Urban Operation and Maintenance Status Visualization Method and Device Based on Multi-Source Data Fusion

Abstract

This invention belongs to the field of operation and maintenance visualization technology, and discloses a method and device for visualizing urban operation and maintenance situation based on multi-source data fusion. The method constructs a fused situation data set with situation objects as basic units, and generates semantic roles, display priorities, and time-effect decay visualization constraints for the situation objects. Combining the spatial distribution, quantity density, semantic co-occurrence relationships, and dynamic update characteristics of situation objects, a visual conflict degree evaluation model is established to quantify the degree of visual resource occupation by the current visualization interface. When the visual conflict degree reaches a preset condition, by identifying the emergency response stage, the situation objects are semantically graded and their visualization constraints are dynamically adjusted, achieving graded release and adaptive presentation of situation information under different emergency stages. This invention can effectively alleviate the information overload problem in emergency response scenarios, highlight key information, and improve the efficiency of urban operation and maintenance situation awareness and decision support.

Description

Technical Field

[0001] This invention relates to the field of operation and maintenance visualization technology, specifically to a method and apparatus for visualizing urban operation and maintenance status based on multi-source data fusion. Background Technology

[0002] With the advancement of smart city construction, urban operation and maintenance management is gradually incorporating video surveillance systems, IoT sensing systems, alarm management systems, and various business support systems to monitor and manage the operational status of urban infrastructure, public safety incidents, and environmental changes. To enhance the overall perception capabilities of urban operation and maintenance, existing technologies typically use multi-source data fusion to aggregate data from different systems and devices onto a unified platform and present it in a visual format to operation and maintenance personnel or emergency command personnel.

[0003] However, due to the significant differences in time update frequency, spatial scale, data accuracy and semantic level among different data sources, when multi-source data are concentrated and presented in a unified visualization interface in a short period of time, it is easy to cause a sharp increase in the density of visualization information and competition among different types of situation information at the visual level, thereby causing key situation information to be submerged in a large amount of fused data.

[0004] Existing urban operation and maintenance situation visualization technologies primarily focus on improving data access capabilities or enhancing visualization effects. Typically, after multi-source data is fused, display control is implemented according to fixed rules or manual configuration, lacking effective means to address the information overload problem caused by multi-source fusion results in emergency scenarios. On the one hand, existing technologies fail to fully consider the capacity of fusion results for visualization presentation during the multi-source data fusion stage; on the other hand, existing visualization control methods struggle to adaptively adjust the fused situation based on dynamic changes in the emergency response process, easily affecting the timeliness and accuracy of emergency decision-making.

[0005] Therefore, there is an urgent need for a multi-source data fusion situation visualization method for emergency response scenarios. This method should ensure the integrity of multi-source data fusion while effectively alleviating the information overload problem that occurs during the visualization process, thereby improving the efficiency of emergency command in identifying and handling key situations. Summary of the Invention

[0006] In view of the above-mentioned shortcomings mentioned in the background art, the purpose of this invention is to provide a method and device for visualizing urban operation and maintenance status based on multi-source data fusion.

[0007] The first aspect of the present invention provides a method for visualizing urban operation and maintenance status based on multi-source data fusion. The method includes the following steps: S1, collecting multi-source data from the urban operation and maintenance system, including video surveillance data, IoT sensor data, alarm event data, and business system status data; S2, fusing the multi-source data according to a unified time window and spatial mapping rules to form a fused status data set with status objects as basic units, and generating corresponding visualization constraint information for them, including the semantic role identifier of the status object in the emergency scenario, the display priority weight, and the time-effect decay parameter; S3, visualizing the fused status based on the fused status data and its corresponding visualization constraint information, and calculating the visual conflict degree of the fused status according to the spatial distribution, quantity density, and update characteristics of the status objects in the fused status; S4, when the visual conflict degree meets preset conditions, dynamically adjusting the visualization constraint information of the status objects according to the changes in the emergency response stage, and adaptively adjusting the visualization presentation mode of the fused status accordingly to reduce the occupancy of non-critical status objects on the visualization interface.

[0008] A second aspect of this invention provides a city operation and maintenance situation visualization device based on multi-source data fusion. The device includes: a data acquisition module for acquiring multi-source data from the city operation and maintenance system, including video surveillance data, IoT sensor data, alarm event data, and business system status data; a fusion situation construction module for fusing the multi-source data according to a unified time window and spatial mapping rules to form a fusion situation data set with situation objects as basic units, and generating corresponding visualization constraint information for them, including the semantic role identifier of the situation object in the emergency scenario, the display priority weight, and the time-effect decay parameter; a visual conflict degree calculation module for visualizing the fusion situation based on the fusion situation data and its corresponding visualization constraint information, and calculating the visual conflict degree of the fusion situation according to the spatial distribution, quantity density, and update characteristics of the situation objects in the fusion situation; and an adaptive visualization adjustment module for dynamically adjusting the visualization constraint information of the situation objects according to the changes in the emergency response stage when the visual conflict degree meets preset conditions, and adaptively adjusting the visualization presentation mode of the fusion situation accordingly to reduce the occupancy of non-critical situation objects on the visualization interface.

[0009] A third aspect of the present invention provides a non-transitory computer-readable storage medium storing computer instructions for causing the computer to perform the method described in any of the preceding claims.

[0010] A fourth aspect of the present invention provides a computer program product including a computing program stored on a non-transitory computer-readable storage medium, the computer program including program instructions that, when executed by a computer, cause the computer to perform the method described in any of the preceding claims.

[0011] Compared with existing technologies, the present invention has at least the following beneficial technical effects: By integrating multi-source urban operation and maintenance data into a situational object with semantic roles and visual constraints, and introducing a visual conflict degree evaluation mechanism based on spatial aggregation, semantic attention competition, and dynamic occupancy, the present invention achieves quantitative perception and adaptive adjustment of the information load of the visual interface; during emergency response, it can accurately identify the response stage by combining alarm evolution, situational changes, and user interaction behavior, and accordingly present the situational object in a hierarchical manner and dynamically weaken non-critical content, thereby effectively avoiding the problems of information overload and the submersion of key situations, and improving the clarity, stability, and decision support efficiency of situational display in urban operation and maintenance emergency scenarios. Attached Figure Description

[0012] Figure 1 This is a schematic diagram of the overall process of a method for visualizing urban operation and maintenance status based on multi-source data fusion, as disclosed in an embodiment of the present invention.

[0013] Figure 2 This is a schematic diagram of the system architecture for implementing multi-source data fusion disclosed in an embodiment of the present invention.

[0014] Figure 3 This is a schematic diagram of the structure of an urban operation and maintenance status visualization device based on multi-source data fusion disclosed in an embodiment of the present invention. Detailed Implementation

[0015] The specific embodiments of this disclosure will be described in detail below with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are for illustration and explanation only and are not intended to limit this disclosure.

[0016] The solution described in this embodiment can be deployed on the integrated operation and maintenance server, city-level operation monitoring platform, industry regulatory business system, or cloud-based operation and maintenance management platform corresponding to the city operation and maintenance management center. It is used for unified perception, integrated analysis, and situational visualization of various types of infrastructure, public service systems, and operational activities involved in city operations. The operational objects include, but are not limited to, city video surveillance equipment, IoT sensing terminals, alarm and event management systems, business operation systems, and related data acquisition nodes. Their corresponding operational status, event information, and environmental parameters are acquired through multi-source data acquisition methods.

[0017] The management platform provides functional pages such as an overall view of the city's operational status, a multi-source data fusion status display view, a spatial distribution map of status objects, an emergency event correlation analysis view, a status change timeline analysis view, and an emergency response auxiliary decision-making interface. It provides a platform for displaying the results of multi-source data acquisition, fusion status generation, visualization, and adaptive adjustment in this invention (e.g., ...). Figure 2 (Integrated display screen and business entry point in the system).

[0018] For ease of understanding, the following embodiments use multiple operational areas and emergency response processes in urban operation and maintenance scenarios as examples. It should be noted that the urban operation objects are not limited to having direct physical connections, communication connections, or management affiliations. The correlation between the situational object and multi-source data described in this invention is constructed based on the actual operational status, event triggering conditions, and situational change characteristics within a specific time window and spatial range during urban operation and maintenance. Essentially, it involves the structured fusion and visual constraint modeling of the urban operation and maintenance situation formed by multi-source operational data in emergency scenarios, rather than an isolated display of the status of a single device or information about a single event.

[0019] Please see Figure 1 This invention provides a method for visualizing urban operation and maintenance status based on multi-source data fusion. The method includes the following steps: S1, collecting multi-source data from the urban operation and maintenance system, including video surveillance data, IoT sensor data, alarm event data, and business system status data; please refer to... Figure 2 This involves unified access and collection of operation and maintenance-related data scattered across different systems, devices, and business domains, resulting in multi-source data. This multi-source data includes at least the following categories: Video surveillance data: used to reflect the intuitive state of urban operations, such as road traffic conditions, crowd gatherings, and on-site footage of key areas. This type of data typically originates from video camera equipment deployed on urban roads, public places, or around important facilities, and its acquisition format can be real-time video streams, keyframe images, or video analysis results.

[0020] IoT sensor data: This data reflects environmental parameters and equipment operating parameters during urban operations, such as temperature, humidity, energy consumption, and equipment status signals. This type of data typically exists as periodic sampled values ​​and is associated with specific sensor installation locations or equipment nodes.

[0021] Alarm event data: This data reflects abnormal situations that have been identified by the system or reported manually during urban operations and maintenance, such as equipment failure alarms, environmental over-limit alarms, and security incident alarms. This type of data has obvious event triggering characteristics and usually includes information such as the time of event occurrence, event type, and event location.

[0022] Business system status data: This data reflects the operational status of urban operations and maintenance related business processes, such as equipment maintenance work order status, inspection task execution status, and dispatch instruction execution status. Although this type of data does not directly reflect the operational status of the physical world, it can supplement the information dimension of urban operations and maintenance from a management perspective.

[0023] Understandably, to ensure the feasibility of subsequent fusion processing, the collection of multi-source data must follow the following consistency principles: Time identifier consistency: The original time identifier of all types of collected data should be retained in a unified manner, and the time format should be standardized when necessary to ensure that data from different data sources can be aligned and analyzed within a unified time window.

[0024] Spatial correlation traceability: For data related to specific spatial locations, such as video surveillance data, IoT sensor data, and alarm event data, the correlation information between the data and the spatial location or spatial region is retained during the collection stage, providing a basis for subsequent spatial mapping and situational object construction.

[0025] Data source distinguishability: During the data collection process, data from different data sources are labeled with corresponding data source type identifiers to avoid losing data source information during subsequent fusion, thereby ensuring the interpretability of the fusion situation analysis results.

[0026] S2. The multi-source data is fused according to a unified time window and spatial mapping rules to form a fused situational data set with situational objects as the basic units, and corresponding visualization constraint information is generated for them, including the semantic role identifier of the situational object in the emergency scenario, the display priority weight, and the time-effect decay parameter. Unlike traditional multi-source data fusion, which only focuses on the data content itself, this invention not only completes the fusion modeling of multi-source data in the time and space dimensions, but also simultaneously introduces semantic and constraint information for subsequent situational visualization, so that the fusion result has the ability to be selectively presented.

[0027] As an example, the multi-source data is fused according to a unified time window and spatial mapping rules to form a fused situational data set with situational objects as the basic unit, and corresponding visualization constraint information is generated for it. This includes: S21, performing time alignment processing on data from different data sources that are within the same time window, and mapping the data to the corresponding spatial region according to a preset spatial mapping rule. In this embodiment, the data from different data sources have significant differences in sampling period, triggering method, and time precision. For example, video surveillance data is usually generated in the form of continuous frames or second-level frame sequences, IoT sensor data is often reported with a fixed sampling period, and alarm event data is generated by event triggering, and its time distribution has obvious dispersion. If the above data is directly fused and analyzed, it is easy to cause situational association errors due to inconsistent time bases.

[0028] To address this, the present invention introduces a unified time window mechanism to align the collected multi-source data in time. Specifically, a fixed-length or sliding time window can be preset, and data whose timestamps fall within the same time window are considered valid data within the same maintenance period. For example, when the time window is set to 10 seconds, video keyframes, sensor data sampling values, and alarm event records collected within this 10-second interval are all included in the same time window for subsequent processing.

[0029] After time alignment is completed, the data is further processed by spatial mapping according to preset spatial mapping rules. The spatial mapping rules are used to establish the correspondence between data and spatial regions. For example, different data can be mapped to a unified spatial region identifier based on geographic coordinates, administrative divisions, functional zones, or gridded spatial models.

[0030] S22. Based on time alignment and spatial mapping results, data associated with the same spatial region are aggregated into a single situational object, and the situational object is associated with the corresponding data source type and status characteristics. In this embodiment, the situational object serves as the basic unit for multi-source data fusion, carrying operational information from different data sources within the same spatial region and the same time window. It should be noted that the situational object is not equivalent to a single physical device or sensor node, but rather a virtualized description unit formed by logically aggregating multi-source operational data within the same spatial region and the same time window in urban operations and maintenance.

[0031] For example, a city intersection may simultaneously be associated with video surveillance footage, pedestrian or vehicle flow sensor data, traffic anomaly alarms, and status information from relevant business systems within the same time window. By aggregating these data into a single situational object, data from different sources and in different formats can be logically organized in a unified manner, thereby forming a comprehensive description of the operational status of that spatial area.

[0032] In this invention, the situational object is instantiated as a visual aggregation unit: In the interface, a specific aggregation icon floats above the intersection, dynamically displaying key information in the form of thumbnails or labels—a real-time video keyframe is embedded in the upper left corner, the current traffic flow value and trend arrow are displayed in the upper right corner, and the presence of alarms is highlighted below (e.g., flashing red indicates an accident), along with the status of the most recent work order. Clicking the icon expands to a detailed view, displaying all the aggregated raw data and the semantic roles, priority weights, and time-decrease parameters generated in step S2. Through this instantiation and labeling, the situational object becomes a direct entry point for users to perceive and operate the city's operational situation.

[0033] When constructing a situational awareness object, this invention associates each situational awareness object with a corresponding data source type identifier to record the data source composition contained in the situational awareness object. Simultaneously, it also associates the situational awareness object with corresponding state characteristics or state change characteristics, such as the current state of sensor values, whether an alarm event has occurred, and whether the state shows an upward or downward trend.

[0034] S23. Based on the corresponding data source type, state change characteristics, and role in emergency response, determine the semantic role identifier for the situational object. In this embodiment, in urban operation and maintenance emergency scenarios, not all situational objects have the same level of importance in the decision-making process. For example, situational objects that directly trigger abnormal alarms often need to be given priority attention, while situational objects used only for auxiliary observation or background perception do not necessarily occupy the main visual resources.

[0035] To address this, the present invention introduces a semantic role identification mechanism to classify the role of situational objects in emergency response from a semantic perspective. Specifically, different semantic role identifiers can be assigned to situational objects based on the type of data source they are associated with, the severity of their state changes, and whether they are directly related to the current emergency event.

[0036] For example, if a situational object is associated with an abnormal alarm event within the current time window, and its state change characteristics show a rapid transition from stable to abnormal, then the situational object can be identified as "decision-triggered level"; if a situational object does not directly trigger an alarm, but its associated video footage or sensor data can provide important evidence for judgment, then it can be identified as "evidence-supported level"; and if a situational object is only used to provide environmental or operational background information, then it can be identified as "background perception level".

[0037] By using the aforementioned semantic role identifiers, the importance of different situational objects can be distinguished at the fusion situational level, thereby avoiding the indiscriminate display of all situational objects in the subsequent visualization process.

[0038] S24. Based on the semantic role identifier and the data freshness and credibility of the situation object, assign corresponding display priority weights and timeliness decay parameters to the situation object, thereby generating the visualization constraint information of the situation object.

[0039] In this embodiment, the visualization constraint information is used to describe how the situation object should occupy visual resources during the visualization process. It does not change the data content of the situation object itself, but serves as a constraint condition for subsequent visualization processing.

[0040] Specifically, in the scenario of multi-source data fusion for urban operations and maintenance, the data associated with different situational objects exhibit significant differences in update timeliness, timeliness, source reliability, and stability. Therefore, this embodiment introduces data freshness and credibility as important adjustment factors when generating visual constraint information.

[0041] Data freshness, in particular, reflects the temporal validity of data associated with a situational object and can be characterized based on the time difference between the current time and the most recent data update time. As an example, a situational object... Data freshness can be expressed as: ;in, The most recent data update time for the situational object. This is the attenuation coefficient. The smaller the time difference, the lower the attenuation coefficient. The larger the value, the fresher the data.

[0042] Data credibility reflects the overall level of accuracy, stability, and source reliability of the data associated with a situational object. It can be evaluated by combining factors such as data source type, historical anomaly rate, and data consistency verification results. As an example, a normalized credibility index can be defined for a situational object: Data from authoritative sensing devices and stable operating systems is more reliable, while data from sources that are susceptible to interference or have incomplete data is relatively less reliable.

[0043] Examples are as follows: Authoritative sensing equipment and stable operating systems: such as fixed video surveillance equipment deployed in transportation hubs (using dedicated fiber optic lines, equipped with UPS power supplies, and calibrated annually), its source type label Si=1.0; the equipment has an extremely low historical anomaly rate (99.9% data integrity rate in the past 30 days), Hi=0.999; the current data is consistent with cross-validation of adjacent equipment, Vi=1.0. After weighting... Close to 1.0.

[0044] Sources susceptible to interference or with missing data: such as solar-powered wireless sensor nodes deployed in remote areas (affected by weather, signal is unstable), their source type label Si=0.6; this node has a high historical anomaly rate (15% missing data rate in the past 30 days), Hi=0.85; current data deviates from other reference points, consistency check result is weak consistency Vi=0.6. After weighting... The value is approximately 0.68 (assuming equal weights). Or, in more extreme cases, such as manually reported data (Si=0.4), which cannot be verified (Vi=0.3), and has a high historical anomaly rate,... Lower.

[0045] In this embodiment, the priority weight is used to indicate the degree to which a situational object is presented in the visualization interface. It is determined not only by the semantic role identifier of the situational object but also by comprehensively considering its data freshness and reliability, in order to avoid misleading decisions due to outdated or unreliable data. As an example, a situational object... Display priority weight It can be determined in the following way: ;in, For semantic role identification Determined basic weights, This is the adjustment coefficient.

[0046] In this way, even among decision-triggered situational objects, those with timely data updates and reliable sources will still be given higher display priority than those with outdated data or lower credibility, thereby enhancing the supporting value of visualization results for actual decision-making.

[0047] In this embodiment, the time-decrease parameter represents the rate at which the visual importance of a situational object decreases over time. It describes the rate at which the visual importance of a situational object decreases over time, and its setting also comprehensively considers semantic role, freshness, and credibility. As an example, a situational object... Time-degradation parameters It can be represented as: ;in, The base decay coefficient corresponding to the semantic role, It is a regulating factor.

[0048] Therefore, situational objects with high data freshness and credibility decay slowly and can remain visible in the visualization interface for a longer period of time; while situational objects with lagging data updates or low credibility will decay more quickly in terms of visualization importance, thus avoiding outdated or unreliable information from occupying visual resources for a long time.

[0049] For example, in the same emergency, two situational objects are both identified as decision-triggered. One comes from a real-time sensing device and is continuously updated, while the other comes from a manual report and has not been updated for a long time. By introducing a mechanism to adjust for data freshness and reliability, the former will have a significantly higher display priority and timeliness decay than the latter, thus ensuring that the visualization interface prioritizes information with greater decision-making value.

[0050] S3. Based on the fused situation data and its corresponding visualization constraint information, the fused situation is visualized, and the visual conflict degree of the fused situation is calculated according to the spatial distribution, quantity density and update characteristics of the situation objects in the fused situation. Visual conflict intensity is used to characterize the overall occupancy of visual presentation resources by the fused situation within the current visualization view. This occupancy is not only related to the number of situation objects, but also closely related to the spatial concentration, semantic importance, and dynamic change characteristics of the situation objects. Therefore, this embodiment models visual conflict intensity hierarchically according to three dimensions: spatial clustering characteristics, semantic attention competition intensity, and dynamic occupancy intensity.

[0051] For ease of explanation, let's assume the spatial area covered by the current visualization view is... In this view, the set of situational objects in the fused situation is denoted as: , where each situational object All are related to their spatial location Semantic role identifier And the update behavior within the preset time window.

[0052] As an example, the visual conflict degree of the fused situation is calculated based on the spatial distribution, quantity density, and update characteristics of situational objects in the fused situation. This includes: S31, within the current visualization view range, spatially dividing the situational objects in the fused situation, and statistically analyzing the distribution of situational objects within each spatial region to obtain the spatial clustering characteristics of the situational objects. In this embodiment, in the urban operation and maintenance visualization interface, the visual congestion perceived by the user often originates from the concentrated distribution of situational objects within a local spatial region, rather than the total number of objects within the view range. Therefore, this embodiment first uses the current visualization view range as the spatial analysis boundary and spatially divides this range.

[0053] In this embodiment, the view range can be... Divide into K non-overlapping spatial regions: Subsequently, for each spatial region Count the number of situational objects within it: ;in, This is an indicator function used to determine whether a situational object falls within this spatial region.

[0054] Since simply counting the number of objects in each region is insufficient to reflect the degree of clustering, this embodiment further introduces relative clustering. As an example, the object density within a unit spatial region can be expressed as: ;in, Represents the area of ​​a spatial region or the pixel area mapped onto the screen.

[0055] Furthermore, to describe whether there are obvious local hotspots, this embodiment can define spatial clustering features. for: ;in, To prevent tiny positive numbers with a denominator of zero.

[0056] When situational objects are highly concentrated in certain spatial areas It will be significantly higher than the average density, thus making An increase indicates that there is a significant spatial clustering phenomenon in the current view.

[0057] S32. Calculate the quantity density of situational objects within a unit spatial region based on the spatial clustering characteristics; and determine the intensity of visual attention competition caused by the simultaneous appearance of situational objects of different semantic importance within the same spatial region by analyzing the co-occurrence relationship of situational objects with different semantic roles in the same spatial region; in this embodiment, in each spatial region Internal, density of situational objects This reflects the baseline visual load level of the area. Generally speaking, The larger the density, the more visual space the area occupies, and the higher the potential risk of occlusion and overlap. However, in emergency response scenarios, density cannot fully reflect the true degree of visual conflict because the semantic importance of different situational objects varies.

[0058] In this embodiment, each situational object All were assigned semantic role identifiers For example, there are decision-triggered (T) level, evidence-supported (E) level, and background-aware (B) level. When multiple semantically important situational objects co-occur in the same spatial region, even if the number of objects is small, it will trigger significant visual attention competition. To describe this phenomenon, this embodiment describes each spatial region... Within, count the number of situational objects with different semantic roles: ;in, .

[0059] Subsequently, to reflect the different impacts of different semantic roles on attention, weights were assigned to each semantic role. .

[0060] The weighting of these situations follows these rules: Decision-triggered situational objects are directly related to emergency events and play a decisive role in decision-making, therefore they should be given the highest weight; Evidence-supporting situational objects provide auxiliary information for decision-making and are of significant value for situation understanding, therefore they should be given a medium weight; Background awareness situational objects only reflect the environment or normal operational conditions and have limited impact on current emergency decisions, therefore they should be given the lowest weight. For example, the weight of decision-triggered situational objects can be set. The weight of the evidence-supported situational object is 1.0. The weight of the background awareness level situational object is 0.5. It is 0.2.

[0061] Furthermore, it establishes the intensity of visual attention competition at the regional level: When multiple decision-triggered or evidence-supported situational objects exist simultaneously within the same spatial region, This will increase significantly, reflecting the enhanced semantic attention competition within the region.

[0062] S33. Analyze the update characteristics of situational objects in the fused situational awareness, and extract the update frequency, update synchronization degree, and update duration of situational objects within a preset time window to determine the dynamic occupancy intensity of dynamic situational objects on the visualization view. In this embodiment, dynamic changes themselves are also an important source of visual burden in emergency scenarios. Therefore, this embodiment statistically analyzes the update behavior of situational objects within a preset time window W.

[0063] 1) Update frequency: For each situational object Within a time window W, count the number of times the object's state is updated, and calculate the normalized value of the proportion of that object. This is used to reflect the frequency of its changes.

[0064] 2) Update Synchronization Level: To describe whether multiple situational objects change intensively within the same time period, this embodiment can count the proportion of objects that are synchronously updated within window W, and form a synchronization metric based on the normalized value of this proportion. .

[0065] 3) Update duration: For situations that are constantly changing, record the duration of continuous updates, and the normalized value of this duration. It is used to reflect the persistence of dynamic stimuli.

[0066] Based on the above update features, dynamic occupancy intensity can be... Represented as: ;in, , , These are weighting parameters used to balance the influence of different dynamic factors.

[0067] S34. Based on the spatial clustering features, and taking the visual attention competition intensity as a sub-indicator representing semantic hierarchical conflict, and the dynamic occupancy intensity as a sub-indicator representing dynamic change occupancy, a comprehensive model is constructed to evaluate the occupancy of visual presentation resources in the current visualization view of the fusion situation, thereby constructing a visual conflict degree evaluation index for the fusion situation.

[0068] In this embodiment, visual conflict degree Designed as a comprehensive measure of multiple conflict sources, its inputs include spatial clustering characteristics. Regional-level attention competition intensity and dynamic occupancy intensity .

[0069] As an example, we can first aggregate the intensity of visual attention competition by region: This is used to characterize the most competitive area within the current view.

[0070] Subsequently, an evaluation index for visual conflict level was constructed: ,in , , , which is a weighting coefficient used to balance the impact of spatial clustering, semantic competition, and dynamic occupancy on overall visual conflict.

[0071] when When the threshold is exceeded, it indicates that the current fusion situation is consuming a high amount of visualization resources and there is a risk of information overload. In the subsequent step S4, the visualization constraint information of the situation object needs to be adaptively adjusted.

[0072] S4. When the visual conflict level meets the preset conditions, the visualization constraint information of the situation object is dynamically adjusted according to the changes in the emergency response stage, and the visualization presentation mode of the fused situation is adaptively adjusted accordingly to reduce the occupation of the visualization interface by non-critical situation objects.

[0073] When the visual conflict level meets the preset conditions, the visualization constraint information of the situation object is dynamically adjusted according to the changes in the emergency response stage, and the visualization presentation method of the fused situation is adaptively adjusted accordingly to reduce the occupation of the visualization interface by non-critical situation objects.

[0074] In this embodiment, the preset condition can be triggered by a threshold. This is combined with the aforementioned visual conflict level. Trigger threshold can be set. ,when If the current visualization view is deemed to have an information overload risk, the dynamic adjustment process in step S4 needs to be executed; when If the current rendering strategy is maintained, or a slow recovery strategy is implemented (e.g., gradually increasing the upper limit of background object transparency) to avoid frequent interface jitter. It should be noted that this invention does not set a threshold. It can be a fixed constant, or it can be dynamically configured according to time period, region or emergency level, but in this embodiment, its feasibility is explained in terms of threshold.

[0075] As an example, the visualization constraints of the situational awareness object are dynamically adjusted according to the changes in the emergency response phase. This includes: S41, identifying the current stage of the emergency response based on the triggering frequency of alarm events in the fused situational awareness, the state change characteristics of the situational awareness object, and user interaction behavior; after a visual conflict is triggered, not only should the display be reduced, but the content that should be highlighted should also be clearly defined. The focus of emergency response differs at different stages: for example, in the confirmation stage, it is more important to highlight the alarm source and evidence chain; in the execution stage, it is more important to highlight the response object and key impact areas; and in the stabilization and recovery stage, it is more important to highlight the recovery trend and residual risks. Without stage identification, uniform weakening can easily lead to the mis-suppression of key information.

[0076] Therefore, this embodiment further introduces the following three types of heterogeneous signals: 1) alarm event signals (reflecting event activity and evolution trend); 2) situation object state change signals (reflecting situation evolution intensity); 3) user interaction behavior signals (reflecting the intensity of human intervention and attention shift).

[0077] As an example, based on the triggering frequency of alarm events, the state change characteristics of situational objects, and user interaction behavior in the integrated situational awareness, the current stage of emergency response is identified, including: S411, within a preset sliding time window, performing time series analysis on alarm events in the integrated situational awareness, extracting the triggering frequency, frequency change rate, and alarm type distribution characteristics of alarm events to form alarm time series characteristics used to characterize the activity level and evolution trend of emergency events; in specific implementation, the sliding time window is set as... Window length is Within this window, the set of alarm events is denoted as... The alarm trigger frequency can be defined as: To reflect the trend of alarm activity over time, a frequency change rate (differential form) can be defined: Furthermore, to describe whether the alarm type structure has changed (e.g., from general alarms to high-risk alarms), the alarm types can be statistically analyzed within the window. Percentage: Thus, the alarm type distribution vector is obtained. .

[0078] The above three types of features together form the alarm time series features: Among them, when High and This usually corresponds to a period when the emergency is in its escalating phase; when Decline and This usually corresponds to the emergency situation tending to converge or entering a stable recovery phase.

[0079] S412. Within the sliding time window, the state change characteristics of the situational object in the fused situation are analyzed in segments. The direction, magnitude, and duration of change of the situational object during its evolution from a stable state to an abnormal state are extracted, and a situational evolution intensity index reflecting the degree of situational evolution is calculated accordingly. In specific implementation, let the situational object... In the window The internal state time series is , This can originate from sensor values, alarm level mapping values, or service status coding values. To describe the evolution from stable to abnormal, the abnormal deviation amount can be defined using a threshold or baseline deviation method: ,in The stable baseline for the object (which can be obtained from the historical mean / median of stationary periods). The direction of change can be determined by... The sign and trend are determined; the magnitude of change can be represented by the maximum deviation within the window: The duration can be used if it deviates from the threshold. The time percentage or continuous duration is expressed as follows, for example: To form a single, comparable indicator of the intensity of situational evolution, the magnitude and persistence of each object can be combined and aggregated across all objects: ;in, , These are the weight parameters.

[0080] Among them, when most of the situational objects show continuous deviation or the deviation increases rapidly, The increase indicates that the situation is evolving rapidly, corresponding to the peak or spread of the response.

[0081] S413. Synchronously collect user interaction behaviors in the situation visualization interface, and perform behavior sequence analysis on the interaction behaviors to construct user interaction behavior trajectories of user attention areas and objects over time, which are used to characterize the degree of human intervention and its changing trend. In specific implementation, user interaction behaviors include at least: view switching, zooming, panning, area selection, object clicking / locking, layer filtering, etc. The above interaction behaviors directly reflect the user's attention to specific spatial areas or situation objects in the situation visualization interface.

[0082] Set within a preset sliding time window The set of user interaction events collected internally is Each interaction event corresponds to a specific user attention action. Therefore, the user's interaction frequency within this time window can be defined as: ; in, This indicates the length of the sliding time window. The interaction frequency can provide a preliminary indication of the level of human intervention activity. A higher level usually indicates that the user is intensively involved in the situation analysis or decision-making process.

[0083] To further characterize whether user attention is focused on a few key areas or objects, this embodiment divides the current visualization view into several spatial regions according to the spatial division method in step S31. Statistics within the time window Within the space, the proportion of user interaction events falling into each spatial area is defined as: ;in, To prevent tiny positive numbers with a denominator of zero.

[0084] Based on this, information entropy can be introduced to quantify the spatial concentration of user attention, and user attention entropy is defined as: ;when When the value is low, it indicates that user interaction is mainly concentrated in a few spatial areas, usually corresponding to the user having identified a clear target or key situational object; when A higher value indicates that users are paying attention to scattered or frequently switching between different points, which usually corresponds to an unclear emergency situation or the need for multiple parallel handling points.

[0085] In addition to the degree of focus, the spatial migration of user attention can also reflect changes in the method of human intervention. Therefore, this embodiment further constructs a user attention trajectory.

[0086] Specifically, the focus location (e.g., the center of the view or the position of the locked pose object) corresponding to each user interaction can be denoted as... Then in the time window Within this context, the user's attention trajectory can be represented as: .

[0087] To quantify the strength of attention migration, the average movement distance between adjacent attention centers can be calculated, and the intensity of user attention migration can be defined as: ;when When the value is small, it indicates that the user's attention remains relatively stable in space, corresponding to continuous analysis or operation around the same object of disposal; when A larger value indicates that user attention is frequently shifting between multiple regions, corresponding to a stage of uncertainty, risk diffusion, or simultaneous handling of multiple objectives.

[0088] Through the above interaction frequency Focus on entropy and migration intensity This embodiment transforms the original user interaction behavior into quantifiable feature indicators, which are used to characterize the degree of human intervention and its changing trends from different dimensions.

[0089] It is understandable that the above-mentioned feature indicators can be used as user-side behavioral features, together with the alarm event time series features and situation evolution intensity indicators, to form a joint judgment input for emergency response stage identification, thereby avoiding relying solely on the system's automatic detection results and ignoring key intervention information in the human decision-making process.

[0090] S414. Based on the alarm time series characteristics, the situation evolution intensity index, and the user interaction behavior trajectory, a joint judgment model for the emergency response stage is constructed, and the stability of the joint judgment results within a continuous time window is analyzed to determine the current stage of the emergency response.

[0091] The above three types of features are summarized into a joint feature vector. And a stage-based scoring function is constructed based on joint features. As an example, it can be applied to each stage. Define score: ,in This represents a normalization or nonlinear transformation of the feature. This is the stage parameter vector. The initial stage judgment result can be taken as the stage corresponding to the highest score: .

[0092] To prevent frequent stage transitions within adjacent windows (e.g., brief alarm fluctuations causing a jump from "handling" back to "confirmation"), this embodiment introduces stability analysis. As an example, the consistency of judgment results can be statistically analyzed within the most recent L windows: When stability meets the threshold condition When the current stage is reached, output the current stage: Otherwise, maintain the previous stable phase. Alternatively, a transition state can be output to ensure that stage identification is available and does not jitter.

[0093] S42. Based on the identified emergency response stage, semantically classify the situational objects in the fused situational data, classifying them into at least three levels: decision-triggered situational objects, evidence-supported situational objects, and background-aware situational objects; the results of stage identification. This will directly impact the tiered strategy: the visualization role of the same object may change at different stages. For example, in the confirmation stage, the alarm source and the chain of evidence are the most critical; in the handling stage, the handling target, the area of ​​impact, and key linkage facilities are more critical; and in the recovery stage, trend-following objects and objects with residual risks are more critical.

[0094] To make this mechanism computable, this embodiment can calculate the stage-related importance of each object in the current stage: ;in: Indicates the degree of relevance of the object to the current event chain (e.g., whether it is located in the alarm impact domain or whether it matches the alarm type). Indicates the risk intensity of the object (which can be derived from the alarm level and deviation range). Continuity wait); This indicates the ability to provide supporting evidence (e.g., whether videos, key sensors, or key business status are provided).

[0095] Then, semantic levels are defined based on thresholds or ranking. As an example, the top few can be ranked by importance as decision-triggered levels, the middle as evidence-supported levels, and the remainder as background awareness levels. , , ;in, , Can be carried out in stages Adjustments (e.g., handling peak periods) Smaller for greater focus.

[0096] S43. Based on the semantic classification results of the situation objects, update the corresponding visualization constraint information so that situation objects at different levels have different display priorities, display methods and refresh strategies in the visualization process, so as to realize the hierarchical release visualization of the integrated situation under different emergency response stages.

[0097] Each situational object With visual constraint information already available (displaying priority weights) Time-related decay parameters (etc.). In this step, these constraints will be phased out. The update is driven together with the semantic classification results.

[0098] Demonstrating phased updates of priority weights: Assuming the object semantic hierarchy is... Then, new priority weights can be defined: ;in, Used to reflect differences in grading Used to reflect the overall strategy of a stage (e.g., increasing the overall weight of key categories during the disposal stage).

[0099] Adaptive updating of decay parameters: For background objects, the decay rate can be increased to reduce their long-term occupancy; for critical objects, the decay rate can be decreased to ensure visibility. For example... ,in This indicates that background objects decay faster.

[0100] Linked control of display mode and refresh strategy: under the condition of visual conflict degree triggering ( This embodiment breaks down the reduction of non-critical resource usage into three actionable steps: weakening, aggregation, and frequency reduction / delay, corresponding to display mode and refresh strategy parameters respectively: weakening display (transparency / contrast): ,in The background object has lower opacity (lighter).

[0101] Aggregation rendering (grouping similar objects in the same area into clusters): For background-aware objects, they are rendered by spatial region. Aggregation display, its aggregation threshold can be related to density Related, for example when Aggregation is enabled at that time.

[0102] Reduced refresh rate / delayed refresh (reduces dynamic stimulation): Background objects refresh at a lower frequency, while critical objects maintain a higher refresh rate.

[0103] The above updates ensure that critical objects remain visible and traceable even in situations with high conflict levels, while non-critical objects are weakened, aggregated, or have their refresh rate reduced, thereby reducing the occupancy of non-critical situational objects.

[0104] Please see Figure 3 This invention also provides a city operation and maintenance situation visualization device 100 based on multi-source data fusion, comprising: a data acquisition module 10, used to collect multi-source data from the city operation and maintenance system, including video surveillance data, IoT sensor data, alarm event data, and business system status data; a fusion situation construction module 20, used to fuse the multi-source data according to a unified time window and spatial mapping rules to form a fusion situation data set with situation objects as basic units, and generate corresponding visualization constraint information for them, including the semantic role identifier, display priority weight, and time-effect decay parameter of the situation object in the emergency scenario; a visual conflict degree calculation module 30, used to visualize the fusion situation based on the fusion situation data and its corresponding visualization constraint information, and calculate the visual conflict degree of the fusion situation according to the spatial distribution, quantity density, and update characteristics of the situation objects in the fusion situation; and an adaptive visualization adjustment module 40, used to dynamically adjust the visualization constraint information of the situation objects according to the changes in the emergency response stage when the visual conflict degree meets preset conditions, and adaptively adjust the visualization presentation mode of the fusion situation accordingly to reduce the occupation of the visualization interface by non-critical situation objects.

[0105] As an example, the fusion situation construction module 20 is configured to: perform time alignment processing on data from different data sources that are within the same time window, and map the data to the corresponding spatial region according to a preset spatial mapping rule; based on the time alignment and spatial mapping results, aggregate data associated with the same spatial region into the same situation object, and associate the situation object with the corresponding data source type and state characteristics; determine the semantic role identifier for the situation object according to the corresponding data source type, state change characteristics, and role in emergency response; and assign the situation object a corresponding display priority weight and timeliness decay parameter based on the semantic role identifier and the data freshness and credibility of the situation object, thereby generating the visualization constraint information of the situation object.

[0106] As an example, the visual conflict degree calculation module 30 is configured to: spatially divide the situational objects in the fused situation within the current visualization view range, statistically analyze the distribution of situational objects in each spatial region to obtain the spatial clustering characteristics of the situational objects; calculate the quantity density of situational objects in a unit spatial region based on the spatial clustering characteristics; and determine the intensity of visual attention competition caused by the simultaneous appearance of situational objects of different semantic importance in the same spatial region by analyzing the co-occurrence relationship of situational objects with different semantic roles in the same spatial region; analyze the update characteristics of situational objects in the fused situation, extract the update frequency, update synchronization degree, and update duration of situational objects within a preset time window to determine the dynamic occupancy intensity of dynamic situational objects on the visualization view; based on the spatial clustering characteristics, and using the visual attention competition intensity as a sub-indicator representing semantic hierarchical conflict, and the dynamic occupancy intensity as a sub-indicator representing dynamic change occupancy, comprehensively model the occupancy of visual presentation resources of the fused situation under the current visualization view, and construct a visual conflict degree evaluation index for the fused situation.

[0107] As an example, the adaptive visualization adjustment module 40 is configured to: identify the current stage of emergency response based on the triggering frequency of alarm events, the state change characteristics of situational objects, and user interaction behavior in the fused situational data; perform semantic classification of situational objects in the fused situational data according to the identified emergency response stage, classifying situational objects into at least decision-triggered situational objects, evidence-supported situational objects, and background-aware situational objects; and update the corresponding visualization constraint information based on the semantic classification results of situational objects, so that situational objects of different levels have different display priorities, display methods, and refresh strategies during the visualization presentation process, so as to realize the hierarchical release visualization of the fused situational data under different emergency response stages.

[0108] As an example, the adaptive visualization adjustment module 40 is configured to: perform time series analysis on alarm events in the fusion situation within a preset sliding time window, extract the trigger frequency, frequency change rate, and alarm type distribution characteristics of alarm events to form alarm time series features for characterizing the activity level and evolution trend of emergency events; perform segmented analysis on the state change characteristics of situation objects in the fusion situation within the sliding time window, extract the direction, magnitude, and duration of change of situation objects from stable to abnormal states, and calculate the situation evolution intensity index reflecting the degree of situation evolution accordingly; synchronously collect user interaction behavior in the situation visualization interface, and perform behavior sequence analysis on the interaction behavior to construct user interaction behavior trajectories of user attention areas and objects of attention changing over time, for characterizing the degree of human intervention and its changing trend; construct a joint judgment model for the emergency response stage based on the alarm time series features, the situation evolution intensity index, and the user interaction behavior trajectory, and determine the current stage of emergency response by analyzing the stability of the joint judgment results within a continuous time window.

[0109] This invention also provides a non-transitory computer-readable storage medium that stores computer instructions for causing the computer to perform any of the methods described above.

[0110] This invention also provides a computer program product, including a computing program stored on a non-transitory computer-readable storage medium, the computer program including program instructions that, when executed by a computer, cause the computer to perform the method described in any of the preceding claims.

[0111] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for visualizing urban operation and maintenance status based on multi-source data fusion, characterized in that, The method includes the following steps: S1, collecting multi-source data from the urban operation and maintenance system, including video surveillance data, IoT sensor data, alarm event data, and business system status data; S2, fusing the multi-source data according to a unified time window and spatial mapping rules to form a fused situation data set with situation objects as basic units, and generating corresponding visualization constraint information for it, including the semantic role identifier of the situation object in the emergency scenario, the display priority weight, and the time-effect decay parameter; S3, based on the fused situation data and its corresponding visualization constraint information, visually presenting the fused situation, and calculating the visual conflict degree of the fused situation according to the spatial distribution, quantity density, and update characteristics of the situation objects in the fused situation; S4. When the visual conflict level meets the preset conditions, the visualization constraint information of the situation object is dynamically adjusted according to the changes in the emergency response stage, and the visualization presentation mode of the fused situation is adaptively adjusted accordingly to reduce the occupation of the visualization interface by non-critical situation objects.

2. The method for visualizing urban operation and maintenance status based on multi-source data fusion according to claim 1, characterized in that: The multi-source data is fused according to a unified time window and spatial mapping rules to form a fused situational data set with situational objects as the basic unit, and corresponding visualization constraint information is generated for them. This includes: S21, performing time alignment processing on data from different data sources that are within the same time window, and mapping the data to the corresponding spatial region according to a preset spatial mapping rule; S22, based on the time alignment and spatial mapping results, aggregating data associated with the same spatial region into the same situational object, and associating the situational object with the corresponding data source type and state characteristics; S23, determining the semantic role identifier for the situational object according to the corresponding data source type, state change characteristics, and role in emergency response; S24, based on the semantic role identifier and the data freshness and credibility of the situational object, assigning the situational object a corresponding display priority weight and timeliness decay parameter, thereby generating the visualization constraint information of the situational object.

3. The method for visualizing urban operation and maintenance status based on multi-source data fusion according to claim 2, characterized in that: Based on the spatial distribution, quantity density, and update characteristics of situational objects in the fusion situation, the visual conflict degree of the fusion situation is calculated, including: S31, within the current visualization view, the situational objects in the fusion situation are spatially divided, and the distribution of situational objects in each spatial region is statistically analyzed to obtain the spatial clustering characteristics of situational objects; S32, based on the spatial clustering characteristics, the quantity density of situational objects in a unit spatial region is calculated; and by analyzing the co-occurrence relationship of situational objects with different semantic roles in the same spatial region, the intensity of visual attention competition caused by the simultaneous appearance of situational objects with different semantic importance in the spatial region is determined; S33, the update characteristics of situational objects in the fusion situation are analyzed, and the update frequency, update synchronization degree, and update duration of situational objects within a preset time window are extracted to determine the dynamic occupancy intensity of dynamic situational objects on the visualization view; S34, based on the spatial clustering characteristics, and using the visual attention competition intensity as a sub-indicator representing semantic hierarchical conflict, and the dynamic occupancy intensity as a sub-indicator representing dynamic change occupancy, a comprehensive model is performed on the occupancy of visual presentation resources of the fusion situation under the current visualization view to construct a visual conflict degree evaluation index for the fusion situation.

4. The method for visualizing urban operation and maintenance status based on multi-source data fusion according to claim 3, characterized in that: Based on the changes in the emergency response phase, the visualization constraint information of the situation objects is dynamically adjusted, including: S41, identifying the current emergency response phase based on the triggering frequency of alarm events, the state change characteristics of situation objects, and user interaction behavior in the fused situation data; S42, semantically classifying the situation objects in the fused situation data according to the identified emergency response phase, classifying the situation objects into at least decision-triggered situation objects, evidence-supported situation objects, and background-aware situation objects; S43, updating the corresponding visualization constraint information based on the semantic classification results of the situation objects, so that situation objects of different levels have different display priorities, display methods, and refresh strategies in the visualization process, so as to realize the hierarchical release visualization of the fused situation under different emergency response phases.

5. The urban operation and maintenance status visualization method based on multi-source data fusion according to claim 4, characterized in that: Based on the triggering frequency of alarm events, the state change characteristics of situational objects, and user interaction behavior in the integrated situational awareness, the current stage of emergency response is identified, including: S411, within a preset sliding time window, performing time series analysis on alarm events in the integrated situational awareness, extracting the triggering frequency, frequency change rate, and alarm type distribution characteristics of alarm events to form alarm time series characteristics used to characterize the activity level and evolution trend of emergency events; S412, within the sliding time window, performing segmented analysis on the state change characteristics of situational objects in the integrated situational awareness, extracting the direction of change, change rate, and change characteristics of situational objects during the evolution process from a stable state to an abnormal state. The system calculates the magnitude and duration of the situation evolution and uses this to determine the situation evolution intensity index, which reflects the degree of situation evolution. S413: Simultaneously collect user interaction behavior in the situation visualization interface and perform behavior sequence analysis on the interaction behavior to construct user interaction behavior trajectories of the user's attention area and object of attention over time, used to characterize the degree of human intervention and its changing trend. S414: Based on the alarm time sequence characteristics, the situation evolution intensity index, and the user interaction behavior trajectory, construct a joint judgment model for the emergency response stage, and determine the current stage of emergency response by analyzing the stability of the joint judgment results within a continuous time window.

6. A city operation and maintenance status visualization device based on multi-source data fusion, characterized in that: The device includes: a data acquisition module for acquiring multi-source data from the urban operation and maintenance system, including video surveillance data, IoT sensor data, alarm event data, and business system status data; a fusion situation construction module for fusing the multi-source data according to a unified time window and spatial mapping rules to form a fusion situation data set with situation objects as basic units, and generating corresponding visualization constraint information for them, including the semantic role identifier, display priority weight, and time-effect decay parameter of the situation object in the emergency scenario; a visual conflict degree calculation module for visualizing the fusion situation based on the fusion situation data and its corresponding visualization constraint information, and calculating the visual conflict degree of the fusion situation according to the spatial distribution, quantity density, and update characteristics of the situation objects in the fusion situation; and an adaptive visualization adjustment module for dynamically adjusting the visualization constraint information of the situation objects according to the changes in the emergency response stage when the visual conflict degree meets preset conditions, and adaptively adjusting the visualization presentation method of the fusion situation accordingly to reduce the occupation of the visualization interface by non-critical situation objects.

7. A city operation and maintenance status visualization device based on multi-source data fusion according to claim 6, characterized in that: The fusion situation construction module is configured to: perform time alignment processing on data from different data sources that are within the same time window, and map the data to the corresponding spatial region according to a preset spatial mapping rule; Based on time alignment and spatial mapping results, data associated with the same spatial region are aggregated into the same situation object, and the situation object is associated with the corresponding data source type and state characteristics; according to the corresponding data source type, state change characteristics and role in emergency response, the semantic role identifier of the situation object is determined. Based on the semantic role identifier and the data freshness and credibility of the situation object, a corresponding display priority weight and timeliness decay parameter are assigned to the situation object, thereby generating the visualization constraint information of the situation object.

8. The urban operation and maintenance status visualization device based on multi-source data fusion according to claim 7, characterized in that: The visual conflict degree calculation module is configured to: within the current visualization view range, spatially divide the situation objects in the fused situation, and statistically analyze the distribution of situation objects in each spatial region to obtain the spatial clustering characteristics of the situation objects; The number density of situational objects within a unit spatial region is calculated based on the spatial clustering characteristics; and the intensity of visual attention competition caused by the simultaneous appearance of situational objects with different semantic importance within the same spatial region is determined by analyzing the co-occurrence relationship of situational objects with different semantic roles in the same spatial region; the update characteristics of situational objects in the fused situation are analyzed to extract the update frequency, update synchronization degree and update duration of situational objects within a preset time window to determine the dynamic occupancy intensity of dynamic situational objects on the visualization view. Based on the spatial clustering characteristics, and taking the visual attention competition intensity as a sub-indicator representing semantic hierarchical conflict, and the dynamic occupancy intensity as a sub-indicator representing dynamic change in occupancy, a comprehensive model is constructed to evaluate the occupancy of visual presentation resources in the current visualization view of the fusion situation, thereby constructing a visual conflict degree evaluation index for the fusion situation.

9. A non-transitory computer-readable storage medium, characterized in that, The non-transitory computer-readable storage medium stores computer instructions for causing the computer to perform the method described in any one of claims 1-5.

10. A computer program product, characterized in that, The computer program includes a computer program stored on a non-transitory computer-readable storage medium, the computer program including program instructions that, when executed by a computer, cause the computer to perform the method described in any one of claims 1-5.

Images