a visualization adaptive fusion module to obtain multiple visualization fusion representations.

[0106] S51. Perform structured analysis on each type of conflict level region in the multi-type conflict level regions, and extract region attribute information, including average conflict intensity, peak conflict intensity, conflict spatial range and information density.

[0107] S52. Construct a visualization fusion strategy library, which includes multiple pre-stored strategy types, applicable scenario attribute information for each strategy type, and visualization strategy parameter templates corresponding to each strategy type.

[0108] S53. Call the visualization fusion strategy library to perform matching analysis on the regional attribute information, and determine the matching strategy type and the matching visualization strategy parameters;

[0109] S54. Perform overlay fusion calculation on the multi-source overlay data within each conflict level region to obtain multi-class visual fusion representations.

[0110] Specifically, the visualization adaptive fusion module analyzes differentiated fusion strategies for regions with multiple conflict levels, determines the visualization function parameter configuration applicable to each conflict level region, and then renders and fuses multi-source overlay data according to the configuration to obtain multiple types of visualization fusion expressions.

[0111] First, a structured analysis is performed on each conflict level region within the multi-level conflict region to extract regional attribute information, including average conflict intensity, peak conflict intensity, conflict spatial range, and information density. Specifically, for each conflict level region, all three-dimensional mesh cells contained within that region are traversed, and the arithmetic mean of the conflict intensity values ​​of these three-dimensional mesh cells is calculated as the average conflict intensity of that conflict level region. The maximum value among these conflict intensity values ​​is extracted as the peak conflict intensity of that conflict level region. The number of three-dimensional mesh cells contained within that conflict level region is counted and multiplied by the volume of a single three-dimensional mesh cell to obtain the conflict spatial range of that conflict level region. The arithmetic mean of the information entropy values ​​of all three-dimensional mesh cells within that conflict level region is calculated, and this arithmetic mean is normalized to map it to the interval between 0 and 1. For example, the arithmetic mean is normalized by dividing the theoretical maximum value of the information entropy, where the theoretical maximum value of the information entropy is determined by the number of intervals K and the logarithmic base used in the information entropy calculation in S2. The normalized value is used as the information density of that conflict level region. The above four indicators together constitute the regional attribute information of the conflict level area, and provide a structured description of the conflict characteristics of the conflict level area from four dimensions: average level of conflict, extreme degree of conflict, spatial scale and information carrying capacity.

[0112] Next, a visualization fusion strategy library is constructed. This library includes multiple pre-stored strategy types, each with corresponding applicable scenario attribute information and visualization strategy parameter templates. Each strategy type corresponds to a preset configuration mode for the visualization function parameters of the adaptive visualization fusion module. Strategy types may include, for example, transparency layering strategies, semantic priority highlighting strategies, spatial offset separation strategies, and dynamic temporal alternation strategies. Specifically, the transparency layering strategy is suitable for conflict areas where the average conflict intensity is below a first preset threshold and the information density is in a moderate range, such as an average conflict intensity below 0.3 and an information density between 0.3 and 0.6. Its visualization strategy parameter template includes transparency gradient configurations for each layer. By setting differentiated transparency for different layers, each layer can see through to the others when overlaid, reducing occlusion. The semantic priority highlighting strategy is suitable for conflict areas where the average conflict intensity is higher than a second preset threshold and the difference between the peak conflict intensity and the average conflict intensity is significant, such as an average conflict intensity higher than 0.5 and a ratio of peak conflict intensity to average conflict intensity greater than 1.5. Its visualization strategy parameter template includes color enhancement parameters for high semantic importance overlays and color weakening parameters for low semantic importance overlays. By amplifying the visual difference between primary and secondary overlays in color expression, key information is presented first. The spatial offset separation strategy is suitable for conflict areas where the conflict spatial range is greater than a third preset threshold and multiple overlays overlap significantly, such as a conflict spatial range exceeding 20% ​​of the entire construction display area volume and an average conflict intensity between 0.4 and 0.7. Its visualization strategy parameter template includes spatial offset direction and offset distance parameters for each overlay. By slightly offsetting the overlapping overlays in space, each overlay is displayed separately. The dynamic temporal alternation strategy is applicable to conflict areas where the information density is higher than a fourth preset threshold and the temporal sensitivity of each layer within the conflict area differs significantly. For example, the information density is higher than 0.7 and the ratio of the maximum to minimum temporal sensitivity values ​​of the layers within the area is greater than 3. Its visualization strategy parameter template includes the display time allocation and switching frequency parameters for each layer. By alternating the display of different layers in the time dimension, it avoids information stacking at the same time. The applicable scenario attribute information for each strategy type records the value range or combination of conditions of the regional attribute information to which the strategy type applies, which is used for subsequent matching with the regional attribute information of the actual conflict area.

[0113] Next, the visualization fusion strategy library is invoked to perform matching analysis on the regional attribute information to determine the matching strategy type and the matching visualization strategy parameters. Specifically, the regional attribute information of each conflict level region is compared with the applicable scenario attribute information of each strategy type in the visualization fusion strategy library. The degree of matching between the regional attribute information and the applicable scenario attribute information of each strategy type is calculated. The strategy type with the highest degree of matching is selected as the matching strategy type, and the visualization strategy parameter template corresponding to this strategy type is read. The template parameters are adjusted according to the actual regional attribute information of the conflict level region to obtain the matching visualization strategy parameters.

[0114] Subsequently, multi-source overlay data within each conflict level region are subjected to overlay fusion calculations to obtain multi-level visual fusion representations. Specifically, for each conflict level region, the visualization function parameters of the visualization adaptive fusion module are set according to the matching strategy type and matching visualization strategy parameters determined for that region. The visualization adaptive fusion module then renders and fuses the multi-source overlay data in each 3D mesh unit contained within that conflict level region according to the set visualization function parameters. If the matching strategy type is a transparency layering strategy, the transparency configuration of each overlay in the matching visualization strategy parameters is used. Based on the semantic importance value of each overlay, transparency values ​​are assigned from low to high, with the overlay with the highest semantic importance set to the lowest transparency (i.e., the least transparent) for priority display. Overlays with lower semantic importance are set to higher transparency to make them semi-transparent, achieving perspective overlay display of each overlay. If the matching strategy type is a semantic priority highlighting strategy, then according to the color enhancement and weakening parameters in the matching visualization strategy parameters, the color saturation and brightness of the high semantic importance overlay are enhanced, while the color saturation and brightness of the low semantic importance overlay are reduced, making the high semantic importance overlay more visually prominent. If the matching strategy type is a spatial offset separation strategy, then according to the spatial offset direction and offset distance of each overlay in the matching visualization strategy parameters, the overlapping overlays in the region are slightly spatially displaced along the specified direction, causing the originally overlapping overlays to be spatially misaligned and separated, allowing the information of each overlay to be presented independently. If the matching strategy type is a dynamic temporal alternation strategy, then according to the display time allocation and switching frequency of each overlay in the matching visualization strategy parameters, the overlays are displayed in turn in the time dimension, with high time-sensitive overlays allocated more display time and higher refresh rates, and low time-sensitive overlays allocated less display time, avoiding the simultaneous stacking of all overlays at the same time through temporal alternation. After performing the above fusion process on all conflict level regions, each conflict level region obtains a set of visual display configurations that have undergone differentiated fusion processing, namely, visual fusion expression. The visual fusion expressions of all conflict level regions together constitute multiple types of visual fusion expressions.

[0115] In a preferred embodiment, the method involves calling the visualization fusion strategy library to perform matching analysis on the regional attribute information, determining the matching strategy type and the matching visualization strategy parameters, and includes:

[0116] S531. Calculate the similarity between the regional attribute information and the applicable scenario attribute information of multiple strategy types pre-stored in the visualization fusion strategy library to obtain a similarity index.

[0117] S532. Obtain the first strategy type according to the similarity index, and read the initial visualization strategy parameters of the first strategy type;

[0118] S533. Optimize and correct the initial visualization strategy parameters based on the regional attribute information, and output the matching visualization strategy parameters.

[0119] Specifically, firstly, similarity metrics are calculated based on the regional attribute information and the applicable scenario attribute information of multiple strategy types pre-stored in the visualization fusion strategy library. Specifically, for each conflict level region, the regional attribute information of that region is compared item by item with the applicable scenario attribute information of each strategy type in the visualization fusion strategy library. The regional attribute information includes four metrics: average conflict intensity, peak conflict intensity, conflict spatial range, and information density. The applicable scenario attribute information for each strategy type also includes the value range or conditions of these four metrics. The similarity calculation method is as follows: For each indicator, the deviation between the actual indicator value of the conflict level region and the center value of the corresponding indicator in the applicable scenario attribute information of the strategy type is calculated. The deviations of the four indicators are weighted and summed to obtain the comprehensive deviation value between the conflict level region and the strategy type. The weights of each indicator are set according to its importance in strategy matching. For example, the weight of average conflict intensity is set to 0.4, the weight of peak conflict intensity is set to 0.2, the weight of conflict spatial range is set to 0.2, and the weight of information density is set to 0.2, so that average conflict intensity plays a dominant role in the matching calculation. The smaller the comprehensive deviation value, the higher the degree of matching. The reciprocal of the comprehensive deviation value or a reverse mapping is used to obtain the similarity index. The larger the similarity index, the more the regional attribute information of the conflict level region matches the applicable scenario of the strategy type. The above similarity calculation is performed on all strategy types in the visualization fusion strategy library to obtain the similarity index between the conflict level region and each strategy type.

[0120] Then, the first strategy type is obtained according to the similarity index, and the initial visual strategy parameters of the first strategy type are read. Specifically, the similarity index between the conflict level region and all strategy types is sorted from largest to smallest, and the strategy type with the largest similarity index is selected as the first strategy type, that is, the strategy type that best matches the regional attribute information of the conflict level region. The visual strategy parameter template corresponding to the first strategy type is read as the initial visual strategy parameters. These initial visual strategy parameters are general parameter configurations preset for typical applicable scenarios of the first strategy type and have not yet been personalized according to the specific attributes of the current conflict level region.

[0121] Next, the initial visualization strategy parameters are optimized and corrected based on the regional attribute information, outputting matching visualization strategy parameters. Specifically, since the initial visualization strategy parameters are general parameters preset for typical applicable scenarios of the first strategy type, and the actual regional attribute information of the current conflict level region may deviate from the typical applicable scenarios, it is necessary to correct and adjust the initial visualization strategy parameters according to the actual regional attribute information of the current conflict level region, so that the visualization strategy parameters are more in line with the actual conflict characteristics of the current region. The visualization strategy parameters output after parameter optimization and correction are the matching visualization strategy parameters.

[0122] Through the above steps, the visualization adaptive fusion module can accurately match the most suitable fusion strategy type for the actual conflict characteristics of each conflict level area, and make personalized adjustments based on the general parameter template, so that the fusion strategy parameters are highly adapted to the conflict status of the current area.

[0123] In a preferred embodiment, the initial visualization strategy parameters are optimized and corrected based on the regional attribute information, the method comprising:

[0124] S5331, The initial visualization strategy parameters include color mapping, transparency, boundary enhancement parameters, and dynamic requirement parameters;

[0125] S5332. Establish the correction mapping relationship between the regional attribute information and each initial visualization strategy parameter, and perform parameter optimization and correction based on the correction mapping relationship to output matching visualization strategy parameters.

[0126] Specifically, the initial visualization strategy parameters are the initial value configurations of the visualization function parameters of the visualization adaptive fusion module itself, including color mapping, transparency, boundary enhancement parameters, and dynamic requirement parameters. These parameters correspond to the specific configuration values ​​of various visualization functions adopted by the visualization adaptive fusion module during rendering output. Among them, the color mapping parameter defines the color scheme used by each overlay data in the visualization display, including the configuration values ​​of hue, saturation, and brightness. The transparency parameter defines the transparency level of each overlay when displayed in an overlay manner, with a value range between 0 and 1, where 0 represents completely opaque and 1 represents completely transparent. The boundary enhancement parameter defines the degree of enhancement of the boundary outline of each overlay when displayed, including the boundary line width and boundary color contrast, used to enhance the discernibility of the boundaries of each overlay in the overlapping area. The dynamic requirement parameters define the time response configuration of each overlay in dynamic display, including the display refresh rate and transition time.

[0127] Then, a correction mapping relationship is established between regional attribute information and each initial visualization strategy parameter. Based on the correction mapping relationship, parameter optimization is performed to correct and output matching visualization strategy parameters. Specifically, the correction mapping relationship defines the correction direction and magnitude of each indicator in the regional attribute information for each initial visualization strategy parameter. The correction mapping relationship can be established based on practical application experience in construction visualization management or through statistical analysis of historical construction visualization data. Alternatively, it can be customized by construction management personnel according to the visualization needs of different construction scenarios.

[0128] For example, regarding color mapping parameters, the modified mapping relationship can be established as follows: when the average conflict intensity is higher, the color saturation of high semantically important overlays increases more significantly from the initial value, while the color saturation of low semantically important overlays decreases more significantly from the initial value, thus widening the visual difference between primary and secondary overlays within high-conflict areas. For instance, when the average conflict intensity is 0.8, the color saturation of high semantically important overlays increases by 20% from the initial value, while the color saturation of low semantically important overlays decreases by 15% from the initial value.

[0129] Regarding the transparency parameter, the modified mapping relationship can be established as follows: when the information density is higher, the transparency of low semantic importance layers increases more significantly from the initial value, so as to further reduce the visual interference of secondary layers in areas with high information content. For example, when the information density is 0.7, the transparency of low semantic importance layers increases by 0.15 from the initial value.

[0130] For boundary reinforcement parameters, the modified mapping relationship can be established as follows: the greater the difference between peak conflict intensity and average conflict intensity, the greater the increase in boundary line width and boundary color contrast from the initial value, thus enhancing the discernibility of the stacked boundary in areas with drastic changes in conflict intensity. For example, when the ratio of peak conflict intensity to average conflict intensity is 1.8, the boundary line width increases by 30% from the initial value, and the boundary color contrast increases by 25% from the initial value.

[0131] For dynamic demand parameters, the mapping relationship can be adjusted as follows: when the conflict space is larger, the display refresh rate is appropriately reduced from the initial value to control the rendering computation load; when the time sensitivity difference of the layers within the region is greater, the display refresh rate of the high time-sensitive layers is increased from the initial value, and the switching transition time is shortened from the initial value to ensure the real-time update requirements of the high time-sensitive layers. For example, when the conflict space exceeds 30% of the construction display area volume, the overall display refresh rate is reduced by 10% from the initial value.

[0132] Based on the aforementioned modified mapping relationship, the color mapping, transparency, boundary enhancement parameters, and dynamic requirement parameters in the initial visualization strategy parameters are modified and adjusted respectively, outputting matching visualization strategy parameters. Through parameter optimization and correction, the visualization strategy parameters are transformed from a general template configuration to a personalized configuration tailored to the actual conflict characteristics of the current conflict level region, improving the accuracy of the fusion processing in adapting to different conflict level regions.

[0133] Example 2, as Figure 2 As shown, based on the same inventive concept as the visual construction management method provided in Embodiment 1, this embodiment of the invention also provides a visual construction management system, including:

[0134] The data standardization processing unit 11 is used to standardize the multi-source overlay data in the construction display area and construct a standardized data structure.

[0135] The layered feature extraction unit 12 is used to extract the layered feature vector of the standardized data structure. The layered feature vector includes a space occupancy feature vector, an information density feature vector, a semantic importance feature vector, and a time sensitivity feature vector.

[0136] The conflict field calculation unit 13 is used to construct a layer conflict determination function. The layer conflict determination function performs conflict field calculation on any two layers in the construction display area based on the layer feature vector to generate a layer conflict intensity distribution map.

[0137] The conflict clustering and grading unit 14 is used to extract the conflict field features of the stacked conflict intensity distribution map, and perform conflict clustering according to the conflict field features to output multiple conflict level regions.

[0138] The visualization fusion output unit 15 is used to perform differentiated fusion strategy analysis on the multi-type conflict level areas according to the visualization adaptive fusion module, obtain multi-type visualization fusion expressions, and perform visualization output management on the construction display area according to the multi-type visualization fusion expressions.

[0139] Furthermore, the collision field calculation unit 13 is also used for:

[0140] The construction display area is spatially meshed to obtain a set of three-dimensional mesh units;

[0141] For each three-dimensional mesh cell in the set of three-dimensional mesh cells, the local conflict factor is calculated for the stacked feature vectors of any two stacks to obtain the spatial overlap conflict factor, information density conflict factor, semantic importance conflict factor, and time sensitivity conflict factor.

[0142] The input spatial overlap conflict factor, information density conflict factor, semantic importance conflict factor, and time-sensitive conflict factor are fused according to the stacked conflict determination function to output the local conflict intensity distribution corresponding to each three-dimensional mesh unit.

[0143] The stacked conflict intensity distribution map is obtained by normalizing and averaging the local conflict intensity distribution of each three-dimensional mesh cell.

[0144] Furthermore, the collision field calculation unit 13 is also used for:

[0145] Obtain the ratio of the construction display area to the total visualization area;

[0146] Obtain the key area enhancement coefficient of the area type to which the construction display area belongs;

[0147] A nonlinear mapping function is used to analyze and configure the gridding step size for the scaling factor and the key area enhancement factor. Based on the gridding step size, the construction display area is spatially gridded.

[0148] Furthermore, the conflict clustering hierarchical unit 14 is also used for:

[0149] The conflict field features include conflict intensity features, conflict gradient change features, and temporal stability mean features;

[0150] A spatial continuity constraint is introduced to perform density clustering on the conflict field features to obtain a category label corresponding to each three-dimensional mesh cell. The category label includes at least low conflict type, medium conflict type and high conflict type.

[0151] Connectivity analysis is performed on labels of the same category to obtain a set of conflict regions, and multiple conflict level regions are output according to the set of conflict regions.

[0152] Furthermore, the visualization fusion output unit 15 is also used for:

[0153] Each conflict level region in the multi-level conflict region is subjected to structured analysis to extract regional attribute information, including average conflict intensity, peak conflict intensity, conflict spatial range, and information density.

[0154] Construct a visualization fusion strategy library, which includes multiple pre-stored strategy types, applicable scenario attribute information for each strategy type, and visualization strategy parameter templates corresponding to each strategy type;

[0155] The visualization fusion strategy library is invoked to perform matching analysis on the regional attribute information to determine the matching strategy type and the matching visualization strategy parameters;

[0156] Multi-source overlay data within each conflict level region are overlaid and fused to obtain multi-class visual fusion representations.

[0157] Furthermore, the visualization fusion output unit 15 is also used for:

[0158] The similarity index is obtained by calculating the similarity between the regional attribute information and the applicable scenario attribute information of multiple strategy types pre-stored in the visualization fusion strategy library.

[0159] Obtain the first strategy type according to the similarity index, and read the initial visualization strategy parameters of the first strategy type;

[0160] Based on the regional attribute information, the initial visualization strategy parameters are optimized and corrected, and matching visualization strategy parameters are output.

[0161] Furthermore, the visualization fusion output unit 15 is also used for:

[0162] The initial visualization strategy parameters include color mapping, transparency, boundary enhancement parameters, and dynamic requirement parameters.

[0163] Establish a correction mapping relationship between the regional attribute information and each initial visualization strategy parameter, and optimize and correct the parameters based on the correction mapping relationship to output matching visualization strategy parameters.

[0164] Furthermore, the collision field calculation unit 13 is also used for:

[0165] The stacking conflict determination function also includes a nonlinear enhancement exponential factor;

[0166] The local conflict intensity distribution output by the stacked conflict determination function is optimized using the nonlinear enhancement exponential factor to obtain the optimized local conflict intensity distribution.

[0167] Furthermore, the collision field calculation unit 13 is also used for:

[0168] The nonlinear mapping function includes a direct proportional nonlinear mapping relationship between the scaling factor and the meshing step size, an inverse proportional nonlinear mapping relationship between the key region enhancement factor and the meshing step size, and a coupled mapping relationship between the key region enhancement factor and the scaling factor.

[0169] Based on the nonlinear mapping function, the scaling factor of the real-time input and the enhancement factor of the key area are analyzed and configured, and the gridded step size is output.

[0170] Although preferred embodiments of the invention have been described, those skilled in the art, once they have learned the basic inventive concept, can make other changes and modifications to these embodiments.

[0171] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of this invention and its equivalents, this invention also intends to include these modifications and variations.

Claims

1. A visual construction management method, characterized in that, The method includes: Standardize the multi-source overlay data within the construction display area to construct a standardized data structure; Extract the stacked feature vectors of the standardized data structure, which include space occupancy feature vectors, information density feature vectors, semantic importance feature vectors, and time sensitivity feature vectors. Construct a layer conflict determination function, which calculates the conflict field for any two layers in the construction display area based on the layer feature vector, and generates a layer conflict intensity distribution map; Extract the conflict field features from the layered conflict intensity distribution map, and perform conflict clustering according to the conflict field features to output multiple conflict level regions; The visualization adaptive fusion module performs differentiated fusion strategy analysis on the multiple conflict level areas to obtain multiple visualization fusion expressions, and performs visualization output management on the construction display area according to the multiple visualization fusion expressions.

2. The visual construction management method as described in claim 1, characterized in that, The layer conflict determination function calculates the conflict field for any two layers within the construction display area based on the layer feature vector, generating a layer conflict intensity distribution map. The method further includes: The construction display area is spatially meshed to obtain a set of three-dimensional mesh units; For each three-dimensional mesh cell in the set of three-dimensional mesh cells, the local conflict factor is calculated for the stacked feature vectors of any two stacks to obtain the spatial overlap conflict factor, information density conflict factor, semantic importance conflict factor, and time sensitivity conflict factor. The input spatial overlap conflict factor, information density conflict factor, semantic importance conflict factor, and time-sensitive conflict factor are fused according to the stacked conflict determination function to output the local conflict intensity distribution corresponding to each three-dimensional mesh unit. The stacked conflict intensity distribution map is obtained by normalizing and averaging the local conflict intensity distribution of each three-dimensional mesh cell.

3. The visual construction management method as described in claim 2, characterized in that, The method for spatially gridding the construction display area includes: Obtain the ratio of the construction display area to the total visualization area; Obtain the key area enhancement coefficient of the area type to which the construction display area belongs; A nonlinear mapping function is used to analyze and configure the gridding step size for the scaling factor and the key area enhancement factor. Based on the gridding step size, the construction display area is spatially gridded.

4. The visual construction management method as described in claim 1, characterized in that, Extracting the conflict field features from the stacked conflict intensity distribution map, and performing conflict clustering based on the conflict field features to output multi-class conflict level regions, the method includes: The conflict field features include conflict intensity features, conflict gradient change features, and temporal stability mean features; A spatial continuity constraint is introduced to perform density clustering on the conflict field features to obtain a category label corresponding to each three-dimensional mesh cell. The category label includes at least low conflict type, medium conflict type and high conflict type. Connectivity analysis is performed on labels of the same category to obtain a set of conflict regions, and multiple conflict level regions are output according to the set of conflict regions.

5. The visual construction management method as described in claim 1, characterized in that, The method involves performing differentiated fusion strategy analysis on the multiple conflict level regions using a visualization adaptive fusion module to obtain multiple visualization fusion representations. Each conflict level region in the multi-level conflict region is subjected to structured analysis to extract regional attribute information, including average conflict intensity, peak conflict intensity, conflict spatial range, and information density. Construct a visualization fusion strategy library, which includes multiple pre-stored strategy types, applicable scenario attribute information for each strategy type, and visualization strategy parameter templates corresponding to each strategy type; The visualization fusion strategy library is invoked to perform matching analysis on the regional attribute information to determine the matching strategy type and the matching visualization strategy parameters; Multi-source overlay data within each conflict level region are overlaid and fused to obtain multi-class visual fusion representations.

6. The visual construction management method as described in claim 5, characterized in that, The method involves calling the visualization fusion strategy library to perform matching analysis on the regional attribute information, determining the matching strategy type and the matching visualization strategy parameters, and includes: The similarity index is obtained by calculating the similarity between the regional attribute information and the applicable scenario attribute information of multiple strategy types pre-stored in the visualization fusion strategy library. Obtain the first strategy type according to the similarity index, and read the initial visualization strategy parameters of the first strategy type; Based on the regional attribute information, the initial visualization strategy parameters are optimized and corrected, and matching visualization strategy parameters are output.

7. The visual construction management method as described in claim 6, characterized in that, The method for optimizing and correcting the initial visualization strategy parameters based on the regional attribute information includes: The initial visualization strategy parameters include color mapping, transparency, boundary enhancement parameters, and dynamic requirement parameters. Establish a correction mapping relationship between the regional attribute information and each initial visualization strategy parameter, and optimize and correct the parameters based on the correction mapping relationship to output matching visualization strategy parameters.

8. The visual construction management method as described in claim 2, characterized in that, The stacking conflict determination function also includes a nonlinear enhancement exponential factor; The local conflict intensity distribution output by the stacked conflict determination function is optimized using the nonlinear enhancement exponential factor to obtain the optimized local conflict intensity distribution.

9. The visual construction management method as described in claim 3, characterized in that, The method involves using a nonlinear mapping function to analyze and configure the gridding step size for the scaling factor and the enhancement factor of the key region. The nonlinear mapping function includes a direct proportional nonlinear mapping relationship between the scaling factor and the meshing step size, an inverse proportional nonlinear mapping relationship between the key region enhancement factor and the meshing step size, and a coupled mapping relationship between the key region enhancement factor and the scaling factor. Based on the nonlinear mapping function, the scaling factor of the real-time input and the enhancement factor of the key area are analyzed and configured, and the gridded step size is output.

10. A visual construction management system, characterized in that, The system is used to implement the visual construction management method as described in any one of claims 1 to 9, the system comprising: The data standardization processing unit is used to standardize the multi-source overlay data within the construction display area and construct a standardized data structure. The layered feature extraction unit is used to extract the layered feature vector of the standardized data structure. The layered feature vector includes a space occupancy feature vector, an information density feature vector, a semantic importance feature vector, and a time sensitivity feature vector. The conflict field calculation unit is used to construct a layer conflict determination function. The layer conflict determination function performs conflict field calculation on any two layers in the construction display area based on the layer feature vector, and generates a layer conflict intensity distribution map. The conflict clustering and hierarchical unit is used to extract the conflict field features of the stacked conflict intensity distribution map, and perform conflict clustering according to the conflict field features to output multiple conflict level regions; The visualization fusion output unit is used to perform differentiated fusion strategy analysis on the multi-conflict level areas according to the visualization adaptive fusion module, obtain multi-type visualization fusion expressions, and perform visualization output management on the construction display area according to the multi-type visualization fusion expressions.

Images