Multi-source spatial data processing method and system for urban design

Abstract

The invention discloses a multi-source spatial data processing method and system for urban design, and relates to the field of urban design. The method comprises the following steps: collecting multi-source spatial data of a city target area in real time; judging whether abnormal first multi-source spatial data exists in the multi-source spatial data or not; if yes, correcting the first multi-source space data, calculating a first confidence coefficient of the multi-source space correction data obtained by correction, and providing planning decision risk early warning and suggestions for the urban target area according to the multi-source space correction data and the first confidence coefficient; or calculating a second confidence coefficient of the first multi-source spatial data, and providing planning decision risk early warning and suggestion for the urban target area according to the first multi-source spatial data and the second confidence coefficient. According to the method, the abnormal condition of the multi-source spatial data in urban design can be effectively identified and processed, more accurate risk early warning and suggestions are provided for urban planning decision by correcting the abnormal data or evaluating the confidence of the abnormal data, and decision deviation caused by data distortion is avoided.

Description

Technical Field

[0001] This application relates to the field of urban design, and more specifically, to a method and system for processing multi-source spatial data for urban design. Background Technology

[0002] In modern urban planning and management, the processing of urban spatial data is a core element, especially in improving residents' livability. Detailed assessments of walkability and microclimate in specific urban areas are becoming increasingly important. This typically requires integrating spatial data from multiple sources, such as high-resolution satellite remote sensing imagery, UAV lidar point cloud data, and ground-based environmental sensor networks, to create a comprehensive and detailed digital view of the urban environment.

[0003] However, the urban environment is not a static system. In urban design, when multi-source spatial data processing methods need to integrate LiDAR point cloud data and ground environmental sensor data, existing technologies face a significant technical problem: how to construct a fusion mechanism that can adaptively assess and correct inherent biases in the data sources? This bias stems from complex contexts. On the one hand, physical structures introduced by temporary urban activities, such as highly reflective metal sculptures or glass art installations, can lead to false spatial information or abnormal reflection areas in LiDAR data that existing cleaning logic cannot effectively identify. The strong reflection signals or multipath reflection signals generated by these temporary structures may saturate the LiDAR receiver, causing data overflow or artifacts, or recording false point cloud data, which the system's preset cleaning logic struggles to effectively identify and remove.

[0004] On the other hand, concurrent construction activities alter the local microenvironment of ground sensors. For example, placing power generation equipment or erecting large tents near sensors may make the data physically accurate but lose its spatial representativeness of the normal urban environment. The heat and exhaust fumes generated by power generation equipment, or the obstruction of air circulation by large tents, can cause sensor measurements to deviate significantly from the normal level of the area, making them unusable for assessing the normal urban microclimate.

[0005] In this complex situation where multiple factors intertwine, the high weight that engineers assign to specific data sources based on experience in existing data processing workflows may actually amplify these hidden, context-dependent data distortions. Systems may over-rely on temperature, humidity, and wind speed data distorted by temporary facilities, while false point clouds or anomalous reflection areas in lidar data may be incorrectly "corrected" or given lower trust during the fusion process, causing crucial information about the true urban spatial structure to be "ignored" or "distorted." Ultimately, when urban planners make design decisions based on such biased fusion results, they may misjudge the microclimate comfort and pedestrian congestion levels of the area under normal operating conditions, leading to design failures and an inability to effectively improve residents' actual comfort experience.

[0006] To address the aforementioned issues, existing technologies urgently need improvement. Summary of the Invention

[0007] This application discloses a multi-source spatial data processing method for urban design, which aims to solve the problem that existing urban spatial data processing methods are unable to effectively identify and correct false information, abnormal reflection areas, or data that loses spatial representativeness in multi-source spatial data under complex situations, thus leading to deviations in urban planning decisions.

[0008] The technical solution of this application is as follows: In a first aspect, this application discloses a multi-source spatial data processing method for urban design, comprising the following steps: Real-time acquisition of multi-source spatial data of target urban areas; The first multi-source spatial data used to determine whether there are anomalies in the multi-source spatial data; When primary multi-source spatial data exists, it is corrected to obtain corrected multi-source spatial data. The first confidence level of the corrected multi-source spatial data is calculated. Based on the corrected multi-source spatial data and the first confidence level, planning decision-making risk warnings and suggestions are provided for the target urban area; or... Calculate the second confidence level of the first multi-source spatial data, and based on the first multi-source spatial data and the second confidence level, provide early warning and suggestions for planning decision-making risks for the target area of ​​the city.

[0009] Through this technical solution, this application can effectively identify and process anomalies in multi-source spatial data in urban design. By correcting abnormal data or assessing its confidence level, it can provide more accurate risk warnings and suggestions for urban planning decisions, thereby avoiding decision-making biases caused by data distortion and improving the scientific nature of urban planning and the livability experience of residents.

[0010] Furthermore, in the above-mentioned multi-source spatial data processing method for urban design, the multi-source spatial data includes point cloud data and environmental data; Following the steps of real-time acquisition of multi-source spatial data of the target urban area, the following are also included: Obtain the schedule and spatial scope of temporary activities in the target urban area; The specific steps for determining whether multi-source spatial data contains anomalies include: Extract the multispectral reflectance features of the first point cloud data whose reflectance intensity value exceeds the preset intensity value from the point cloud data. Search for the spectral curve that matches the multispectral reflectance features in the preset high reflectance spectral feature library. After finding the match, define the first point cloud data as spurious spatial data. Extract primary environmental data affected by temporary activities from environmental data based on timelines and spatial extent; When the first difference between the first environmental data and the second environmental data of the target urban area under historical normal conditions exceeds the first preset value, and the second difference between the first environmental data and the environmental data of the target urban area that has not been affected by temporary activities exceeds the second preset value, the first environmental data is defined as losing spatial representativeness. When the first point cloud data is false spatial data and / or the first environmental data loses spatial representativeness, it is determined that there is an abnormal first multi-source spatial data in the multi-source spatial data. The first multi-source spatial data includes first point cloud data and / or first environmental data.

[0011] Through this technical solution, this application can propose specific methods for identifying abnormal data for two common types of multi-source spatial data: point cloud data and environmental data. This includes identifying false point cloud data and environmental data that has lost spatial representativeness, thereby more accurately locating abnormal data sources and laying the foundation for subsequent data correction and risk warning.

[0012] Based on this, in the above-mentioned multi-source spatial data processing method for urban design, the step of calculating the first confidence level of the multi-source spatially corrected data specifically includes: The data quality score, contextual interference score, correction effect score, and cross-source consistency score of the first multi-source spatial data were evaluated. Weighting coefficients were assigned to the data quality score, contextual interference score, correction effect score, and cross-source consistency score, respectively. The first confidence level of the multi-source spatially corrected data is calculated based on the data quality score, contextual interference score, correction effect score, cross-source consistency score, and their respective weight coefficients.

[0013] This technical solution introduces multi-dimensional evaluation indicators to calculate the confidence level of corrected data, comprehensively considering data quality, contextual interference, correction effect and cross-source consistency, making the confidence level assessment more comprehensive and objective, effectively avoiding the limitations of single indicator evaluation, and improving the reliability of corrected data in decision-making.

[0014] Furthermore, in the aforementioned multi-source spatial data processing method for urban design, the multi-source spatial data also includes video stream data, audio data, and the reflected light intensity and direction of point cloud data; environmental data includes: ambient temperature data; The specific steps for evaluating the contextual disturbance score of the first multi-source spatial data include: Based on the reflected light intensity and direction, calculate the variance of the reflected light intensity of the point cloud data within a set time period, and the directional variation frequency of the reflected peak value. Calculate the dynamic reflectivity value of the point cloud data based on variance and orientation change frequency; Calculate the image gradient change rate in the video stream data, and calculate the inter-frame motion blur index based on the image gradient change rate; Search the preset heat dissipation device operation voiceprint database for the first voiceprint data that matches the audio data voiceprint; Once the first voiceprint data is found, calculate the temperature gradient in the ambient temperature data and the area of ​​the region where the temperature gradient exceeds a preset threshold. Calculate the heat source intensity index of the ambient temperature data based on the temperature gradient and the area of ​​the region. The contextual interference score of the first multi-source spatial data is evaluated based on the dynamic reflection activity value, motion ambiguity index, and heat source intensity index.

[0015] By incorporating more multi-dimensional information such as the intensity and direction of reflected light from video stream data, audio data, and point cloud data, and combining it with indicators such as dynamic reflection activity value, motion blur index, and heat source intensity index, this application can more precisely assess the degree of contextual interference, thereby more accurately determining the cause and impact of data anomalies and improving the comprehensiveness and accuracy of contextual interference assessment.

[0016] In some preferred embodiments, the step of correcting the first multi-source spatial data to obtain corrected multi-source spatial data in the above-described multi-source spatial data processing method for urban design specifically includes: Collect historical normalized point cloud data of the area corresponding to the first point cloud data and / or normalized point cloud data of the target urban area that has not been affected by temporary activities; Based on historical normalized point cloud data and / or normalized point cloud data, the first point cloud data is reconstructed to obtain point cloud reconstruction correction data, and / or, Collect historical routine environmental data of the area corresponding to the first environmental data and / or routine environmental data of the target urban area that has not been affected by temporary activities; Based on historical and / or normalized environmental data, the first environmental data is corrected to obtain corrected environmental data. Multi-source spatial correction data includes point cloud reconstruction correction data and / or environmental correction data.

[0017] This application proposes a specific method for correcting abnormal point clouds and environmental data based on historical normalized data or normalized data from unaffected areas. Through reconstruction and correction, the authenticity and representativeness of the data can be effectively restored, providing a more reliable data foundation for subsequent planning and decision-making.

[0018] As a technological improvement, in the aforementioned multi-source spatial data processing method for urban design, the step of evaluating the correction effect score of the first multi-source spatial data includes: Calculate the root mean square error of the point cloud deviation between the reconstructed and corrected point cloud data and the preset standard reference data; Calculate the matching degree between the statistical distribution of reflection intensity of the point cloud reconstruction correction data and the preset material reflection intensity distribution; Based on the root mean square error of the point cloud deviation and the matching degree, calculate the first correction effect score of the point cloud reconstruction correction data; Calculate the temperature standard residual and humidity calibration residual between the calculated environmental correction data and the preset standard environmental data; Calculate the second correction effect score for the environmental correction data based on the temperature standard residual and the humidity standard residual; The first and second correction effect scores are weighted and aggregated to obtain the correction effect score of the first multi-source spatial data.

[0019] This technical solution introduces quantitative indicators such as root mean square error of point cloud deviation, reflection intensity matching degree, temperature standard residual and humidity calibration residual, which can objectively evaluate the correction effect of point cloud and environmental data and perform weighted aggregation calculation, thereby providing a scientific basis for the correction effect score and ensuring the quality of the corrected data.

[0020] To enhance functionality, the step of evaluating the data quality score of the first multi-source spatial data in the aforementioned multi-source spatial data processing method for urban design specifically includes: Calculate the sparsity and noise level of the first point cloud data, and calculate the first quality score of the first point cloud data based on the sparsity and noise level; Calculate the stability and missing rate of the first environmental data during data collection, and calculate the second quality score of the first environmental data based on the stability and missing rate; The data quality score of the first multi-source spatial data is evaluated based on the first quality score and the second quality score. The specific steps for evaluating the contextual disturbance score of the first multi-source spatial data include: Based on the schedule and spatial extent of the temporary activities, calculate the degree value of the first point cloud data affected by the temporary activities and the first degree value of the first environmental data affected by the temporary activities. The contextual interference score of the first multi-source spatial data is evaluated based on the degree value and the first degree value.

[0021] Through this technical solution, this application can more comprehensively evaluate the quality of the original data by calculating the sparsity and noise level of point cloud data, as well as the stability and missing rate of environmental data. At the same time, by combining the time schedule and spatial range of temporary activities, it can more accurately quantify the degree of situational interference, thereby providing more refined input for subsequent confidence calculation.

[0022] As an alternative, in the above-mentioned multi-source spatial data processing method for urban design, the multi-source spatial data includes laser pulse reflection signals with different polarization directions; The steps for determining whether multi-source spatial data contains anomalies include: Extract the first laser pulse reflection signal that matches the laser pulse emission signal identifier from the laser pulse reflection signal; Calculate the degree of linear polarization and the polarization angle of the reflected signal of the first laser pulse within a first set time period; Determine whether there is a second laser pulse reflection signal within a second set time period whose linear polarization degree and polarization angle change pattern matches the preset pattern; If so, determine whether the geometric shape of the second laser pulse reflection signal matches that of the historical normal point cloud data; if not, define the second laser pulse reflection signal as false spatial data. If not, determine whether there is a third laser pulse reflection signal whose reflection intensity value exceeds a preset reflection intensity threshold; if yes, define the third laser pulse reflection signal as saturation artifact data. If a second or third laser pulse reflection signal is present in the first laser pulse reflection signal, it is determined that there is an abnormal first multi-source spatial data in the multi-source spatial data. The first type of multi-source spatial data includes: spurious spatial data or saturation artifact data.

[0023] Through this technical solution, this application proposes a specific method for identifying false spatial data and saturation artifact data based on the linear polarization degree, polarization angle variation mode, and reflection intensity threshold for multi-source spatial data containing laser pulse reflection signals with different polarization directions. This method can more effectively handle complex anomalies in lidar data and improve the accuracy and relevance of anomaly data identification.

[0024] Based on the above, in the multi-source spatial data processing method for urban design described above, the step of calculating the second confidence level of the first multi-source spatial data specifically includes: The matching strength between the laser pulse reflection signal and the laser pulse emission signal, the degree of patterning of the linear polarization degree and polarization angle of the second laser pulse reflection signal, the degree of deviation of the geometric shape formed by the second laser pulse reflection signal from the normal ground object, and the degree of closeness of the reflection intensity value of the third laser pulse reflection signal to the preset reflection intensity threshold are calculated. Assign a first weight coefficient to each of the following: matching strength, patterning degree, deviation degree, and proximity degree. The second confidence level of the first multi-source spatial data is calculated based on the matching strength, patterning degree, deviation degree, proximity degree, and their respective first weight coefficients.

[0025] Through this technical solution, this application proposes a method for calculating the confidence level of abnormal data of laser pulse reflection signals based on multi-dimensional indicators such as matching strength, patterning degree, deviation degree and proximity degree, and assigns weight coefficients for comprehensive evaluation, so as to make the confidence level assessment of abnormal data more refined and accurate, and provide a more reliable basis for subsequent decision-making.

[0026] Secondly, this application also discloses a multi-source spatial data processing system for urban design, comprising: The acquisition module is used to collect multi-source spatial data of the target area in the city in real time; The judgment module is used to determine whether there is any abnormal first multi-source spatial data in the multi-source spatial data; The processing module is used to correct the first multi-source spatial data when it exists, to obtain multi-source spatially corrected data, calculate the first confidence level of the multi-source spatially corrected data, and provide planning decision risk warnings and suggestions for the urban target area based on the multi-source spatially corrected data and the first confidence level; or, it is used to calculate the second confidence level of the first multi-source spatial data, and provide planning decision risk warnings and suggestions for the urban target area based on the first multi-source spatial data and the second confidence level.

[0027] This application provides a multi-source spatial data processing system for urban design. Through modular design, it realizes the collection, anomaly detection, correction, and confidence calculation of multi-source spatial data, and ultimately provides early warning and suggestions for planning decision risks. It provides a systematic solution for urban planning and improves processing efficiency and the level of intelligence in decision-making.

[0028] Beneficial effects This application discloses a multi-source spatial data processing method for urban design. It collects multi-source spatial data of a target urban area in real time and determines whether abnormal data exists. When abnormal data is found, this application can correct the abnormal data and calculate the corrected confidence level, or directly calculate the confidence level of the abnormal data. Based on the corrected data or abnormal data and their confidence levels, it provides early warning and suggestions for planning decision risks in the target urban area. This method effectively solves the problem in existing technologies where multi-source spatial data processing methods in urban design struggle to adaptively assess and correct inherent biases in data sources. Specifically, this application identifies false spatial information or abnormal reflection areas introduced by temporary urban activities, as well as environmental data that loses spatial representativeness due to concurrent construction activities. This avoids the problem in traditional methods where engineers assign high weights to data sources based on experience, potentially amplifying data distortion. By correcting abnormal data or assessing its confidence level, this application can provide more accurate assessments of urban microclimate comfort and pedestrian congestion, thereby preventing planning scheme failures due to data bias, effectively improving residents' actual comfort experience, and providing a more scientific and reliable basis for urban planning decisions. Attached Figure Description

[0029] Figure 1 This application provides a schematic diagram of a multi-source spatial data processing method for urban design.

[0030] Figure 2 This application provides a schematic diagram of a multi-source spatial data processing system for urban design.

[0031] Figure 2 In the diagram: 1 is the data acquisition module, 2 is the judgment module, and 3 is the processing module. Detailed Implementation

[0032] The technical solutions of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments.

[0033] See Figure 1 This application proposes a multi-source spatial data processing method for urban design, comprising the following steps: Step S10: Real-time acquisition of multi-source spatial data of the target urban area; Step S20: Determine whether there is any abnormal first multi-source spatial data in the multi-source spatial data; Step S30: When first multi-source spatial data exists, the first multi-source spatial data is corrected to obtain multi-source spatially corrected data. The first confidence level of the multi-source spatially corrected data is calculated. Based on the multi-source spatially corrected data and the first confidence level, planning decision risk warnings and suggestions are provided for the urban target area. Alternatively, the second confidence level of the first multi-source spatial data is calculated. Based on the first multi-source spatial data and the second confidence level, planning decision risk warnings and suggestions are provided for the urban target area.

[0034] This application effectively solves the bias and distortion problems of traditional methods when dealing with multi-source spatial data in complex urban contexts by introducing anomaly data judgment, correction and confidence assessment mechanisms, thus providing a more accurate and reliable basis for urban planning decisions.

[0035] To better understand the technical solution proposed in this application, some key terms involved will be explained first.

[0036] "Multi-source spatial data" refers to various types of data describing the urban spatial environment obtained from different sources, such as point cloud data, environmental data, video stream data, and audio data. These data collectively constitute a digital profile of the urban environment, which is the foundation of urban planning and management.

[0037] "Target urban area" refers to the specific geographical area for urban design or planning, which can be a neighborhood, park, commercial center, etc.

[0038] "First-source spatial data" refers to data identified as anomalous or unreliable in multi-source spatial data. These anomalous data may be caused by various factors such as sensor malfunction, temporary activity interference, and environmental changes.

[0039] "Multi-source spatial correction data" refers to data obtained by correcting the first multi-source spatial data, aiming to restore its authenticity and representativeness.

[0040] "First confidence level" and "second confidence level" are metrics used to assess data reliability. The first confidence level is used to assess the reliability of corrected multi-source spatially corrected data, while the second confidence level is used to assess the reliability of uncorrected outlier data (i.e., the first multi-source spatial data). These confidence levels reflect the extent to which the data can be trusted, thus guiding subsequent planning decisions.

[0041] "Planning Decision Risk Warning and Suggestions" refers to providing urban planners with alerts and corresponding optimization suggestions regarding potential risks based on processed spatial data and its confidence level, in order to avoid decision-making errors caused by data bias.

[0042] The multi-source spatial data processing method for urban design proposed in this application is based on the intelligent processing of multi-source spatial data to address the complexity and dynamism of the urban environment.

[0043] Specifically, in the step of "real-time acquisition of multi-source spatial data of urban target areas," data can be acquired in various ways. For example, environmental data such as ambient temperature, humidity, and air quality can be continuously collected through a network of fixed sensors deployed throughout the city. Simultaneously, drones equipped with LiDAR can be used to scan urban target areas periodically or as needed to acquire high-precision point cloud data for constructing 3D city models. Furthermore, video stream data can be acquired through urban surveillance cameras, and audio data can be acquired through microphone arrays to capture dynamic information and the acoustic environment of the city. These data sources can be collected independently or collaboratively through an integrated platform to ensure the real-time nature and comprehensiveness of the data.

[0044] In the step of "determining whether there are anomalies in the first multi-source spatial data," a preliminary quality check and anomaly identification are required for the collected multi-source spatial data. For example, a threshold can be set; if a sensor continuously outputs temperature values ​​exceeding the normal range within a short period, the environmental data is considered anomaly. For point cloud data, regions significantly inconsistent with the surrounding environment can be identified by analyzing features such as point cloud density and reflectance distribution, such as suddenly appearing extremely high reflectance points or sparse areas, which may indicate the presence of anomalous data. Furthermore, comparison with historical data can identify data that deviates significantly from the normal pattern, thus determining it as anomalous data.

[0045] When anomalies are detected in the first multi-source spatial data, the system enters two processing paths. The first path involves correcting the first multi-source spatial data to obtain corrected multi-source spatial data, calculating the first confidence level of the corrected data, and providing planning decision risk warnings and suggestions for the target urban area based on the corrected data and the first confidence level. For example, if temperature sensor data for a certain area is identified as anomaly, an interpolation algorithm can be used, combined with readings from other normal sensors in the vicinity and historical normalized data, to correct the anomaly and obtain more reasonable environmental correction data. For anomalous points in point cloud data, reconstruction or smoothing can be performed using the geometric features of surrounding normal point clouds to generate point cloud reconstruction correction data. After correction, the quality of the corrected data needs to be assessed. For example, the first confidence level is calculated by measuring the difference between the corrected data and the original data, and the consistency between the corrected data and historical normalized data. This confidence level reflects the reliability of the corrected data. Finally, by combining the corrected multi-source spatial data and its first confidence level, the system can provide urban planners with more accurate planning decision risk warnings and suggestions. For example, if the revised data still shows microclimate anomalies in a certain area, but with a high degree of confidence, it can be suggested that planners increase greenery or adjust building layouts in that area.

[0046] The second approach is to "calculate the second confidence level of the first multi-source spatial data, and based on the first multi-source spatial data and the second confidence level, provide early warnings and suggestions for planning decision risks in the target urban area." This approach is suitable for data that, although judged as anomalous, is difficult to correct or still has significant uncertainty after correction. In this case, the system will not attempt to correct the data, but will directly assess its second confidence level. For example, if there are a large number of unexplained strong reflection points in the lidar data of a certain area, and cannot be effectively corrected by existing methods, the system will assess the reliability of this anomalous data. The assessment can be based on factors such as the source of the anomalous data, the degree of anomalousness, and the degree of conflict with surrounding data. For example, if the anomalous data is caused by a known interference source, its confidence level may be high. Ultimately, the system will combine this uncorrected anomalous data with its second confidence level to provide early warnings and suggestions to planners. For example, if the confidence level of the anomalous point cloud data in a certain area is high, the system may suggest that planners take a cautious approach to the data in that area during the design process, or conduct on-site investigations to verify the data.

[0047] The multi-source spatial data processing method for urban design proposed in this application forms a complete data processing closed loop through the synergistic effect of the aforementioned steps. First, real-time acquisition of multi-source spatial data ensures the timeliness and comprehensiveness of the data. Second, an intelligent judgment mechanism can promptly detect and identify anomalies in the data, avoiding the negative impact of anomalous data on subsequent decision-making. More importantly, this application provides two flexible strategies for handling anomalous data: correcting and evaluating the confidence level of the corrected data, or directly evaluating the confidence level of the anomalous data. These two strategies enable the system to adopt the most appropriate processing method based on the specific circumstances of the anomalous data, thereby maximizing the preservation of the data's effective information and quantitatively assessing its reliability. Finally, combining the processed data and confidence levels, it provides accurate risk warnings and suggestions for urban planning decisions, effectively improving the scientific rigor and rationality of urban design.

[0048] The core innovation of this application lies in introducing a step of "determining whether there are anomalies in the first multi-source spatial data," and providing two flexible anomaly processing paths: "correcting and calculating the first confidence level" and "directly calculating the second confidence level." This mechanism enables the system to adaptively assess and correct inherent biases in the data source, effectively identifying and processing false spatial information and data lacking spatial representativeness caused by temporary urban activities or construction activities. By assigning confidence levels to corrected or uncorrected anomaly data, this application can quantify the reliability of the data, thereby avoiding decision-making errors caused by over-reliance on empirical weights in traditional methods. For example, when false point clouds appear in lidar data caused by highly reflective metal sculptures, this application can identify and correct these anomalies, or, if correction is not possible, provide a higher confidence level for the anomaly data to remind planners to use it with caution. When ground sensor data loses spatial representativeness due to construction activities, this application can also identify such anomalies and provide corresponding confidence levels, preventing planners from making decisions based on distorted data.

[0049] Therefore, the method proposed in this application can provide urban planners with more accurate and reliable urban environmental assessment results, effectively improving the scientific nature of urban design and the livability experience of residents.

[0050] This application further proposes a more refined abnormal data identification mechanism, which improves the accuracy and robustness of abnormal data detection by introducing specific data types and multi-dimensional judgment criteria.

[0051] Specifically, in the above-mentioned multi-source spatial data processing method for urban design, the multi-source spatial data includes point cloud data and environmental data; Following the steps of real-time acquisition of multi-source spatial data of the target urban area, the following are also included: Obtain the schedule and spatial scope of temporary activities in the target urban area; The specific steps for determining whether multi-source spatial data contains anomalies include: Extract the multispectral reflectance features of the first point cloud data whose reflectance intensity value exceeds the preset intensity value from the point cloud data. Search for the spectral curve that matches the multispectral reflectance features in the preset high reflectance spectral feature library. After finding the match, define the first point cloud data as spurious spatial data. Extract primary environmental data affected by temporary activities from environmental data based on timelines and spatial extent; When the first difference between the first environmental data and the second environmental data of the target urban area under historical normal conditions exceeds the first preset value, and the second difference between the first environmental data and the environmental data of the target urban area that has not been affected by temporary activities exceeds the second preset value, the first environmental data is defined as losing spatial representativeness. When the first point cloud data is false spatial data and / or the first environmental data loses spatial representativeness, it is determined that there is an abnormal first multi-source spatial data in the multi-source spatial data. The first multi-source spatial data includes first point cloud data and / or first environmental data.

[0052] Specifically, multi-source spatial data can be understood as a collection of various types of data used to describe the spatial state of a city. Point cloud data refers to a set of three-dimensional spatial points obtained through devices such as LiDAR. Each point usually contains information such as three-dimensional coordinates and reflection intensity, which is used to construct a three-dimensional geometric model of the city. Environmental data refers to data that reflects the physical environment of the city, such as ambient temperature, humidity, air quality, and noise.

[0053] Furthermore, obtaining the timetable and spatial extent of temporary activities in target urban areas refers to collecting and identifying information on non-routine events or activities occurring within a specific time period and spatial area through various channels (such as government announcements, news reports, social media, sensor networks, etc.). Examples include construction sites, large-scale outdoor events, and traffic control measures. This information is used to help determine the anomaly nature of spatial data.

[0054] The step of determining whether anomalies exist in the first multi-source spatial data is detailed. Specifically, for point cloud data, multispectral reflectance features are extracted from the first point cloud data whose reflectance intensity value exceeds a preset value. A reflectance intensity value exceeding the preset value usually indicates that the point cloud may originate from an atypical reflectance source, such as a temporary structure or anomalous object with high reflectivity. Multispectral reflectance features refer to the distribution of reflectance intensity across different wavelength ranges, which can be used to identify material types. By searching for spectral curves in a preset high-reflectance spectral feature library that match the extracted multispectral reflectance features, it can be determined whether the first point cloud data is consistent with the spectral characteristics of known high-reflectivity spurious objects (such as reflectors, temporary metal structures, etc.). If a match is successful, the first point cloud data is defined as spurious spatial data, the purpose of which is to identify and eliminate point cloud data anomalies caused by non-real ground objects or temporary interference.

[0055] Simultaneously, for environmental data, based on the acquired schedules and spatial extent of temporary activities, primary environmental data potentially affected by these activities is extracted. For example, if construction is underway in an area, environmental data such as noise and ambient temperature may be affected. To determine whether this primary environmental data has lost its spatial representativeness, it is compared with two benchmarks: first, with second environmental data reflecting the historical normality of the target urban area, calculating the first difference; and second, with the second difference from environmental data of the target urban area unaffected by temporary activities. When both differences exceed their respective preset thresholds, it indicates that the primary environmental data has significantly deviated from its historical normality and the normality of the surrounding unaffected areas, thus defining it as having lost its spatial representativeness. The purpose is to identify and eliminate environmental data distortion caused by temporary activities, ensuring the validity of environmental data.

[0056] Therefore, when the first point cloud data is defined as spurious spatial data, and / or the first environmental data is defined as lacking spatial representativeness, it can be determined that there is an abnormal first multi-source spatial data in the multi-source spatial data. This first multi-source spatial data may include spurious spatial data and / or environmental data lacking spatial representativeness.

[0057] This application's solution effectively addresses the issue of insufficient accuracy in identifying anomalous data in complex urban environments by introducing specific anomaly detection mechanisms for point cloud data and environmental data from multi-source spatial data. Specifically, by acquiring the timeline and spatial extent of temporary activities in the target urban area, crucial contextual information is provided for subsequent anomaly detection. For point cloud data, by analyzing its multispectral reflectance characteristics and comparing them with a pre-defined high reflectance spectral feature library, false spatial data caused by temporary reflective objects or spurious structures can be accurately identified, avoiding potential misjudgments based solely on reflectance intensity thresholds. For environmental data, by combining temporary activity information with historical normalized data and data from unaffected areas, it is possible to effectively determine whether environmental data has lost its spatial representativeness due to temporary activities, thus avoiding misjudging temporary, localized environmental fluctuations as normalized anomalies. This multi-dimensional, context-aware anomaly detection method makes the identification of anomalous data more refined and accurate, laying a solid foundation for subsequent data correction and risk warning.

[0058] Through the aforementioned technical solution, this application significantly improves the accuracy and robustness of anomaly detection in multi-source spatial data. Specifically, by distinguishing between spurious point cloud data caused by temporary activities and environmental data that has lost spatial representativeness, it avoids misjudging normal data affected by temporary events as anomalies, or missing truly anomalous data. This refined anomaly identification capability enables urban planning decisions to be based on more reliable and authentic data, thereby effectively reducing planning risks caused by inaccurate data. Furthermore, by introducing the timetable and spatial scope of temporary activities as judgment criteria, anomaly detection becomes more context-aware, better adapting to the dynamic changes in the urban environment and providing more accurate data support for sustainable urban development.

[0059] Specifically, in the above-mentioned multi-source spatial data processing method for urban design, the step of calculating the first confidence level of the multi-source spatially corrected data includes: evaluating the data quality score, contextual interference score, correction effect score, and cross-source consistency score of the first multi-source spatial data; assigning weight coefficients to the data quality score, contextual interference score, correction effect score, and cross-source consistency score respectively; and calculating the first confidence level of the multi-source spatially corrected data based on the data quality score, contextual interference score, correction effect score, cross-source consistency score, and their respective weight coefficients.

[0060] The data quality score quantifies the reliability and completeness of the primary multi-source spatial data, taking into account factors such as sparsity, noise level, missing data rate, and acquisition stability. The contextual interference score measures the impact of external environmental factors on the primary multi-source spatial data, assessing interference from temporary events, abnormal weather, or equipment malfunctions. The correction effect score evaluates the improvement achieved by correcting the primary multi-source spatial data, assessing differences between the corrected and uncorrected data, deviations from standard reference data, and internal data consistency. The cross-source consistency score measures the consistency and synergy between different types or sources of multi-source spatial correction data, ensuring that the corrected data corroborates each other across different dimensions and avoiding bias from a single data source.

[0061] Specifically, the allocation of weighting coefficients can be flexibly adjusted based on the actual application scenario, the importance of the data type, and the degree of influence on the confidence score calculation. For example, in scenarios with extremely high data accuracy requirements, the data quality score and the correction effect score may be given higher weights; while in scenarios with complex and ever-changing environments, the contextual interference score may be given higher weights. These weighting coefficients can be preset or dynamically optimized based on historical data through machine learning models.

[0062] This application's solution comprehensively evaluates data quality scores, contextual interference scores, correction effect scores, and cross-source consistency scores, assigning corresponding weight coefficients to these evaluation indicators. This allows for a comprehensive and objective quantification of the first confidence level of multi-source spatial correction data. This multi-dimensional, weighted aggregation evaluation mechanism ensures that the calculation of the first confidence level is no longer limited to a single dimension, but fully reflects the reliability of the data itself, the impact of the external environment, the effectiveness of the correction process, and the synergy between different data sources. Therefore, it allows for a more accurate assessment of the reliability of multi-source spatial correction data, providing a more solid data foundation for subsequent planning decision-making risk warnings and recommendations.

[0063] Through the above technical solution, the first confidence level of multi-source spatially corrected data can be calculated more accurately and comprehensively. Compared with confidence assessment methods that rely solely on a single indicator or simple averaging, this solution significantly improves the accuracy and robustness of confidence assessment by introducing data quality scores, contextual interference scores, correction effect scores, and cross-source consistency scores, and performing weighted aggregation. This helps to more effectively identify and quantify potential risks in urban planning decisions, thereby providing more reliable and targeted early warnings and suggestions for planning decision risks in target urban areas, avoiding decision-making errors caused by data uncertainty, and ultimately improving the scientific nature and effectiveness of urban design.

[0064] This application further includes the aforementioned multi-source spatial data, including video stream data, audio data, and the reflected light intensity and direction of point cloud data; environmental data includes: ambient temperature data; The specific steps for evaluating the contextual disturbance score of the first multi-source spatial data include: Based on the reflected light intensity and direction, calculate the variance of the reflected light intensity of the point cloud data within a set time period, and the directional variation frequency of the reflected peak value. Calculate the dynamic reflectivity value of the point cloud data based on variance and orientation change frequency; Calculate the image gradient change rate in the video stream data, and calculate the inter-frame motion blur index based on the image gradient change rate; Search the preset heat dissipation device operation voiceprint database for the first voiceprint data that matches the audio data voiceprint; Once the first voiceprint data is found, calculate the temperature gradient in the ambient temperature data and the area of ​​the region where the temperature gradient exceeds a preset threshold. Calculate the heat source intensity index of the ambient temperature data based on the temperature gradient and the area of ​​the region. The contextual interference score of the first multi-source spatial data is evaluated based on the dynamic reflection activity value, motion ambiguity index, and heat source intensity index.

[0065] Specifically, multi-source spatial data, in addition to point cloud data and environmental data, can further include video stream data, audio data, and the intensity and direction of reflected light from point cloud data. The intensity and direction of reflected light refer to the intensity and direction of the light signal reflected back to the sensor from the laser pulse during the point cloud data acquisition process. This information reflects the physical characteristics of the object's surface and the lighting conditions in the environment. Environmental data can specifically include ambient temperature data, used to monitor temperature changes in the target area.

[0066] When evaluating the contextual interference score of the first multi-source spatial data, firstly, the variance of the reflected light intensity and the frequency of directional changes in the reflected light peaks of the point cloud data within a set time period can be calculated based on the reflected light intensity and direction of the point cloud data. The variance of the reflected light intensity characterizes the light intensity fluctuation of the point cloud data over time, while the frequency of directional changes in the reflected light peaks reflects the stability or degree of change in the reflected light direction. Based on these calculation results, the dynamic reflection activity value of the point cloud data can be further calculated. This value is used to quantify the degree of dynamic change of the point cloud data within a specific time period, such as changes caused by moving objects or changes in ambient lighting.

[0067] Secondly, the image gradient change rate in the video stream data can be calculated, and the inter-frame motion blur index can be calculated based on this rate. The image gradient change rate reflects the speed and direction of pixel value changes between video frames, while the inter-frame motion blur index can quantify the degree of blurring of moving objects in the video, thus indirectly reflecting the intensity of dynamic activities in the scene.

[0068] Furthermore, audio data can be used for contextual interference assessment. Specifically, the first voiceprint data matching the audio data can be searched in a pre-defined database of operating heat dissipation equipment voiceprints. This database stores the voiceprint characteristics of various heat dissipation devices (such as air conditioner outdoor units, cooling towers, etc.) during normal operation. If a matching first voiceprint data is found, it indicates that there may be operating heat dissipation equipment in the area. Based on this, the temperature gradient in the ambient temperature data and the area of ​​the region where the temperature gradient exceeds a preset threshold can be calculated. The temperature gradient reflects the rate of temperature change in space, while the area of ​​the region exceeding the threshold indicates the extent of the heat source. Based on the temperature gradient and the area, the heat source intensity index of the ambient temperature data can be calculated, which is used to quantify the degree of ambient temperature interference caused by heat dissipation equipment or other heat sources.

[0069] Finally, the calculated dynamic reflection activity value, motion ambiguity index, and heat source intensity index are comprehensively considered to evaluate the contextual interference score of the first multi-source spatial data. These indices capture contextual factors such as dynamic changes, motion activities, and heat source interference within the urban target area from different dimensions, providing a multi-dimensional basis for quantifying the degree of contextual interference.

[0070] This application's solution incorporates richer and more detailed multi-source spatial data, including video stream data, audio data, reflected light intensity and direction from point cloud data, and ambient temperature data. Quantitative indicators such as dynamic reflectivity, motion blur index, and heat source intensity index are designed for this data, enabling a more comprehensive and accurate capture of contextual disturbances within urban target areas. For example, the dynamic reflectivity value effectively identifies changes in point cloud data caused by moving targets such as vehicles and pedestrians; the motion blur index reflects the intensity of motion in the scene from a video perspective; and the heat source intensity index identifies and quantifies localized thermal disturbances generated by industrial equipment, building facilities, etc., through temperature and voiceprint data. It is precisely because of these multi-dimensional, high-precision contextual feature quantifications that the assessment of the degree of contextual disturbance in the first multi-source spatial data becomes more objective and accurate, avoiding misjudgments caused by insufficient information or coarse assessments.

[0071] Through the aforementioned technical solution, this application significantly enhances the ability to identify and quantify contextual interference within urban target areas. By comprehensively analyzing the dynamic reflectivity of point cloud data, the motion blur index of video stream data, and the heat source intensity index of ambient temperature data, it can more accurately determine whether multi-source spatial data is affected by temporary activities or abnormal events, and the degree of interference. This multi-dimensional and refined assessment method allows the contextual interference score to more realistically reflect the actual situation, thus providing a more reliable and accurate basis for subsequent correction of the first multi-source spatial data and for risk warnings and suggestions for planning decisions. This effectively avoids decision-making biases caused by inaccurate contextual interference assessments, and improves the scientific rigor and effectiveness of urban design and planning.

[0072] This application further proposes steps for correcting the first multi-source spatial data to obtain multi-source spatially corrected data, specifically including: Collect historical normalized point cloud data of the area corresponding to the first point cloud data and / or normalized point cloud data of the target urban area that has not been affected by temporary activities; Based on historical normalized point cloud data and / or normalized point cloud data, the first point cloud data is reconstructed to obtain point cloud reconstruction correction data, and / or, Collect historical routine environmental data of the area corresponding to the first environmental data and / or routine environmental data of the target urban area that has not been affected by temporary activities; Based on historical and / or normalized environmental data, the first environmental data is corrected to obtain corrected environmental data. Multi-source spatial correction data includes point cloud reconstruction correction data and / or environmental correction data.

[0073] Specifically, the first point cloud data refers to point cloud data that is judged to be abnormal during real-time acquisition, such as spurious spatial data. Historical normalized point cloud data refers to point cloud data collected within the target urban area during normal periods without temporary activity interference, reflecting the typical spatial morphology and characteristics of the area. Normalized point cloud data within the target urban area unaffected by temporary activities refers to point cloud data collected in other areas within the target urban area that have not been affected by temporary activities during the current time period, serving as a normal reference for the current time period. Reconstructing the first point cloud data can be understood as using this normalized data as a reference to repair, supplement, or replace abnormal point cloud data. This can be achieved through interpolation, model fitting, or directly replacing point cloud data in abnormal areas to eliminate the impact of abnormal point cloud data, thereby obtaining point cloud reconstruction and correction data that more closely approximates the actual situation.

[0074] The first environmental data refers to environmental data that is judged to be abnormal during real-time acquisition, such as environmental data that has lost its spatial representativeness. Historical normalized environmental data refers to environmental data collected within the target urban area during normal periods without temporary activity interference; it reflects the typical environmental parameters of the area. Normalized environmental data within the target urban area unaffected by temporary activities refers to environmental data collected in other areas within the target urban area that have not been affected by temporary activities during the current time period; it can serve as a normal reference for the current time period. Correcting the first environmental data can be understood as using these normalized data as a reference to calibrate, smooth, or replace abnormal environmental data. This can be done through time series analysis, spatial interpolation, or regression models based on normalized data to eliminate the impact of abnormal environmental data, thereby obtaining more representative corrected environmental data.

[0075] In practical applications, multi-source spatial correction data can include point cloud reconstruction correction data and / or environmental correction data, depending on whether the original anomalous data is point cloud data, environmental data, or both.

[0076] The proposed solution introduces historical normalized data and / or normalized data unaffected by temporary activities as a correction benchmark, providing a reliable reference for anomalous first-source spatial data. When first-source point cloud data is identified as anomalous, reconstructing it using historical normalized point cloud data or normalized point cloud data unaffected by temporary activities effectively removes false or disturbed point cloud information, restoring its true spatial geometry and reflection characteristics. Similarly, when first-source environmental data loses its spatial representativeness, correcting it by referencing historical normalized environmental data or normalized environmental data unaffected by temporary activities eliminates biases caused by temporary activities, allowing it to reflect the actual environmental conditions of the target urban area again. This correction mechanism based on a reliable benchmark ensures that the corrected multi-source spatial data more accurately reflects the normalized characteristics of the target urban area, thus providing high-quality input for subsequent first-confidence calculations.

[0077] Through the above technical solution, this application can significantly improve the accuracy and reliability of multi-source spatial data correction. By introducing multi-dimensional normalized data as the correction benchmark, the interference and deviation caused by temporary activities or abnormal situations to the original data can be effectively eliminated, making the corrected point cloud reconstruction data and environmental correction data closer to the real and normalized state of the urban target area. As a result, the first confidence level calculated subsequently will more accurately reflect the credibility of the data, thereby providing more accurate and reliable risk warnings and suggestions for urban planning decisions, and avoiding misjudgments or suboptimal decisions caused by data bias.

[0078] This application further proposes steps for evaluating the correction effect score of the first multi-source spatial data, which specifically include: Calculate the root mean square error of the point cloud deviation between the reconstructed and corrected point cloud data and the preset standard reference data; Calculate the matching degree between the statistical distribution of reflection intensity of the point cloud reconstruction correction data and the preset material reflection intensity distribution; Based on the root mean square error of the point cloud deviation and the matching degree, calculate the first correction effect score of the point cloud reconstruction correction data; Calculate the temperature standard residual and humidity calibration residual between the calculated environmental correction data and the preset standard environmental data; Calculate the second correction effect score for the environmental correction data based on the temperature standard residual and the humidity standard residual; The first and second correction effect scores are weighted and aggregated to obtain the correction effect score of the first multi-source spatial data.

[0079] Specifically, the root mean square error (RMSE) of point cloud deviation is a quantitative indicator of the geometric deviation in three-dimensional space between the reconstructed and corrected point cloud data and the preset standard reference data. Its calculation typically involves averaging the squared Euclidean distances between corresponding points and then taking the square root. Its purpose is to evaluate the geometric accuracy of the reconstructed and corrected point cloud data. The preset standard reference data can be understood as point cloud data representing a real-world scene that has undergone high-precision measurement or verification, such as data collected under normal conditions using a high-precision laser scanner. The statistical distribution of reflectance intensity refers to the frequency distribution of reflectance intensity at each point in the reconstructed and corrected point cloud data. The preset material reflectance intensity distribution can be understood as the characteristic distribution of reflectance intensity of specific urban features (such as building materials, vegetation, and road surfaces) under standard conditions. The matching degree aims to quantify the authenticity of the reconstructed and corrected point cloud data in terms of material properties. For example, it can be measured by calculating the correlation coefficient, KL divergence, or JS divergence between the two distributions. In practical applications, the first correction effect score is derived by combining the RMSE of point cloud deviation and the matching degree of the statistical distribution of reflectance intensity, and is used to comprehensively evaluate the correction effect of the reconstructed and corrected point cloud data in terms of geometric and material properties. For example, a comprehensive evaluation can be made by weighted averaging of the two or based on fuzzy logic.

[0080] Specifically, the temperature standard residual refers to the difference between the temperature value in the environmental correction data and the temperature value in the preset standard environmental data, obtained after standardization. Similarly, the humidity calibration residual refers to the difference between the humidity value in the environmental correction data and the humidity value in the preset standard environmental data, obtained after calibration. The preset standard environmental data can be understood as data representing the real environmental conditions collected by high-precision environmental sensors under normal conditions. These residuals aim to quantify the correction accuracy of the environmental correction data in terms of temperature and humidity. The second correction effect score is derived by combining the temperature standard residual and the humidity calibration residual, and is used to evaluate the correction effect of the environmental correction data on environmental parameters. For example, it can be weighted and summed based on the absolute value or square value of the residuals. In practical applications, the first and second correction effect scores are weighted and aggregated to assign different weights to different data types (point cloud data and environmental data) based on their importance or correction difficulty in urban design, thereby obtaining a comprehensive correction effect score that fully reflects the overall correction quality of the first multi-source spatial data.

[0081] This application addresses the problem of insufficient quantification of correction effects in the basic scheme by introducing multi-dimensional evaluation indicators for correction effects. Specifically, for point cloud data, by calculating the root mean square error of point cloud deviation, the degree of geometric conformity between the reconstructed point cloud data and the real scene can be accurately measured; simultaneously, by calculating the matching degree between the statistical distribution of reflection intensity and the preset material reflection intensity distribution, the authenticity of the reconstructed point cloud data in terms of material properties can be evaluated. The combination of these two indicators allows the first correction effect score of the point cloud reconstruction correction data to comprehensively and objectively reflect its correction quality. For environmental data, by calculating the temperature standard residual and humidity calibration residual, the correction accuracy of environmental correction data on key environmental parameters can be quantified. Finally, by weighted aggregation of the first and second correction effect scores, this application can obtain a comprehensive correction effect score. This score not only considers the correction characteristics of different data types but also provides more accurate and reliable input for subsequent first confidence level calculations, thereby ensuring the scientific nature and effectiveness of planning decision risk warnings and recommendations.

[0082] Through the above technical solution, this application provides a more refined and comprehensive mechanism for evaluating the correction effect. Compared with basic solutions that only correct data without quantitative evaluation, this application introduces specific indicators such as root mean square error of point cloud deviation, reflection intensity matching degree, temperature standard residual, and humidity calibration residual, and performs weighted aggregation calculations, significantly improving the accuracy and reliability of the evaluation of the correction quality of the first multi-source spatial data. Therefore, the calculated first confidence level will more realistically reflect the usability of the corrected data, making urban planning decision-making risk warnings and recommendations more instructive, effectively avoiding decision-making biases caused by inaccurate evaluation of data correction effects, and improving the scientific nature and robustness of urban design schemes.

[0083] This application further proposes specific methods for evaluating the data quality score and contextual interference score of first multi-source spatial data.

[0084] Specifically, the steps for evaluating the data quality score of the first multi-source spatial data include: Calculate the sparsity and noise level of the first point cloud data, and calculate the first quality score of the first point cloud data based on the sparsity and noise level; Calculate the stability and missing rate of the first environmental data during data collection, and calculate the second quality score of the first environmental data based on the stability and missing rate; The data quality score of the first multi-source spatial data is evaluated based on the first quality score and the second quality score. The specific steps for evaluating the contextual disturbance score of the first multi-source spatial data include: Based on the schedule and spatial extent of the temporary activities, calculate the degree value of the first point cloud data affected by the temporary activities and the first degree value of the first environmental data affected by the temporary activities. The contextual interference score of the first multi-source spatial data is evaluated based on the degree value and the first degree value.

[0085] The first point cloud data refers to the first point cloud data extracted from the point cloud data whose reflection intensity value exceeds a preset value when judging whether there are anomalies in the multi-source spatial data. Sparsity can be understood as the distribution density of points in the point cloud data, for example, it can be measured by calculating the number of points per unit volume or unit area. Noise level refers to the random errors or inaccuracies present in the point cloud data, for example, it can be determined by statistically analyzing the proportion or magnitude of points in the point cloud data that deviate from the local average. The first quality score is a quantitative assessment of the quality of the first point cloud data by combining sparsity and noise level, and its purpose is to reflect the reliability of the point cloud data.

[0086] The first environmental data refers to the first environmental data extracted from the environmental data that has been affected by temporary activities when determining whether there are anomalies in the multi-source spatial data. Stability refers to the degree of fluctuation in environmental data over time during the acquisition process; for example, it can be measured by calculating the standard deviation of environmental data values ​​over a period of time. The missing rate refers to the proportion of data points in the environmental data that were not successfully acquired or recorded, and its purpose is to reflect the completeness of the environmental data. The second quality score is a quantitative assessment of the quality of the first environmental data by combining stability and missing rate, and its purpose is to reflect the reliability of the environmental data. The data quality score is a comprehensive evaluation of the first quality score of the first point cloud data and the second quality score of the first environmental data to obtain an overall data quality assessment of the first multi-source spatial data.

[0087] The contextual disturbance score aims to quantify the impact of temporary events on multi-source spatial data. The timeline and spatial extent of a temporary event refer to the temporal and geographical boundaries of short-term events occurring within the target urban area that may affect spatial data acquisition, such as large gatherings, construction activities, or traffic control. The severity value indicates the degree to which the first point cloud data is affected by the temporary event within its timeline and spatial extent; for example, it can be determined by calculating the proportion or intensity of the area affected by the temporary event in the point cloud data. The primary severity value indicates the degree to which the first environmental data is affected by the temporary event within its timeline and spatial extent; for example, it can be determined by analyzing the anomalous fluctuations in environmental data during the temporary event. The contextual disturbance score assesses the overall degree of disturbance to the first multi-source spatial data caused by temporary events based on the severity value and the primary severity value.

[0088] The proposed solution quantifies the sparsity and noise level of the first point cloud data, as well as the stability and missing rate of the first environmental data during acquisition, enabling a detailed assessment of the intrinsic quality of the first multi-source spatial data. Furthermore, by combining the timeline and spatial extent of temporary activities, the degree to which the first point cloud data and the first environmental data are affected by these activities is calculated, accurately identifying and quantifying the interference caused by external circumstances. Therefore, the aforementioned assessment method provides more refined and comprehensive input parameters for subsequent calculations of the first confidence level of the multi-source spatially corrected data.

[0089] The above technical solutions enable a more accurate and objective assessment of the data quality and contextual interference level of the first multi-source spatial data. This meticulous assessment helps to more precisely reflect the reliability of the original anomalous data and the degree of influence from external factors when calculating the first confidence level, thereby making the final planning decision risk warnings and suggestions more targeted and effective, and avoiding misjudgments or omissions caused by inaccurate assessments of data quality or contextual interference.

[0090] This application further proposes a specific method for determining whether there are abnormal first multi-source spatial data when multi-source spatial data includes laser pulse reflection signals with different polarization directions, so as to improve the accuracy and robustness of abnormal data identification.

[0091] The aforementioned multi-source spatial data processing method for urban design includes multi-source spatial data such as laser pulse reflection signals with different polarization directions.

[0092] The steps for determining whether multi-source spatial data contains anomalies include: Extract the first laser pulse reflection signal that matches the laser pulse emission signal identifier from the laser pulse reflection signal; Calculate the degree of linear polarization and the polarization angle of the reflected signal of the first laser pulse within a first set time period; Determine whether there is a second laser pulse reflection signal within a second set time period whose linear polarization degree and polarization angle change pattern matches the preset pattern; If so, determine whether the geometric shape of the second laser pulse reflection signal matches that of the historical normal point cloud data; if not, define the second laser pulse reflection signal as false spatial data. If not, determine whether there is a third laser pulse reflection signal whose reflection intensity value exceeds a preset reflection intensity threshold; if yes, define the third laser pulse reflection signal as saturation artifact data. If a second or third laser pulse reflection signal is present in the first laser pulse reflection signal, it is determined that there is an abnormal first multi-source spatial data in the multi-source spatial data. The first type of multi-source spatial data includes: spurious spatial data or saturation artifact data.

[0093] Specifically, multi-source spatial data includes laser pulse reflection signals with different polarization directions, typically referring to data acquired by polarization-sensitive lidar (LiDAR) systems. This data not only contains traditional distance and intensity information but also information on the polarization state of the reflected light, such as the degree of linear polarization (DoLP) and the angle of polarization (AoP). This polarization information provides additional clues about the surface material, roughness, geometry, and scattering mechanism of the target object, which is crucial for distinguishing different land cover types and identifying anomalous reflections.

[0094] The purpose of extracting the first laser pulse reflection signal that matches the laser pulse emission signal identifier from the laser pulse reflection signal is to ensure that the processed reflection signal is an effective echo corresponding to the laser pulse emitted by the system, and to eliminate interference from environmental noise or irrelevant signals.

[0095] Furthermore, the linear polarization degree and polarization angle of the reflected signal from the first laser pulse within a first predetermined time period are calculated. The linear polarization degree is used to quantify the proportion of linearly polarized components in the reflected light, while the polarization angle describes the direction of linear polarization. Changes in these parameters can indicate changes in the physical characteristics of the reflection source or environmental conditions. For example, certain anomalous reflections (such as specular reflections and multipath reflections) may cause significant changes in the polarization state.

[0096] Based on this, it is determined whether a second laser pulse reflection signal exists within a second set time period, and whether its linear polarization degree and polarization angle change pattern matches a preset pattern. The preset pattern can be established based on the polarization characteristic analysis of known anomalies (such as false reflections or specific interference sources). If such a second laser pulse reflection signal exists, it is further determined whether its geometric shape matches that of historical normalized point cloud data. Historical normalized point cloud data represents the stable geometric structure of the urban target area under normal conditions. If the geometric shape of the second laser pulse reflection signal does not match the normalized data, it is defined as false spatial data. This typically occurs due to erroneous reflections caused by moving objects, atmospheric effects, or sensor malfunctions, which exhibit anomalies in both polarization characteristics and spatial location.

[0097] If the above polarization change patterns do not match, it is determined whether a third laser pulse reflection signal exists in the first laser pulse reflection signal and whether its reflection intensity value exceeds a preset reflection intensity threshold. When a laser pulse hits a highly reflective surface or the sensor is too close to the target, the received reflection signal intensity may be too high, exceeding the dynamic range of the sensor, thus producing saturation artifacts. Such third laser pulse reflection signals are defined as saturation artifact data.

[0098] Ultimately, when a second or third laser pulse reflection signal is present in the first laser pulse reflection signal, it is determined that the multi-source spatial data contains anomalies. Therefore, the first multi-source spatial data includes spurious spatial data or saturation artifact data.

[0099] This application's solution effectively addresses the limitations of general anomaly detection methods when processing complex multi-source spatial data, particularly polarization lidar data, by introducing polarization information analysis and intensity saturation detection of laser pulse reflection signals. Specifically, by extracting and analyzing the linear polarization degree and polarization angle of the first laser pulse reflection signal, unique polarization change patterns caused by anomalous reflection sources (such as specular reflection and multipath reflection) or environmental interference can be captured. This polarization-based detection mechanism enables the system to identify spurious spatial data that might be overlooked in traditional intensity or geometric analysis.

[0100] Furthermore, by comparing the geometric shape of the second laser pulse reflection signal with historical normalized point cloud data, this scheme can verify the authenticity of polarization anomalies from a spatial dimension, thus more accurately defining anomalous signals that do not match actual ground features as false spatial data. In addition, addressing the saturation artifact problem unique to lidar data, by setting a preset reflection intensity threshold and detecting third laser pulse reflection signals with reflection intensity values ​​exceeding that threshold, this scheme can promptly identify and mark data distortion caused by sensor saturation, preventing such distorted data from misleading subsequent urban planning decisions. Therefore, this scheme constructs a more comprehensive and refined anomaly detection framework from three dimensions: polarization characteristics, geometric consistency, and signal intensity saturation, significantly improving the ability to identify various anomalies in polarized lidar data.

[0101] Through the above technical solution, this application provides a more refined and robust anomaly detection mechanism for multi-source spatial data containing laser pulse reflection signals with different polarization directions. Specifically, by analyzing the variation patterns of linear polarization degree and polarization angle, and combining this with geometrical comparison with historical normal point cloud data, it can effectively identify and eliminate false spatial data caused by non-ground object reflections or environmental interference, significantly reducing the false judgment rate that may occur with traditional methods. Simultaneously, by detecting saturation artifact data where the reflection intensity value exceeds a preset threshold, it avoids the impact of data distortion caused by sensor saturation on subsequent analysis. Therefore, this solution ensures that the quality and reliability of multi-source spatial data used for urban design are significantly improved before entering the planning decision-making process, providing more accurate and reliable early warnings and suggestions for planning decision risks in target urban areas, thereby optimizing the scientific nature and effectiveness of urban planning.

[0102] This application further proposes the following steps for calculating the second confidence level of the first multi-source spatial data: The matching strength between the laser pulse reflection signal and the laser pulse emission signal, the degree of patterning of the linear polarization degree and polarization angle of the second laser pulse reflection signal, the degree of deviation of the geometric shape formed by the second laser pulse reflection signal from the normal ground object, and the degree of closeness of the reflection intensity value of the third laser pulse reflection signal to the preset reflection intensity threshold are calculated. Assign a first weight coefficient to each of the following: matching strength, patterning degree, deviation degree, and proximity degree. The second confidence level of the first multi-source spatial data is calculated based on the matching strength, patterning degree, deviation degree, proximity degree, and their respective first weight coefficients.

[0103] Specifically, the matching strength between the laser pulse reflection signal and the laser pulse emission signal refers to the degree of similarity between the received reflected signal and the original emission signal in terms of time, frequency, and phase. Its purpose is to assess the reliability of the reflected signal's source. For example, this matching degree can be quantified using cross-correlation functions or energy spectral density analysis. The patterning degree of the linear polarization degree and polarization angle of the second laser pulse reflection signal can be understood as the clarity, stability, or degree of conformity with the preset pattern when the second laser pulse reflection signal is identified as having a variation pattern consistent with the preset pattern. Its purpose is to assess the typicality or significance of the anomalous pattern. In practical applications, the deviation of the geometric shape formed by the second laser pulse reflection signal from normal ground features specifically measures the difference between the three-dimensional geometric structure constructed by the second laser pulse reflection signal and the geometric shape of historical normal ground features (such as buildings, ground, etc.) within the urban target area. For example, the deviation degree can be quantified by calculating geometric distance, shape similarity, or topological differences. Its purpose is to determine the authenticity or falsity of the anomalous geometric shape. Furthermore, the proximity of the reflection intensity value of the third laser pulse reflection signal to a preset reflection intensity threshold refers to the magnitude of the excess or the relative proportion to the threshold when the reflection intensity value of the third laser pulse reflection signal exceeds the preset reflection intensity threshold. This aims to assess the severity of the saturation artifact data. Further, first weighting coefficients are assigned to the matching strength, patterning degree, deviation degree, and proximity degree, respectively, to allow for personalized or empirical adjustments based on the contribution of different evaluation dimensions to the second confidence level. These first weighting coefficients can be determined and optimized based on expert experience, historical data analysis, or machine learning models. Therefore, based on the matching strength, patterning degree, deviation degree, proximity degree, and their respective first weighting coefficients, the second confidence level of the first multi-source spatial data is calculated through weighted summation or other aggregation functions.

[0104] This application's scheme comprehensively quantifies the anomalous characteristics of the first multi-source spatial data by considering multiple dimensions, including the matching strength between the laser pulse reflection signal and the laser pulse emission signal, the degree of pattern normalization of anomalies, the deviation of anomalous geometry from normal ground features, and the severity of saturation artifacts. Specifically, matching strength is used to assess the reliability of the signal itself, while the degree of pattern normalization and deviation analyze the possibility of false spatial data from the perspectives of signal change patterns and geometric shapes, respectively. Proximity focuses on the severity of saturation artifacts. By assigning first weight coefficients to these different dimensions of evaluation results, the influence of each factor in the final confidence calculation can be flexibly adjusted according to the actual application scenario and data characteristics. It is precisely because of this multi-dimensional, weighted aggregation evaluation mechanism that the calculation of the second confidence score can more accurately reflect the degree of anomaly and potential risks of the first multi-source spatial data, thus providing a more solid data foundation for subsequent planning decision-making risk warnings and recommendations.

[0105] Through the aforementioned technical solution, the calculation of the second confidence level is no longer a rough judgment based on a single dimension, but rather a comprehensive quantitative assessment based on multiple key indicators such as the matching intensity, patterning degree, geometric deviation, and similarity of reflection intensity of the anomalous laser pulse reflection signal. This multi-dimensional, weighted aggregation calculation method allows the obtained second confidence level to reflect the anomalous nature and potential risks of the first multi-source spatial data more precisely and accurately. Compared to relying on a single indicator or simple judgment, this solution can significantly improve the accuracy of anomalous data identification and the reliability of confidence level assessment, thereby providing more accurate and reliable risk warnings and suggestions for urban planning decisions, effectively avoiding misjudgments or decision-making errors caused by data anomalies.

[0106] See Figure 2 This application also discloses a multi-source spatial data processing system for urban design, comprising: a data acquisition module 1, a judgment module 2, and a processing module 3. The data acquisition module 1 is used to acquire multi-source spatial data of a target urban area in real time; the judgment module 2 is used to determine whether there is any abnormal first multi-source spatial data; the processing module 3 is used to, when the first multi-source spatial data exists, correct the first multi-source spatial data to obtain corrected multi-source spatial data, calculate a first confidence level of the corrected multi-source spatial data, and provide planning decision risk warnings and suggestions for the target urban area based on the corrected multi-source spatial data and the first confidence level; or, it is used to calculate a second confidence level of the first multi-source spatial data and provide planning decision risk warnings and suggestions for the target urban area based on the first multi-source spatial data and the second confidence level.

[0107] The multi-source spatial data processing system for urban design proposed in this application aims to achieve intelligent processing of urban multi-source spatial data through modular design, in order to cope with the complexity and dynamism of the urban environment. The system encapsulates data acquisition, anomaly detection, and data processing (including correction and confidence assessment) into independent modules, forming an efficient and reliable data processing architecture. Therefore, this system can effectively solve the bias and distortion problems existing in traditional methods when processing multi-source spatial data in complex urban contexts, thus providing a more accurate and reliable basis for urban planning decisions.

[0108] Specifically, the above embodiments have already described the specific methods and objectives for real-time acquisition of multi-source spatial data of the urban target area, determining whether there is abnormal first multi-source spatial data, correcting the first multi-source spatial data and calculating a first confidence level, and calculating a second confidence level of the first multi-source spatial data, which will not be repeated here. It should be emphasized that the system proposed in this application implements the above functions through the following modules: Acquisition module 1 is configured to acquire multi-source spatial data of a target urban area in real time. In a preferred embodiment, acquisition module 1 can be an integrated hardware interface unit designed to directly connect to various sensors (e.g., LiDAR, environmental sensors, cameras, microphones, etc.) and convert raw physical signals into digital data streams. This hardware interface unit may include multiple data ports and protocol converters to be compatible with different types and formats of data sources. In another embodiment, acquisition module 1 can be a software agent deployed in a distributed computing environment, communicating with remote data sources (e.g., cloud databases, third-party APIs, geographic information systems) via a network interface and periodically or event-drivenly retrieving the required multi-source spatial data. This software agent may include data caching mechanisms and preliminary data formatting functions to ensure data real-time performance and availability.

[0109] The judgment module 2 is configured to determine whether the multi-source spatial data contains anomalies. Specifically, judgment module 2 can be a rule-based expert system with a pre-defined series of logical rules and thresholds for identifying data anomalies. For example, when the received data value exceeds the preset normal range, the data change rate is abnormal, or the data pattern significantly deviates from the historical normal pattern, this module can trigger an anomaly alarm. Alternatively, judgment module 2 can be an analysis engine based on a machine learning model. This model learns the inherent patterns and distribution characteristics of the data by training on a large amount of historical normal data. When new multi-source spatial data is input, the model can calculate its deviation from the normal pattern and determine whether an anomaly exists. This machine learning model can employ supervised or unsupervised learning methods to adapt to different types of anomaly detection needs.

[0110] Processing module 3 is configured to, when first multi-source spatial data exists, correct the first multi-source spatial data to obtain corrected multi-source spatial data, calculate the first confidence level of the corrected multi-source spatial data, and provide planning decision risk warnings and suggestions for the urban target area based on the corrected multi-source spatial data and the first confidence level; or, to calculate the second confidence level of the first multi-source spatial data, and provide planning decision risk warnings and suggestions for the urban target area based on the first multi-source spatial data and the second confidence level. Specifically, processing module 3 can be a data processing pipeline that integrates multiple data correction algorithms and confidence assessment models. When judgment module 2 identifies abnormal data, processing module 3 can automatically select an appropriate correction algorithm (e.g., interpolation, reconstruction, filtering, etc.) to process the data according to the type and context of the abnormal data to generate corrected multi-source spatial data. Subsequently, the module will call the confidence assessment model to quantitatively evaluate the corrected data or the uncorrected abnormal data to obtain the first confidence level or the second confidence level. Finally, processing module 3 combines the processed data and confidence levels to generate risk warnings and recommendations for planning decisions through the decision support system, and presents them to the user. Alternatively, processing module 3 can be an interactive data analysis platform that allows users to manually select correction algorithms, adjust confidence assessment parameters, and personalize warnings and recommendations according to their actual needs. This platform can provide a visual interface so that users can intuitively understand the data processing process and results.

[0111] Compared with existing traditional urban spatial data processing systems, the system proposed in this application has significant advantages and innovations. Traditional systems, when processing multi-source spatial data, often lack effective mechanisms for identifying and correcting inherent data biases and context-dependent distortions. Their data processing flow is typically linear, and their ability to handle anomalous data is limited. For example, when temporary activities or construction occur in the city, traditional systems may fail to distinguish between false spatial information or data lacking spatial representativeness caused by these activities, leading to biases in the data fusion results.

[0112] The core innovation of this application lies in its modular design, which decouples data acquisition, anomaly detection, and data processing functions, and introduces a flexible anomaly data handling strategy. Acquisition module 1 ensures the comprehensiveness and real-time nature of the data; detection module 2 intelligently identifies anomalies in the data, avoiding the negative impact of anomalies on subsequent decisions; and processing module 3 provides two flexible anomaly data processing paths: correcting and evaluating the confidence level of the corrected data, or directly evaluating the confidence level of the anomaly data. This system architecture enables the system to adaptively evaluate and correct inherent biases in the data source, effectively identifying and processing false spatial information and data lacking spatial representativeness caused by temporary urban activities or construction activities. By assigning confidence levels to corrected or uncorrected anomaly data, the system of this application can quantify the reliability of the data, thereby avoiding decision-making errors caused by over-reliance on empirical weights in traditional systems. Therefore, the system proposed in this application can provide urban planners with more accurate and reliable urban environmental assessment results, effectively improving the scientific nature of urban design and the livability experience for residents.

[0113] The above description is merely an embodiment of this application and is not intended to limit the scope of protection of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A multi-source spatial data processing method for urban design, characterized in that, Includes the following steps: Real-time acquisition of multi-source spatial data of target urban areas; The first multi-source spatial data used to determine whether there are anomalies in the multi-source spatial data; When there is first multi-source spatial data, the first multi-source spatial data is corrected to obtain multi-source spatial corrected data. The first confidence level of the multi-source spatial corrected data is calculated. Based on the multi-source spatial corrected data and the first confidence level, planning decision risk warnings and suggestions are provided for the target area of ​​the city. or, Calculate the second confidence level of the first multi-source spatial data, and based on the first multi-source spatial data and the second confidence level, provide early warning and suggestions for planning decision-making risks for the target area of ​​the city.

2. The multi-source spatial data processing method for urban design according to claim 1, characterized in that, The multi-source spatial data includes point cloud data and environmental data; Following the steps of real-time acquisition of multi-source spatial data of the target urban area, the following are also included: Obtain the schedule and spatial scope of temporary activities in the target urban area; The specific steps for determining whether multi-source spatial data contains anomalies include: Extract the multispectral reflectance features of the first point cloud data whose reflectance intensity value exceeds the preset intensity value from the point cloud data. Search for the spectral curve that matches the multispectral reflectance features in the preset high reflectance spectral feature library. After finding the match, define the first point cloud data as spurious spatial data. Extract primary environmental data affected by temporary activities from environmental data based on timelines and spatial extent; When the first difference between the first environmental data and the second environmental data of the target urban area under historical normal conditions exceeds the first preset value, and the second difference between the first environmental data and the environmental data of the target urban area that has not been affected by temporary activities exceeds the second preset value, the first environmental data is defined as losing spatial representativeness. When the first point cloud data is false spatial data and / or the first environmental data loses spatial representativeness, it is determined that there is an abnormal first multi-source spatial data in the multi-source spatial data. The first multi-source spatial data includes first point cloud data and / or first environmental data.

3. The multi-source spatial data processing method for urban design according to claim 2, characterized in that, The specific steps for calculating the first confidence level of multi-source spatially corrected data include: The data quality score, contextual interference score, correction effect score, and cross-source consistency score of the first multi-source spatial data were evaluated. Weighting coefficients were assigned to the data quality score, contextual interference score, correction effect score, and cross-source consistency score, respectively. The first confidence level of the multi-source spatially corrected data is calculated based on the data quality score, contextual interference score, correction effect score, cross-source consistency score, and their respective weight coefficients.

4. The multi-source spatial data processing method for urban design according to claim 3, characterized in that, The multi-source spatial data also includes video stream data, audio data, and the reflected light intensity and direction of point cloud data; environmental data includes: ambient temperature data; The specific steps for evaluating the contextual disturbance score of the first multi-source spatial data include: Based on the reflected light intensity and direction, calculate the variance of the reflected light intensity of the point cloud data within a set time period, and the directional variation frequency of the reflected peak value. Calculate the dynamic reflectivity value of the point cloud data based on variance and orientation change frequency; Calculate the image gradient change rate in the video stream data, and calculate the inter-frame motion blur index based on the image gradient change rate; Search the preset heat dissipation device operation voiceprint database for the first voiceprint data that matches the audio data voiceprint; Once the first voiceprint data is found, calculate the temperature gradient in the ambient temperature data and the area of ​​the region where the temperature gradient exceeds a preset threshold. Calculate the heat source intensity index of the ambient temperature data based on the temperature gradient and the area of ​​the region. The contextual interference score of the first multi-source spatial data is evaluated based on the dynamic reflection activity value, motion ambiguity index, and heat source intensity index.

5. The multi-source spatial data processing method for urban design according to claim 3, characterized in that, The specific steps for correcting the first multi-source spatial data to obtain multi-source spatially corrected data include: Collect historical normalized point cloud data of the area corresponding to the first point cloud data and / or normalized point cloud data of the target urban area that has not been affected by temporary activities; Based on historical normalized point cloud data and / or normalized point cloud data, the first point cloud data is reconstructed to obtain point cloud reconstruction correction data, and / or, Collect historical routine environmental data of the area corresponding to the first environmental data and / or routine environmental data of the target urban area that has not been affected by temporary activities; Based on historical and / or normalized environmental data, the first environmental data is corrected to obtain corrected environmental data. Multi-source spatial correction data includes point cloud reconstruction correction data and / or environmental correction data.

6. The multi-source spatial data processing method for urban design according to claim 5, characterized in that, The steps for evaluating the correction effect score of the first multi-source spatial data include: Calculate the root mean square error of the point cloud deviation between the reconstructed and corrected point cloud data and the preset standard reference data; Calculate the matching degree between the statistical distribution of reflection intensity of the point cloud reconstruction correction data and the preset material reflection intensity distribution; Based on the root mean square error of the point cloud deviation and the matching degree, calculate the first correction effect score of the point cloud reconstruction correction data; Calculate the temperature standard residual and humidity calibration residual between the calculated environmental correction data and the preset standard environmental data; Calculate the second correction effect score for the environmental correction data based on the temperature standard residual and the humidity standard residual; The first and second correction effect scores are weighted and aggregated to obtain the correction effect score of the first multi-source spatial data.

7. The multi-source spatial data processing method for urban design according to claim 3, characterized in that, The specific steps for evaluating the data quality score of the first multi-source spatial data include: Calculate the sparsity and noise level of the first point cloud data, and calculate the first quality score of the first point cloud data based on the sparsity and noise level; Calculate the stability and missing rate of the first environmental data during data collection, and calculate the second quality score of the first environmental data based on the stability and missing rate; The data quality score of the first multi-source spatial data is evaluated based on the first quality score and the second quality score. The specific steps for evaluating the contextual disturbance score of the first multi-source spatial data include: Based on the schedule and spatial extent of the temporary activities, calculate the degree value of the first point cloud data affected by the temporary activities and the first degree value of the first environmental data affected by the temporary activities. The contextual interference score of the first multi-source spatial data is evaluated based on the degree value and the first degree value.

8. The multi-source spatial data processing method for urban design according to claim 1, characterized in that, Multi-source spatial data includes laser pulse reflection signals with different polarization directions; The steps for determining whether multi-source spatial data contains anomalies include: Extract the first laser pulse reflection signal that matches the laser pulse emission signal identifier from the laser pulse reflection signal; Calculate the degree of linear polarization and the polarization angle of the reflected signal of the first laser pulse within a first set time period; Determine whether there is a second laser pulse reflection signal within a second set time period whose linear polarization degree and polarization angle change pattern matches the preset pattern; If so, determine whether the geometric shape of the second laser pulse reflection signal matches that of the historical normal point cloud data; if not, define the second laser pulse reflection signal as false spatial data. If not, determine whether there is a third laser pulse reflection signal whose reflection intensity value exceeds a preset reflection intensity threshold; if yes, define the third laser pulse reflection signal as saturation artifact data. If a second or third laser pulse reflection signal is present in the first laser pulse reflection signal, it is determined that there is an abnormal first multi-source spatial data in the multi-source spatial data. The first type of multi-source spatial data includes: spurious spatial data or saturation artifact data.

9. The multi-source spatial data processing method for urban design according to claim 8, characterized in that, The specific steps for calculating the second confidence level of the first multi-source spatial data include: The matching strength between the laser pulse reflection signal and the laser pulse emission signal, the degree of patterning of the linear polarization degree and polarization angle of the second laser pulse reflection signal, the degree of deviation of the geometric shape formed by the second laser pulse reflection signal from the normal ground object, and the degree of closeness of the reflection intensity value of the third laser pulse reflection signal to the preset reflection intensity threshold are calculated. Assign a first weight coefficient to each of the following: matching strength, patterning degree, deviation degree, and proximity degree. The second confidence level of the first multi-source spatial data is calculated based on the matching strength, patterning degree, deviation degree, proximity degree, and their respective first weight coefficients.

10. A multi-source spatial data processing system for urban design, characterized in that, include: The acquisition module is used to collect multi-source spatial data of the target area in the city in real time; The judgment module is used to determine whether there is any abnormal first multi-source spatial data in the multi-source spatial data; The processing module is used to correct the first multi-source spatial data when it exists, to obtain multi-source spatially corrected data, to calculate the first confidence level of the multi-source spatially corrected data, and to provide planning decision risk warnings and suggestions for the target area of ​​the city based on the multi-source spatially corrected data and the first confidence level. Alternatively, it can be used to calculate the second confidence level of the first multi-source spatial data, and based on the first multi-source spatial data and the second confidence level, provide early warning and suggestions for planning decision risks in the target urban area.

Images