A parking lot position calibration method, system, medium and product

Abstract

A parking lot location calibration method, system, medium, and product are disclosed, relating to the field of data processing technology. The method includes: acquiring a parking payment dataset, initial location coordinates of the target parking lot, and physical attribute data; extracting a set of valid payment location coordinates; determining multiple location clusters; identifying target location clusters whose average distance between valid payment location coordinates and their center coordinates is less than or equal to a first preset maximum distance threshold as valid location clusters, obtaining their number as a first quantity, and using their center coordinates as the valid location center coordinates; counting the number of locations whose distance between the initial location coordinates and the valid location center coordinates is less than a second preset maximum distance threshold, obtaining a second quantity, and calculating the ratio of the second quantity to the first quantity as a first coordinate confidence level; when the distance is less than a preset confidence threshold, calibration is performed based on the center coordinates of each valid location to obtain calibrated location coordinates. This application can improve the accuracy of parking lot location data.

Description

Technical Field

[0001] This application relates to the field of data processing technology, and in particular to a parking lot location calibration method, system, medium, and product. Background Technology

[0002] With the rapid development of smart city and intelligent transportation technologies, smart parking services have become an important part of urban management. Parking lot location data, especially its latitude and longitude coordinates, is the core foundation for accurate user navigation, real-time parking space guidance, and the coordinated allocation of urban parking resources. In practical applications, this location data mainly relies on self-reporting by parking lot operators. Due to manual input errors, confusion with complex scenarios involving multiple entrances and exits, and even deliberate misreporting to obtain inappropriate traffic, the location information in the platform database is generally inaccurate. This not only directly causes navigation errors and difficulties in finding vehicles for users, seriously affecting user experience, but also distorts the results of location-based parking resource statistics and analysis, hindering the progress of urban smart transportation construction.

[0003] To correct the aforementioned location data discrepancies and improve data reliability, it is common practice to dispatch verification personnel with professional GPS positioning equipment to the target parking lot to measure and obtain the precise latitude and longitude coordinates of the main entrance or core area of ​​the parking lot. The back-end management personnel then manually update the database with the measured precise coordinates, thereby correcting the erroneous data for a single parking lot and improving the accuracy of the location data to some extent.

[0004] While manual on-site surveys can achieve high positioning accuracy in a single task, their inherent execution method introduces new and insurmountable technical problems. This method heavily relies on manual labor, resulting in high verification costs and low efficiency. Faced with thousands of parking lots in a city that are constantly changing, its coverage is extremely limited, making it impossible to achieve routine, automated updates to massive amounts of location data. Especially when parking lot location information changes due to expansion, renovation, or changes in entrances and exits, this static, passive verification mode often leads to the embarrassing situation where data calibration becomes outdated, reducing the accuracy of massive parking lot location data. Summary of the Invention

[0005] This application provides a parking lot location calibration method, system, medium, and product to address the technical problem of how to improve the accuracy of massive parking lot location data.

[0006] In a first aspect, embodiments of this application provide a parking lot location calibration method, including: At the end of the current preset calibration period, obtain the parking payment dataset, the initial location coordinates of the target parking lot, and the physical attribute data of the parking payment dataset when the user pays within the preset coordinate range during the preset data collection period; Extract the set of valid payment location coordinates from the parking payment dataset, wherein the set of valid payment location coordinates includes multiple valid payment location coordinates; Based on the physical attribute data, multiple location clusters are determined in the set of valid payment location coordinates; Calculate the average distance between all valid payment location coordinates of the target location cluster and the center coordinates corresponding to the target location cluster, wherein the target location cluster is any of the location clusters; The target location clusters whose average distance is less than or equal to the first preset maximum distance threshold are taken as effective location clusters, the number of effective location clusters is taken as the first number, and the center coordinates of the effective location clusters are taken as the center coordinates of the effective locations. The number of effective position center coordinates whose distance from the initial position coordinates to each effective position center coordinate is less than a second preset maximum distance threshold is counted as the second number. The ratio of the second number to the first number is used as the first coordinate confidence of the initial position coordinates. The second preset maximum distance threshold is greater than the first preset maximum distance threshold. When the confidence level of the first coordinate is less than the preset confidence threshold, the initial position coordinates are calibrated based on the center coordinates of each valid position to obtain the calibrated position coordinates of the target parking lot in the current preset calibration period.

[0007] Optionally, before acquiring the parking payment dataset, the initial location coordinates of the target parking lot, and the physical attribute data parking payment dataset when the user pays within the preset coordinate range during the preset data collection period at the end of the current preset calibration period, the method further includes: acquiring a second coordinate confidence level of the calibration location coordinates of the previous preset calibration period, and using the calibration location coordinates of the previous preset calibration period as the initial location coordinates of the target parking lot; adjusting the previous preset calibration period based on the second coordinate confidence level to obtain the current preset calibration period; acquiring the average daily traffic flow of the target parking lot during the current preset calibration period; when the average daily traffic flow is less than a preset traffic flow threshold, using a first data collection period as the preset data collection period; when the average daily traffic flow is greater than or equal to the preset traffic flow threshold, using a second data collection period as the preset data collection period, wherein the first data collection period is longer than the second data collection period.

[0008] Optionally, the parking payment dataset includes multiple payment data entries, each of which includes at least payment location coordinates and a positioning accuracy identifier. Extracting the set of valid payment location coordinates from the parking payment dataset includes: determining the positioning error of each payment data entry based on the positioning accuracy identifier; identifying payment data whose payment location coordinates are within the preset coordinate range and whose positioning error is less than or equal to a preset error threshold as valid payment data; identifying the payment location coordinates corresponding to each valid payment data entry as each valid payment location coordinate; combining the valid payment location coordinates to obtain the set of valid payment location coordinates; adjusting the preset data acquisition period to make the ratio greater than the preset ratio threshold when the ratio of the number of valid payment data entries to the total number of payment data entries is less than or equal to a preset ratio threshold; and adjusting the preset data acquisition period to make the number of valid payment location coordinates greater than or equal to the preset number threshold when the number of valid payment location coordinates is less than a preset number threshold.

[0009] Optionally, the physical attribute data further includes the area of ​​the target parking lot. Before counting the number of effective location center coordinates whose distance from the initial location coordinates to each effective location center coordinate is less than a second preset maximum distance threshold, the method further includes: when the area is less than or equal to a first preset area threshold, using the first preset distance as the second preset maximum distance threshold; when the area is greater than the first preset area threshold and less than or equal to the second preset area threshold, using the second preset distance as the second preset maximum distance threshold, wherein the second preset distance is greater than the first preset distance; when the area is greater than the second preset area threshold, using a third preset distance as the second preset maximum distance threshold, wherein the third preset distance is greater than the second preset distance.

[0010] Optionally, the initial position coordinates include initial longitude and initial latitude. Before counting the number of effective position center coordinates whose distance from the initial position coordinates to each of the effective position center coordinates is less than a second preset maximum distance threshold, the method further includes: calculating the distance between the initial position coordinates and the target effective position center coordinates in the following manner, wherein the target effective position center coordinates are any of the effective position center coordinates, and the target effective position center coordinates include the target center longitude and the target center latitude: the distance = Earth radius × arccos[sin(initial latitude) × sin(target center latitude) + cos(initial latitude) × cos(target center latitude) × cos(initial longitude - target center longitude)].

[0011] Optionally, the calibration location coordinates include calibration latitude and calibration longitude. When the confidence level of the first coordinate is less than a preset confidence threshold, the initial location coordinates are calibrated based on the center coordinates of each valid location to obtain the calibration location coordinates of the target parking lot. This includes: taking the number of all valid payment location coordinates in the valid location cluster corresponding to the center coordinates of the target valid location as a third quantity, where the center coordinates of the target valid location are any valid location center coordinates, and the center coordinates of the target valid location include the center longitude and the center latitude; taking the number of all valid payment location coordinates in the valid location cluster corresponding to the center coordinates of the target valid location as a fourth quantity; taking the ratio of the fourth quantity to the third quantity as the target weight of the center coordinates of the target valid location; taking the product of the center latitude and the target weight as the calibration sub-latitude of the center coordinates of the target valid location, and taking the product of the center longitude and the target weight as the calibration sub-longitude of the center coordinates of the target valid location; taking the sum of all the calibration sub-latitudes as the calibration latitude, and taking the sum of all the calibration sub-longitudes as the calibration longitude.

[0012] Optionally, the physical attribute data further includes the initial geographical range data of the target parking lot, and the method further includes: calculating the deviation distance between the calibration location coordinates and the initial location coordinates; when the deviation distance is greater than a preset coordinate deviation threshold, calling a preset API interface to query and obtain the preset geographical range parameters of the target parking lot based on the calibration location coordinates; calculating the estimated geographical range of the target parking lot based on the calibration location coordinates and the preset geographical range parameters; and updating the physical attribute data of the target parking lot with the estimated geographical range to cover the initial geographical range data.

[0013] In a second aspect, embodiments of this application provide a parking lot location calibration system, which includes: one or more processors and a memory; the memory is coupled to the one or more processors, and the memory is used to store computer program code, which includes computer instructions, and the one or more processors call the computer instructions to cause the parking lot location calibration system to perform the method described in the first aspect and any possible implementation thereof.

[0014] Thirdly, embodiments of this application provide a computer program product containing instructions that, when the computer program product is run on a parking lot location calibration system, cause the parking lot location calibration system to perform the method described in the first aspect and any possible implementation thereof.

[0015] Fourthly, embodiments of this application provide a computer-readable storage medium including instructions that, when executed on a parking lot location calibration system, cause the parking lot location calibration system to perform the method described in the first aspect and any possible implementation thereof.

[0016] In summary, one or more technical solutions provided in this application have at least the following technical effects or advantages: 1. By adopting a periodic triggering, multi-source data fusion, and clustering verification mechanism, at the end of a calibration cycle, based on the massive location data generated when users pay in the parking lot and the reported initial location coordinates and physical attribute data, automated clustering analysis and confidence calculation are performed. When the confidence is insufficient, the calibration process is initiated, thereby achieving normalized, automated, and intelligent calibration of parking lot location data. This reduces the manpower and time costs of data maintenance and can dynamically respond to possible changes in parking lot locations, significantly improving the accuracy, timeliness, and reliability of location data in the smart parking system. This provides a solid data foundation for precise navigation and resource scheduling. By introducing a multi-dimensional adaptive parameter adjustment and data quality control mechanism, the accuracy of the entire calibration method is systematically improved.

[0017] 2. By dynamically adjusting the confidence level of historical calibration results and differentiating the data collection cycle based on traffic flow, the system resources can intelligently focus on parking lots with large data fluctuations or slow sample collection. This optimizes the allocation of computing and storage resources while ensuring calibration effectiveness. During the data preprocessing stage, payment data with high positioning accuracy and reasonable location are strictly screened, and a dual feedback adjustment mechanism for data quality and quantity (adjusting the collection cycle) is set up to ensure the high reliability of the data input to the subsequent clustering analysis module. This effectively avoids interference from low-quality or malicious data on calibration results, laying the foundation for high-precision calibration. The verification threshold (second preset maximum distance threshold) is correlated with the actual physical scale (area) of the parking lot, and a precise geodetic distance formula is used for calculation. This makes the judgment standard for location accuracy more scientific, reasonable, and in line with the actual scenario, improving the accuracy and fairness of calibration decisions and ensuring the accuracy and physical rationality of calibration results.

[0018] 3. By calculating calibration coordinates using a weighted average method with the number of user location points as the weight, areas with more frequent user activity contribute more to the final calibration results. The calibrated location coordinates more accurately reflect the core area of ​​the parking lot in actual use, rather than simply the geometric center. The calibration results are more realistic and practical. By calculating the deviation distance before and after coordinate calibration, map API queries and geographic range recalculation are only triggered when the deviation is significant, achieving linked calibration and accurate updates from location to range. This avoids indiscriminate resource consumption, significantly reduces the system's computation and call costs while ensuring the consistency and timeliness of parking lot spatial data, and greatly enhances the fault tolerance and reliability of the entire system's output results. Attached Figure Description

[0019] Figure 1 This is a schematic flowchart of the parking lot location calibration method provided in the embodiments of this application; Figure 2 This is a schematic diagram of the periodic confirmation process provided in the embodiments of this application; Figure 3 This is a schematic diagram of the structure of a parking lot location calibration system provided in an embodiment of this application.

[0020] Explanation of reference numerals in the attached figures: 601, Central Processing Unit; 602, Read-Only Memory; 603, Random Access Memory; 604, Bus; 605, Input / Output Interface; 606, Input Section; 607, Output Section; 608, Storage Section; 609, Communication Section; 610, Driver; 611, Removable Media. Detailed Implementation

[0021] To make the objectives, technical solutions, and advantages of this application clearer, the application will be further described in detail below with reference to the accompanying drawings. The described embodiments should not be regarded as limitations on this application. All other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0022] In the description of the embodiments of this application, words such as "illustrative," "for example," or "for example" are used to indicate examples, illustrations, or explanations. Any embodiment or design described as "illustrative," "for example," or "for example" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or design solutions. Rather, the use of words such as "illustrative," "for example," or "for example" is intended to present the relevant concepts in a specific manner.

[0023] In the description of the embodiments of this application, the terms "first, second, third" are used only to distinguish similar objects and do not represent a specific order of objects. It is understood that "first, second, third" can be interchanged in a specific order or sequence where permitted, so that the embodiments of this application described herein can be implemented in an order other than that illustrated or described herein.

[0024] In the embodiments of this application, the terms "module" or "unit" refer to a computer program or part of a computer program that has a predetermined function and works with other related parts to achieve a predetermined goal, and can be implemented wholly or partially using software, hardware (such as processing circuitry or memory), or a combination thereof. Similarly, a processor (or multiple processors or memory) can be used to implement one or more modules or units. Furthermore, each module or unit can be part of an overall module or unit that includes the functionality of that module or unit. The terms "comprising," "including," "having," and variations thereof all mean "including but not limited to," unless otherwise specifically emphasized.

[0025] In the implementation of this application, the collection and processing of relevant data should strictly comply with the requirements of relevant national laws and regulations, obtain the informed consent or separate consent of the personal information subject, and carry out subsequent data use and processing within the scope of laws and regulations and the authorization of the personal information subject.

[0026] Unless otherwise defined, all technical and scientific terms used in the embodiments of this application have the same meaning as commonly understood by one of ordinary skill in the art. The terminology used in the embodiments of this application is for the purpose of describing the embodiments of this application only and is not intended to limit this application.

[0027] In related technologies, the heavy reliance on human input leads to high verification costs and low efficiency. Faced with thousands of parking lots in cities that are constantly changing, the coverage is extremely limited, making it impossible to achieve normalized and automated updates of massive location data, thus reducing the accuracy of massive parking lot location data. To address the above problems, this application provides a parking lot location calibration method, system, medium, and product that can improve the accuracy of massive parking lot location data.

[0028] Figure 1 This is a flowchart illustrating the parking lot location calibration method provided in the embodiments of this application.

[0029] This invention discloses a parking lot location calibration method, such as... Figure 1 As shown, the steps include the following.

[0030] S101. At the end of the current preset calibration period, acquire the parking payment dataset, the initial location coordinates of the target parking lot, and the physical attribute data when the user makes a payment within the preset coordinate range during the preset data collection period.

[0031] Specifically, before acquiring the parking payment dataset, users explicitly authorize the platform to use their transaction data for parking lot operation optimization and location calibration after anonymization and aggregation processing upon their first use of the parking service, through the user agreement. The system only extracts non-sensitive fields authorized by the user (such as transaction time, payment amount, parking lot identifier, etc.). The data comes from aggregated logs directly generated by the parking lot's internal billing system. These logs have been de-identified by the parking lot management and do not contain any personally identifiable information (such as name, mobile phone number, license plate number, etc.). When third-party payment gateways are involved, data acquisition complies with compliance agreements signed with the gateways, ensuring that all data complies with the Personal Information Protection Law and related privacy policies during transmission and use, and only data within the scope of explicit user consent is used.

[0032] When the current preset calibration cycle, pre-set and maintained by the system, reaches its end time or meets the end conditions (e.g., the parking management platform server), the data acquisition process will be automatically activated. First, it retrieves and aggregates data records generated by all users completing payment transactions within a preset coordinate range (parking lot internal payment) during another independent preset data collection cycle from the payment gateway or application logs, forming a parking payment dataset. Second, it queries and reads the latest registered or previously confirmed latitude and longitude coordinates of the target parking lot from the parking lot's asset information database, using them as the reference benchmark for this calibration, i.e., the initial location coordinates. Finally, it extracts structured data describing the physical state of the parking lot from the same asset information database or geographic information system, constituting physical attribute data. The acquisition of these three data packets is the foundation and input for all subsequent analysis, verification, and calibration operations.

[0033] The current preset calibration period represents a configurable, cyclical time period used to periodically trigger a complete calibration process for parking lot location data. Its length can be a fixed value (e.g., 30 days) or dynamically adjusted based on the confidence level of historical calibration results. The preset data collection period represents a specific time interval used to accumulate user payment behavior data. Its length differs from the current preset calibration period and is typically set to a shorter time period (e.g., 3 or 5 days). It aims to collect sufficiently fresh and adequately sampled user location data and is a time period within the current preset calibration period. It can be a historical period ending with the current preset calibration period or any time period within the current calibration period. The preset coordinate range is typically a polygonal geofence. Its boundaries can be the boundaries of the city or administrative division where the target parking lot is located, or a reasonable area formed by buffering the initial location coordinates of the parking lot. This is used to exclude coordinates that are obviously impossible to locate due to equipment errors or human interference. The parking payment dataset refers to a structured collection of data synchronously recorded and uploaded by the payment system when a user completes parking fee settlement at a target parking lot using a specified payment method (such as mobile application or QR code payment) within a data collection period. Each record includes at least the payment time, the user's anonymous identifier, the latitude and longitude coordinates provided by the device at the time of payment, and the positioning accuracy. The target parking lot refers to the specific parking lot entity with a unique identifier in the smart parking system, which is the focus of this calibration process. The initial location coordinates refer to the latitude and longitude coordinates of the target parking lot currently recorded and provided to the public in the system's location database. These may be the coordinates initially reported by the parking lot operator or the result of the previous calibration process. Physical attribute data refers to structured data used to describe the inherent physical characteristics of the target parking lot, which may include, but is not limited to, the parking lot's floor area, number of entrances and exits, design capacity, number of floors (for multi-story parking garages), and geofence boundary coordinates, providing important contextual information for subsequent cluster analysis and threshold setting.

[0034] Figure 2 This is a schematic diagram of the periodic confirmation process provided in the embodiments of this application.

[0035] Based on the above embodiments, as an optional embodiment, see [link to embodiment]. Figure 2 ,exist Figure 1 Before step S101 shown, it also includes Figure 2 Steps S201-S205 are explained in detail below.

[0036] S201. Obtain the second coordinate confidence level of the calibration position coordinates of the previous preset calibration cycle, and use the calibration position coordinates of the previous preset calibration cycle as the initial position coordinates of the target parking lot.

[0037] Specifically, before initiating the full calibration process for the current preset calibration cycle of the target parking lot, it is first necessary to establish the baseline state for this calibration. In order to assess the reliability trend of the calibration position coordinates of the previous calibration cycle at the current moment and to provide a basis for decision-making in subsequent processes, the system will call the second coordinate confidence level associated with the calibration position coordinates, which is saved in the history record or recalculated based on historical data at the current moment. This confidence level measures the reliability of the previous calibration result and is a key feedback signal that drives the system to make dynamic adjustments. At the same time, the system will read the historical calibration record and obtain the coordinate results calculated and finally confirmed by the calibration process at the end of the calibration cycle immediately preceding the current cycle (i.e., the previous preset calibration cycle). This historical coordinate result is officially designated as the reference object to be evaluated and corrected in this calibration process, that is, it becomes the initial position coordinates of the target parking lot.

[0038] The calibration location coordinates of the previous preset calibration cycle refer to the final latitude and longitude coordinates calculated and determined for the target parking lot after the most recent complete execution of all or core calibration steps of the parking lot location calibration method provided in this application embodiment before the start of this round of calibration process. It is the output result of historical calibration work and has been updated as the service coordinates of the parking lot in the system.

[0039] The calculation of the second coordinate confidence level of the calibration position coordinates in the previous preset calibration cycle refers to the quantitative process of assessing the reliability of historical calibration coordinates in the current data environment. The calculation here may be to directly read the final confidence level result calculated and stored in the previous calibration process (i.e., historical snapshot), or it may take into account the changes in data distribution that may occur over time. The system uses the same or similar algorithm as the previous calibration, but re-evaluates the confidence level of the historical coordinates based on the latest or current valid data (i.e., re-evaluation). Whether it is reading or recalculating, the purpose is to obtain a numerical index that reflects the current credible state of the historical coordinates. The second coordinate confidence level is different from the first coordinate confidence level. The former is based on the current status assessment of the historical coordinates, while the latter is based on the direct verification of the historical coordinates based on the current data.

[0040] S202. Adjust the previous preset calibration cycle based on the second coordinate confidence level to obtain the current preset calibration cycle.

[0041] Specifically, after obtaining the confidence level of the second coordinate, it is used as an input parameter and applied to a preset or learnable adjustment rule. This rule defines the mapping relationship between the confidence level and the calibration cycle length. By executing this rule, the duration of the time period applied to the current and subsequent calibration work is calculated, which is the current preset calibration cycle. The determination of this new cycle is a direct modification or replacement of the previous preset calibration cycle, thereby realizing the dynamic change of the calibration frequency according to the reliability of the data.

[0042] For example, suppose the system's built-in adjustment rules are: if the confidence level of the second coordinate is ≥90%, the calibration cycle is extended to 45 days; if the confidence level is between 75% and 90%, the standard cycle of 30 days is maintained; if the confidence level is <75%, the calibration cycle is shortened to 14 days. Scenario A: The system calculates that the confidence level of the second coordinate of parking lot P3003 is 92%, and determines that the calibration cycle should be extended. Therefore, it adjusts (extends) the previous preset calibration cycle (assuming it is the standard 30 days) to 45 days, and this 45 days becomes the current preset calibration cycle for P3003. Scenario B: The system calculates that the confidence level of the second coordinate of parking lot P3004 is 68%. According to the rules, the system determines that the calibration frequency should be increased, and adjusts (shortens) the preset calibration cycle to 14 days, and this 14 days becomes the current preset calibration cycle for P3004.

[0043] S203. Obtain the average daily traffic flow of the target parking lot during the current preset calibration period.

[0044] Specifically, the system identifies the past period corresponding to the current preset calibration cycle, then accesses the parking lot vehicle entry and exit record database or statistical report system. For the target parking lot, it queries and calculates the average daily number of vehicles entering and exiting during this specific historical period, i.e., the average daily traffic flow. This serves as a key parameter for measuring the parking lot's operational busyness and the potential rate of user data generation. Independent of payment data, it provides the system with another dimension of decision-making basis.

[0045] The average daily traffic volume is calculated by dividing the total traffic volume recorded within the current preset calibration period by the number of days covered by that period, thus obtaining a value representing the average daily vehicle volume.

[0046] S204. When the average daily traffic flow is less than the preset traffic flow threshold, the first data collection period shall be used as the preset data collection period.

[0047] Specifically, the average daily traffic flow calculated in the previous step is compared with a preset traffic flow threshold defined by the system or dynamically calculated; no specific restrictions are imposed here. When the comparison result shows that the average daily traffic flow is less than the preset traffic flow threshold, the system determines that the frequency of user payment events in the target parking lot is low. To ensure that a sufficient number of valid user location samples can be accumulated when the calibration analysis is initiated, the system decides to adopt a relatively long data collection time window, namely the first data collection period, and officially sets it as the preset data collection period to be used in this calibration. This is to compensate for the insufficient daily data volume by extending the data collection time for low-activity parking lots.

[0048] The preset traffic flow threshold is a reference value used to distinguish between high and low traffic flow levels in parking lots. This threshold can be a fixed value set based on historical experience (e.g., 500 vehicles / day) or a value dynamically calculated based on the traffic flow distribution of all parking lots in the area (e.g., median or a certain percentile). The first data collection period is a relatively long time length option preset by the system (e.g., 7 days, 14 days, or 30 days). Through this operation, this longer period is assigned to the variable preset data collection period, thereby determining the time range for subsequent data extraction from payment logs.

[0049] S205. When the average daily traffic flow is greater than or equal to the preset traffic flow threshold, the second data collection period shall be used as the preset data collection period, and the first data collection period shall be greater than the second data collection period.

[0050] Specifically, when the condition that the average daily traffic flow is greater than or equal to the preset traffic flow threshold is met, it is determined that the target parking lot has a high frequency of user payment events. At this time, there is no need to rely on long-term accumulation to obtain sufficient user location samples for analysis in a short period of time. In order to ensure that the data used for calibration can reflect the latest status of the parking lot to the greatest extent and reduce data processing delay, a relatively short second data collection cycle designed specifically for high-traffic scenarios is adopted and set as the preset data collection cycle used in this round of calibration. This is to ensure that for highly active parking lots, more recent data with higher freshness should be selected first to quickly respond to possible location changes.

[0051] The second data collection period is a relatively short time option preset by the system (e.g., 1 day, 3 days, or 5 days).

[0052] S102. Extract the set of valid payment location coordinates from the parking payment dataset. The set of valid payment location coordinates includes multiple valid payment location coordinates.

[0053] Specifically, after obtaining the parking payment dataset, each payment record is iterated through, and the fields such as payment location coordinates contained in the record are checked according to a series of preset validity judgment rules. Only those payment location coordinates that pass all validity checks will be retained, and the data corresponding to the remaining payment location coordinates that fail the checks will be directly deleted. All the retained payment location coordinates are organized into a new set, which is called the valid payment location coordinate set. Each coordinate contained in this set is called a valid payment location coordinate.

[0054] The valid payment location coordinate set is a set-type data structure specifically used to store filtered payment location coordinates. Valid payment location coordinates refer to geographical coordinate points from a single payment record that meet all preset validity criteria. Their validity attribute is relative to any invalid or suspicious coordinates that may exist in the original parking payment dataset.

[0055] Based on the above embodiments, as an optional embodiment, the parking payment dataset includes multiple payment data entries, each of which includes at least payment location coordinates and a positioning accuracy identifier. Figure 1 The step S102 shown can be implemented through steps S1021-S1026, which will be explained in detail below.

[0056] S1021. Based on the positioning accuracy identifier, determine the positioning error of each payment data.

[0057] Specifically, when processing each payment data item, its location accuracy identifier field is read. This field is a descriptive information provided by the user's device (such as a mobile phone) positioning module or operating system at the time of payment, which is a reliable estimate of the current location. Based on pre-established mapping rules or parsing algorithms, this identifier (which may be an enumerated value, a string description, or the original accuracy value) is converted or directly read into a specific location error value expressed in length units (usually meters). This location error represents a radius range that the actual location may be distributed with the payment location coordinates as the center. It is a core indicator for measuring the data quality of a single coordinate point and is used to objectively assess the degree of uncertainty of each payment location coordinate.

[0058] The positioning accuracy identifier specifies the data source and dependencies used to determine the positioning error. Positioning error specifically refers to the horizontal position accuracy estimate of the location information contained in the payment data record, usually understood as the radius (or a similar definition) corresponding to the confidence interval; a smaller value indicates higher accuracy.

[0059] S1022. Payment data whose payment location coordinates are within the preset coordinate range and whose positioning error is less than or equal to the preset error threshold shall be regarded as valid payment data.

[0060] Specifically, when processing each payment record, two key conditions are evaluated in parallel: First, it is determined whether the payment location coordinates (i.e., latitude and longitude values) in the data record fall within a preset coordinate range defined by the system; second, it is determined whether the positioning error value of the data determined in the previous step is less than or equal to a preset error threshold set by the system. The system will classify a payment record as valid payment data only if both conditions are met simultaneously. This determination is binary: either yes or no, with no intermediate state. All payment records determined to be valid will constitute a reliable subset of data for subsequent processing.

[0061] The preset error threshold is an upper limit value expressed in length units (usually meters), such as 10 meters, 20 meters, or 50 meters, representing the highest acceptable location uncertainty for the system. Valid payment data refers to data records that simultaneously meet the above requirements for spatial rationality and accuracy. It not only includes the location coordinates that meet the filtering criteria but also usually associates with other business information such as the original payment time and parking lot ID, making it a complete and reliable data object.

[0062] S1023. Use the payment location coordinates corresponding to each valid payment data as the coordinates of each valid payment location.

[0063] Specifically, after obtaining a set of multiple valid payment data objects, the system iterates through each object in the set and performs an attribute extraction operation on each object: it reads the value of the field named payment location coordinates in the object (this field value has been verified to be reasonable and accurate in the previous step), and then outputs or stores this extracted coordinate value as an independent, purely geospatial data point. This independent data point is called a valid payment location coordinate. By iterating through all valid payment data objects, the system will eventually obtain a set of these pure coordinate points.

[0064] Through the above embodiments, the key transformation from attributed business data objects to pure spatial data points was completed, providing a uniform and pure input for subsequent spatial algorithms such as clustering analysis. The location components with the greatest spatial analysis value contained in the business data, which have undergone rigorous quality inspection, were efficiently extracted to form a list of coordinate points specifically for spatial calculations. This not only meets the requirements of the algorithm interface but also brings advantages in data processing efficiency: subsequent algorithms can directly perform high-speed calculations on the lightweight coordinate array without having to access and parse the more complex original data objects in each iteration.

[0065] S1024. Combine the coordinates of each valid payment location to obtain the set of valid payment location coordinates.

[0066] Specifically, after generating or extracting each valid payment location coordinate, they are not left in isolation. Instead, a new set of containers specifically designed to hold geographic point data is created. All generated valid payment location coordinates are iterated over, and each coordinate point is added (or combined) as an element to this newly created set of containers. Once all independent coordinate points have been included, this full set of containers is formally formed and given a specific identifier, namely the set of valid payment location coordinates.

[0067] S1025. When the ratio of the number of valid payment data to the number of payment data is less than or equal to a preset ratio threshold, adjust the preset data collection period to make the ratio greater than the preset ratio threshold.

[0068] Specifically, after performing data validity screening, the ratio between the number of valid payment data and the total number of original parking payment datasets is calculated. This calculated ratio is then compared to a pre-set threshold representing the minimum acceptable data quality percentage. If the actual ratio is less than or equal to the threshold, it indicates that the proportion of valid data that passes the screening within the current preset data collection period is too low. This suggests that the root cause of the problem may be insufficient data collection time, resulting in a small base of collected raw data, or unsatisfactory data freshness and quality distribution. To correct this, the preset data collection period will be extended. The goal of this adjustment is to accumulate a larger amount of raw payment data within a longer collection window. The aim is to maintain or increase the absolute number of valid data while, through the law of large numbers or by covering more diverse time periods (such as periods with better location signal strength), increasing the final calculated proportion of valid data to exceed the preset threshold.

[0069] The preset ratio threshold refers to a monitoring standard based on data efficiency.

[0070] For example, assuming the system's preset ratio threshold is 70% (meaning at least 70% of the original payment data is expected to be valid), for a parking lot, the initial preset data collection period is 3 days. During the first execution: within the 3-day collection period, 90 pieces of original payment data are collected. After filtering, 54 pieces of valid payment data are obtained. The calculated ratio = 54 / 90 = 60%. The judgment is: 60% ≤ 70%, triggering the condition to extend the preset data collection period from 3 days to 5 days. Subsequently (when the next calibration is triggered, or when collection is re-executed): the system uses the new 5-day period for data collection. This time, 160 pieces of original payment data are collected, and after filtering, 128 pieces of valid payment data are obtained. The calculated new ratio = 128 / 160 = 80%. The new judgment is: 80% > 70%, the condition is no longer triggered, and the current data collection period (5 days) is considered appropriate.

[0071] S1026. When the number of valid payment location coordinates is less than the preset number threshold, adjust the preset data collection cycle so that the number of valid payment location coordinates is greater than or equal to the preset number threshold.

[0072] Specifically, after completing all data cleaning and extraction steps and generating a set of valid payment location coordinates, the system immediately assesses the size of this set, i.e., calculates the total number of valid payment location coordinates it contains. The system compares this actual number with a pre-defined threshold representing the minimum sample size required for meaningful statistical analysis. If the actual number is less than the threshold, the system determines that the currently collected valid data is insufficient to support subsequent clustering analysis and confidence calculations (e.g., too few samples can lead to unstable clustering results or unreliable confidence calculations). After diagnosing the root cause of the insufficient data, the system extends the length of the pre-defined data collection period. The goal of this adjustment is to expand the data collection time window, hoping to accumulate more user payment events over a longer time span, thereby increasing the potential number of valid location coordinates. Ultimately, this ensures that the number of valid payment location coordinates acquired within the adjusted period reaches or exceeds the pre-defined threshold.

[0073] For example, suppose the system's preset threshold is 50 (meaning at least 50 valid coordinate points are needed for reliable cluster analysis). For a newly built or low-traffic parking lot P7001: First attempt: The system uses an initial preset data collection period of 3 days. After collection and cleaning, only 30 coordinate points are obtained in the set of valid payment location coordinates. Judgment: 30 < 50. The preset data collection period is extended from 3 days to 7 days. Second attempt (as in the next calibration or re-collection): The system uses the new 7-day period for data collection. After the same cleaning process, 85 valid payment location coordinates are obtained. New judgment: 85 ≥ 50. The condition is no longer triggered. At this point, the data volume is considered sufficient, and the subsequent analysis stage can begin.

[0074] S103. Based on physical attribute data, determine multiple location clusters in the effective payment location coordinate set.

[0075] Specifically, after obtaining the cleaned set of coordinate points, a spatial clustering algorithm is invoked or executed. This algorithm does not blindly search for natural clusters in the data in an unsupervised manner. Instead, it uses physical attribute data (such as the number of entrances / exits, core area boundaries, etc.) as key parameters or constraints to guide the clustering process. For example, the algorithm might use the number of parking lot entrances / exits as an important reference for the number of clusters to be found, or use the parking lot boundary to limit the geographic space of the clustering search. In this way, points in the entire effective payment location coordinate set are assigned to one or more location clusters based on their spatial proximity and whether they conform to the expected physical structure of the parking lot. Each location cluster represents a potentially specific functional area within the parking lot that is frequently used by users, such as near an entrance / exit, the core parking area, or around a payment point.

[0076] Among them, location clusters specifically refer to each subset generated after grouping, where the coordinate points within each cluster are spatially close to each other, and the entire cluster corresponds to an inferred user activity hotspot area.

[0077] S104. Calculate the average distance between the coordinates of all valid payment locations in the target location cluster and the center coordinates of the target location cluster. The target location cluster can be any location cluster.

[0078] Specifically, after dividing all location clusters, one of them is selected as the current evaluation object, called the target location cluster. For this target cluster, the system first determines its center coordinates. These coordinates are calculated by the clustering algorithm (such as K-means) when generating the cluster, representing the average spatial position or optimal representative point of all points in the cluster. Then, it iterates through each valid payment location coordinate within the target cluster, calculating the straight-line distance between each point and the aforementioned center coordinates. After calculating the distances of all points to the center, the system sums these distance values ​​and divides them by the total number of valid payment location coordinates in the cluster, thus obtaining a value representing the average distance of all points in the cluster to the center. This value is defined as the average distance of the cluster. This step is performed sequentially for each location cluster, generating a unique average distance value for each cluster.

[0079] Among them, the target location cluster refers to one of the multiple spatial groups (i.e., location clusters) obtained after analyzing the set of effective payment location coordinates through clustering algorithms, which is then arbitrarily selected as the object to be evaluated and processed.

[0080] The above embodiments provide a precise quantitative tool for evaluating the internal consistency and data quality of each cluster, and also provide an objective basis for the next step of screening out truly reliable and effective location clusters. This ensures that the location information used for calibration is based on a high-quality, internally consistent spatial pattern, thereby improving the accuracy and robustness of the entire calibration method.

[0081] S105. The target location clusters with an average distance less than or equal to the first preset maximum distance threshold are taken as effective location clusters. The number of effective location clusters is taken as the first quantity, and the center coordinates of the effective location clusters are taken as the center coordinates of the effective locations.

[0082] Specifically, after obtaining the average distance value of each location cluster, it is compared with a predefined length standard—a first preset maximum distance threshold. For each target location cluster used as a comparison object, if its average distance is less than or equal to the threshold, the system determines that the points inside the cluster are sufficiently concentrated, forming a clear and reliable spatial clustering pattern, and therefore classifies it as a valid location cluster. Conversely, if its average distance is greater than the threshold, the cluster will be considered invalid or unreliable and excluded or directly deleted. After completing the screening of all clusters, the system counts the total number of valid location clusters that are retained, and this value is recorded as the first quantity. For each valid location cluster, the system obtains the center coordinates generated during the clustering process and outputs or marks these center coordinates as the valid location center coordinates of the corresponding cluster.

[0083] The first preset maximum distance threshold is a predefined numerical upper limit, expressed in units of length (e.g., meters), used by the system to determine the acceptable dispersion of a target location cluster. If the average distance of a cluster is less than or equal to this threshold, the cluster is considered compact and reliable. An effective location cluster refers to a target location cluster whose average distance is less than or equal to the first preset maximum distance threshold. These clusters are considered to have a compact internal structure, high data quality, and represent a clear user activity hotspot. The first quantity is the total number of all clusters identified as effective locations; it is a non-negative integer representing the number of reliable user activity hotspot areas identified in the current calibration analysis. The center coordinates of an effective location cluster are the representative spatial location of each effective location cluster, usually calculated by the clustering algorithm when generating the cluster. They represent the average or center position of all effective payment location coordinates within the cluster. The center coordinates of an effective location refer to the coordinate points directly mapped or inherited from the center coordinates of an effective location cluster. Specifically, they refer to those center coordinates derived from effective location clusters, which correspond one-to-one with each effective location cluster and together constitute the core set of location reference points upon which subsequent calibration and verification are based. S106. Count the number of valid position center coordinates whose distance from the initial position coordinates to each valid position center coordinate is less than the second preset maximum distance threshold, and use it as the second quantity. Use the ratio of the second quantity to the first quantity as the first coordinate confidence of the initial position coordinates. The second preset maximum distance threshold is greater than the first preset maximum distance threshold.

[0084] Specifically, for each valid location center coordinate, the spatial straight-line distance between it and the initial location coordinate is calculated. This distance value is then compared with a second preset maximum distance threshold set according to the parking lot size. The total number of valid location center coordinates with a distance less than this threshold is counted, and this value is recorded as the second quantity. The second quantity is then compared with the total number of previously counted valid location clusters (i.e., the first quantity). By calculating the ratio of the two (second quantity / first quantity), a value between 0 and 1 (or between 0% and 100%) is obtained. This value is defined as the first coordinate confidence of the initial location coordinate, which intuitively reflects the proportion of user activity hotspots that the initial coordinate is close to. It is clear that the second preset maximum distance threshold is numerically stricter than the first preset maximum distance threshold. This ensures that the standard used to judge whether the initial coordinate matches the user area is more lenient than the standard used to judge whether the user area is compact.

[0085] The second quantity refers to a statistically derived integer value representing the number of valid location center coordinates that meet the following condition: the spatial distance between the valid location center coordinate and the initial location coordinate to be verified is less than the second preset maximum distance threshold set by the system for this verification. This reflects how many reliable hotspot areas (i.e., valid location clusters) derived from user data the initial location coordinate can match. The second preset maximum distance threshold is a distance standard used to determine whether the initial location coordinate matches the valid location center coordinate. It is an upper limit value expressed in length units (such as meters). If the distance between a valid location center coordinate and the initial location coordinate is less than or equal to this threshold, the center coordinate is considered to match the initial location. Its value can be determined according to the size of the parking lot; for example, the threshold is larger for large parking lots and smaller for small parking lots. The first coordinate confidence level is an indicator used to quantitatively evaluate the credibility of the initial location coordinate. Its value is equal to the ratio obtained by dividing the second quantity by the first quantity, and is between 0 and 1 (or 0% to 100%). The higher the value, the more user activity hotspot areas the initial location coordinate matches, and the higher its credibility.

[0086] Based on the above embodiments, as an optional embodiment, the physical attribute data also includes the floor area of ​​the target parking lot. Figure 1 Before step S105, steps S301-S303 are also included, which will be explained in detail below.

[0087] S301. When the area occupied is less than or equal to the first preset area threshold, the first preset distance is used as the second preset maximum distance threshold.

[0088] Specifically, after obtaining the key physical attribute data of the target parking lot's area, it compares it with a system-predefined scale classification standard—a first preset area threshold. When the comparison result shows that the area is less than or equal to the first preset area threshold, the system determines that the parking lot belongs to the small-scale category. For small-scale parking lots, their geographical coverage is limited and user activity areas are relatively concentrated. Therefore, the accuracy requirements for the reported location coordinates should be more stringent, and the allowable deviation range should be smaller. Based on this logic, the system designates a relatively small distance value—the first preset distance—as the benchmark tolerance for judging whether the initial location coordinates of the parking lot are accurate in subsequent verification steps, i.e., the second preset maximum distance threshold, ensuring that the verification standard matches the actual physical scale of the parking lot.

[0089] The "land area" refers to the horizontal projected area of ​​the target parking lot on the ground, a core attribute describing its physical scale, usually measured in square meters. The first preset area threshold is a predefined standard area value used by the system to classify parking lots by size, serving as a boundary value to distinguish small parking lots; parking lots with an area equal to or less than this value are classified as small. The first preset distance is a predefined specific distance value (e.g., 50 meters) used by the system, expressed in length units (e.g., meters), designated as the applicable location verification tolerance standard for small parking lot scenarios. The second preset maximum distance threshold is an upper limit standard for judging the accuracy of the initial location coordinates reported by the parking lot; its specific value is selected from multiple preset distance values ​​(e.g., the first preset distance, the second preset distance, and the third preset distance) based on the size (land area) of the parking lot.

[0090] S302. When the area occupied is greater than the first preset area threshold and less than or equal to the second preset area threshold, the second preset distance is used as the second preset maximum distance threshold, and the second preset distance is greater than the first preset distance.

[0091] Specifically, the area occupied is compared with two preset size classification standards. When the judgment condition that the area occupied is greater than the first preset area threshold and less than or equal to the second preset area threshold is met, the parking lot is determined to belong to the medium-sized category. For this type of parking lot, its geographical range is wider than that of a small parking lot, but it has not yet reached the level of a large parking lot. Therefore, it is neither possible to use the strict standards applicable to small parking lots, nor can it directly apply the lenient standards applicable to large parking lots. For this reason, a relatively compromise distance value between the two is set - the second preset distance. This value is officially designated as the second preset maximum distance threshold used by the parking lot in subsequent verification steps, ensuring that the verification standard can be adapted to the spatial characteristics of parking lots of different sizes in a gradient manner.

[0092] The second preset area threshold refers to another area boundary value (e.g., 5000 square meters) pre-set by the system, used to classify parking lot size levels. Together with the first preset area threshold, it defines a size range within which parking lots are classified as medium-sized parking lots. The second preset distance refers to a distance value (e.g., 100 meters) pre-set by the system for medium-sized parking lots. When the parking lot's area meets the medium-sized condition, this value will be used as a tolerance standard to assess its location accuracy, i.e., the second preset maximum distance threshold.

[0093] S303. When the area occupied is greater than the second preset area threshold, the third preset distance is used as the second preset maximum distance threshold, and the third preset distance is greater than the second preset distance.

[0094] Specifically, the system compares the area occupied with the highest threshold in the classification system—the second preset area threshold. When the judgment condition that the area occupied is greater than the second preset area threshold is met, the system determines that the parking lot belongs to the large-scale category. These parking lots typically have a wide geographical coverage and may contain multiple scattered parking areas, entrances and exits, and ancillary facilities. The spatial span is large, and the accuracy assessment of the reported location coordinates requires a relatively lenient tolerance standard that can adapt to this large spatial scale. For this purpose, a large distance value—the third preset distance—is set, and this value is officially designated as the second preset maximum distance threshold used by the large parking lot in subsequent verification steps. This ensures that the verification standard can fully match the physical reality of large parking lots and avoid misjudgment due to overly strict standards.

[0095] The third preset distance refers to a distance value (such as 200 meters) that the system pre-sets for large-scale parking lots. When the area of ​​the parking lot meets the large-scale condition, this value will be used as a tolerance standard to evaluate its location accuracy, i.e., the second preset maximum distance threshold.

[0096] Based on the above embodiments, as an optional embodiment, the initial position coordinates include initial longitude and initial latitude. Figure 1 Before step S106, there is also step S401, which will be explained in detail below.

[0097] S401. Calculate the distance between the initial position coordinates and the target effective position center coordinates in the following way. The target effective position center coordinates are any effective position center coordinates, including the target center longitude and the target center latitude: Distance = Earth radius × arccos[sin(initial latitude) × sin(target center latitude) + cos(initial latitude) × cos(target center latitude) × cos(initial longitude - target center longitude)].

[0098] Specifically, when calculating the distance between the initial position coordinates and the center coordinates of a specified target's effective position, a simple planar Euclidean distance formula is not used. Instead, a specific algorithm based on spherical trigonometry is invoked or executed. This requires four parameters: the latitude and longitude of the initial point, and the latitude and longitude of the target point. Applying the spherical cosine theorem, the latitude value is first processed using sine and cosine functions, and the longitude difference is processed using a cosine function. The results of these trigonometric functions are combined into an intermediate value. Then, the arccosine function is applied to this intermediate value to obtain the central angle (in radians) between the two points and the Earth's center. This central angle in radians is then multiplied by a constant representing the Earth's average radius (Earth's radius), thus converting the angle into an actual surface distance. This step follows the spherical distance calculation methods in geodesy for long distances (greater than several kilometers) or high-precision requirements, ensuring the scientific validity and high accuracy of the calculation results.

[0099] The target's effective location center coordinates refer to the center coordinates of any effective location from all filtered and reliable user activity hotspot areas (i.e., effective location clusters), used as a reference point for current distance calculation. This includes two components: target center latitude and target center longitude. Initial latitude and initial longitude are the latitude and longitude values ​​of the initial location coordinates, respectively, expressed in degrees. Latitude indicates the location's north-south direction (-90° to 90°), and longitude indicates the location's east-west direction (-180° to 180°). The target center latitude and target center longitude are the latitude and longitude values ​​of the target's effective location center coordinates, respectively, expressed in degrees. The Earth's radius is a constant used to convert angles to actual surface distances; it is a pre-defined value representing the Earth's approximate average radius, typically around 6371 kilometers or 6371000 meters.

[0100] For example, assuming: initial position coordinates P0: latitude = 31.2304° (lat1), longitude = 121.4737° (lng1), target effective position center coordinates P... i Latitude = 31.2300° i Longitude = 121.4730° (lng) i The Earth's radius R = 6371 kilometers (approximate value). Converting all angles (degrees) to radians (rad), lat1 ≈ 0.5449 rad, lng1 ≈ 2.1201 rad, lat i ≈0.5449rad, lng i≈2.1199rad. Substituting into the formula, initial longitude - target center longitude ≈ 2.1201 - 2.1199 = 0.0002rad. sin(initial latitude) × sin(target center latitude) + cos(initial latitude) × cos(target center latitude) × cos(initial longitude - target center longitude) ≈ 1. Since the cosine value should be between [-1, 1], numerical stability needs to be handled for values ​​very close to 1 (e.g., using the Haversine formula or ensuring that the arccos input is not greater than 1). Assuming the processed value is 0.999999, arccos(0.999999) ≈ 0.001414rad, then the distance ≈ 6371 * 0.001414 ≈ 9 kilometers. Obviously, the calculation is incorrect. This is just a schematic of the example process. The actual result of a small coordinate difference should be on the order of tens of meters. The above calculation is only to demonstrate the process. In actual programming, the numerically optimized Haversine formula will be used to avoid the numerical problems of arccos.

[0101] S107. When the confidence level of the first coordinate is less than the preset confidence threshold, the initial position coordinates are calibrated based on the center coordinates of each valid position to obtain the calibration position coordinates of the target parking lot in the current preset calibration period.

[0102] Specifically, after completing the quantitative evaluation of the initial location coordinates and calculating the first coordinate confidence level, this confidence level is immediately compared with a predefined minimum acceptable reliability standard, i.e., a pre-set confidence threshold. When the comparison result shows that the first coordinate confidence level is less than the pre-set confidence threshold, it is determined that the reliability of the initial location coordinates has failed to meet the requirements and needs to be corrected. Then, the calibration process is initiated. All valid location center coordinates identified in the previous steps are used as new location benchmarks and references. Through a pre-set fusion algorithm (such as weighted average, geometric median, etc., no specific restrictions are made here), the center point coordinates of these representative user activity hotspots are combined to calculate a new geographic coordinate that can better reflect the distribution center of these user behaviors. This newly calculated coordinate is formally defined as the calibration location coordinate determined for the target parking lot within the current pre-set calibration cycle. This coordinate is the final output of this calibration process and will be used to update the parking lot location data in the system.

[0103] The preset confidence threshold is a critical value (e.g., 80%) pre-set by the system to determine whether the confidence level of the first coordinate is acceptable. It serves as a decision benchmark to trigger the calibration operation. The calibration location coordinates are new coordinate points generated through calibration calculations to replace the initial location coordinates. They are specifically defined within the calibration time window (current preset calibration cycle) set for a particular parking lot (target parking lot) and are the final output of the calibration work for that cycle.

[0104] Based on the above embodiments, as an optional embodiment, the calibration of location coordinates includes calibrating latitude and calibrating longitude, for... Figure 1 The step S107 shown can be implemented through steps S1071-S1075, which will be explained in detail below.

[0105] S1071. The number of valid payment location coordinates for all valid location clusters is taken as the third quantity.

[0106] Specifically, during calibration, it is necessary to access the internal structure information of each valid location cluster, that is, the list of valid payment location coordinates contained in each cluster. It is necessary to traverse each valid location cluster, read the number of coordinate points contained in the cluster, and then sum up these counts read from each cluster to obtain a total value, which serves as the third quantity. This represents the total scale of all high-quality, highly relevant user location data points participating in the core calculation of this calibration, and is an important normalization basis for subsequent weighted calculations and other operations.

[0107] The third quantity refers to an integer value obtained through cumulative calculation. Its value is equal to the sum of the number of valid payment location coordinates within each of the valid location clusters. It represents the total number of high-quality location data points contributed by all reliable user activity hotspots during this calibration period. It is the normalization benchmark for subsequent weighted calculations and other operations, and is a non-negative integer.

[0108] S1072. Take the number of all valid payment location coordinates in the valid location cluster corresponding to the center coordinates of the target valid location as the fourth quantity. The center coordinates of the target valid location are any valid location center coordinates, including the longitude and latitude of the target center.

[0109] Specifically, for the specific reference point currently being processed, referred to as the target effective location center coordinates, its source is located—that is, the effective location cluster that generated the center coordinates. This specific cluster is accessed, and the total number of effective payment location coordinates contained in the cluster is accurately calculated. This calculated value, which characterizes the size of the data points in the cluster, is defined as the fourth quantity. This fourth quantity is a special statistic for the specific target effective location center coordinates and its corresponding cluster, reflecting the data richness of the user's hotspot area.

[0110] S1073. The ratio of the fourth quantity to the third quantity is used as the target weight of the effective position center coordinates of the target.

[0111] Specifically, the fourth and third values ​​are divided by the third value to obtain a value between 0 and 1. This result is the target weight of the center coordinates of the effective location of the target, which quantifies the relative importance and contribution of the specific user hotspot area in the entire effective user dataset.

[0112] The target weight is a real number, usually located in the open interval (0,1) (unless there are extreme cases). It represents the proportion of the center coordinate that should be included in the subsequent weighted calculation. The sum of all weights should be 1, which is naturally guaranteed by the calculation method (the number of each cluster divided by the total number).

[0113] S1074. The product of the target center latitude and the target weight is used as the calibration sub-latitude of the target's effective position center coordinates, and the product of the target center longitude and the target weight is used as the calibration sub-longitude of the target's effective position center coordinates.

[0114] Specifically, the two components of its geographic coordinates are read separately: the target center latitude (latitude value) and the target center longitude (longitude value). The target weight calculated for this coordinate is obtained. The target center latitude value is multiplied by the target weight, and the result is defined as the calibration sub-latitude of this coordinate. Similarly, the target center longitude value is multiplied by the target weight, and the result is defined as the calibration sub-longitude of this coordinate. These two calibration sub-values ​​represent the contribution components of this specific center point in the latitude and longitude dimensions, respectively, to the final synthesized coordinates after their importance (weight) adjustments. This step is performed independently for each valid location center coordinate.

[0115] The calibration sub-latitude refers to the intermediate result obtained by multiplying the target center latitude by the target weight. It is a numerical value with the same dimension as latitude (degrees), representing the contribution component of the target's effective position center coordinates in the latitude dimension, scaled according to its importance, to the final composite coordinates. Similarly, the calibration sub-longitude refers to the intermediate result obtained by multiplying the target center longitude by the target weight. It has the same dimension as longitude (degrees), representing the contribution component of the target's effective position center coordinates in the longitude dimension, scaled according to its importance, to the final composite coordinates.

[0116] S1075. The sum of all calibrator sub-latitudes is taken as the calibrator latitude, and the sum of all calibrator sub-longitudes is taken as the calibrator longitude.

[0117] Specifically, the system iterates through all center points, summing their calibration latitude values ​​one by one to obtain a total, which is directly defined as the calibration latitude. Similarly, it iterates through all center points, summing their calibration longitude values ​​one by one to obtain another total, which is directly defined as the calibration longitude. Through these two summation operations, the system aggregates the weighted geographic information of all center points into a single, comprehensive coordinate point. This coordinate point, defined by both calibration latitude and calibration longitude, is the core component of the final output of this calibration process—the calibration location coordinates.

[0118] The above embodiments bring the entire automated calibration process to a close, successfully transforming the original user payment behavior data stream into a high-quality, highly reliable parking lot location data update through a series of cleaning, analysis, evaluation, and calculations. This enables the location database of the smart parking system to have the ability to self-verify, self-correct, and continuously optimize.

[0119] Based on the above embodiments, as an optional embodiment, the physical attribute data also includes the initial geographical range data of the target parking lot, for... Figure 1 The parking lot location calibration method shown also includes steps S501-S504, which are described in detail below.

[0120] S501. Calculate the deviation distance between the calibration position coordinates and the initial position coordinates.

[0121] Specifically, the system reads the calibration coordinates (e.g., longitude C1, latitude C2) representing the center of the calibrated location and the initial coordinates (e.g., longitude O1, latitude O2) representing the center of the original location from the storage system. A preset distance calculation model (e.g., a spherical distance calculation function based on the Haversine formula) is invoked, using these two sets of coordinates as input parameters. Geometric operations are performed, and the final output is a scalar value representing the straight-line distance between the two points, i.e., the deviation distance. This calculation result is a physical quantity with a definite unit of length (usually meters), objectively quantifying the spatial displacement of the parking lot center position caused by this calibration operation.

[0122] The deviation distance refers to the shortest path length on the Earth's sphere between the calibration position coordinates and the initial position coordinates, calculated using a specific geometric algorithm. It is used to accurately measure the degree of separation between the two in physical space.

[0123] S502. When the deviation distance is greater than the preset coordinate deviation threshold, the preset API interface is called to query and obtain the preset geographical range parameters of the target parking lot based on the calibration location coordinates.

[0124] Specifically, the deviation distance calculated in step S501 is compared in real time with a preset coordinate deviation threshold (e.g., 100 meters) pre-configured within the system. If the deviation distance is determined to be greater than this threshold, it means that the calibration has caused a substantial (rather than minor fluctuation) shift in the center position of the parking lot. At this point, the system automatically activates a preset API interface connected to an external map service provider (such as Gaode Maps or Baidu Maps) based on a predefined protocol and address. Using the determined calibration location coordinates as the core query parameter, the system initiates a structured geographic information query request to the external service through this interface. It receives and parses the response data packet returned by the external service, extracting standardized preset geographic range parameters related to the target parking lot, such as "boundary polygon coordinate string," "outer rectangle diagonal coordinates," or "suggested coverage radius centered on the coordinate point," etc., and temporarily stores these parameters in local memory for use in the next calculation. The entire process is a condition-triggered, automated online data acquisition chain.

[0125] Among them, the preset coordinate deviation threshold refers to a distance critical value pre-set by the system administrator or algorithm based on factors such as business experience, accuracy requirements, and cost control. It is a scalar with a length unit (usually meters) and is used as a decision-making benchmark to determine whether the location change is "significant" or "requires further processing." The preset API interface refers to an application programming interface pre-integrated and configured in the system to enable programmatic data exchange with specific external services (such as commercial map platforms). It specifies the request format (such as URL, parameters), protocol (such as HTTP / HTTPS), and response data structure (such as JSON, XML), serving as a standardized communication bridge between the system and external services. The preset geographic range parameter refers to the standardized data fields or datasets retrieved and returned by the external map service from its authoritative geographic information database after receiving a coordinate point query request. These parameters describe the spatial coverage of typical geographic entities (such as parking lots) near the coordinate point. These parameters are not generated by on-site measurements but are modeled information pre-defined and stored by the map service provider based on its data source.

[0126] For example, the system calculates that the calibration location coordinates of a parking lot deviate from the initial record by 150 meters. The preset coordinate deviation threshold is 100 meters. Since 150 meters > 100 meters, the trigger condition is met. The system then calls the configured Amap reverse geocoding or location search API, sending the new calibration coordinates (e.g., 116.4080°E, 39.9038°N) as the location parameter. The JSON data returned by the API contains a parking object with a polygon field whose value is a string of latitude and longitude coordinate pairs, such as "116.4078,39.9045;116.4085,39.9045;116.4085,39.9032;116.4078,39.9032". This string is the obtained "preset geographic range parameter," describing the quadrilateral boundary of the parking lot in the map service.

[0127] S503. Based on the calibration location coordinates and preset geographical range parameters, calculate the estimated geographical range of the target parking lot.

[0128] Specifically, the preset geographic extent parameters obtained from the map API (such as a circular extent description centered at an origin with radius R, or a string of polygon vertex coordinates) are loaded into memory. These parameters are parsed to identify the geometry of the extent they describe (e.g., whether it's a "radius model" or a "polygon model"). Then, the reference center point of this geometry (for polygons, this could be its centroid or original center) is compared with the input of the current step—the calibration location coordinates. If they do not coincide, a translation vector (Δlongitude, Δlatitude) from the original center point of the preset extent to the current calibration location coordinates is uniformly applied to all geometric vertices or keypoints in the preset extent description. For example, for a polygon extent, each vertex coordinate is traversed, and Δlongitude and Δlatitude are added to the longitude and latitude of each vertex respectively, generating a new set of vertex coordinates. The polygon defined by this new set of vertex coordinates is the calculated presumed geographic extent. This extent retains the original shape and size obtained from the map API, but its spatial location has been anchored to the new center point obtained in this calibration.

[0129] The estimated geographic range refers to the new geographic range geometric description that is consistent with the calibrated position, obtained by translating the spatial position of the geometric shape described by the preset geographic range parameters to the center (or corresponding anchor point) of the calibration position coordinates.

[0130] Through the above embodiments, it is ensured that the generated geographic range is consistent in shape accuracy (inherited from authoritative map data) and location accuracy (derived from calibration based on real user data). This avoids spatial logic errors that may occur due to mismatch between the range and the center point caused by directly using uncorrected API range data. It provides internally consistent and highly reliable spatial foundation data for business systems (such as parking navigation and electronic fence determination), greatly enhancing the user experience and operational accuracy of location-based services.

[0131] S504. Update the physical attribute data of the target parking lot with the estimated geographical range to cover the initial geographical range data.

[0132] Specifically, the estimated geographic extent data structure calculated in step S503 is first retrieved from memory or temporary storage. This data structure is typically a canonical geometric object, such as a polygon defined by a sequence of vertex coordinates. Based on the unique identifier (e.g., ID) of the target parking lot being processed, the corresponding data record is located in its physical attribute database. This record contains a field named "Initial Geographic Extent Data," which stores the geographic extent description used by the parking lot before the start of this calibration process. This may be an outdated or inaccurate polygon coordinates, rectangular area, or other form of geometric data. The original content held by the "Initial Geographic Extent Data" field in the database record is completely replaced or overwritten with a new geometric data object representing the estimated geographic extent. After this operation, the database stores the latest geographic extent information that precisely matches the latest calibration location coordinates in space and whose shape and dimensions originate from an authoritative map service. The entire process ensures the atomicity of data updates to guarantee that the business system obtains a consistent and up-to-date state when reading data.

[0133] The initial geographic range data refers to the original geometric data stored in the physical attribute database before the calibration process begins, which describes the geographic coverage of the target parking lot. This data may be outdated or mismatched with the calibrated center location.

[0134] Through the above embodiments, the high-quality data results generated by dynamic calibration and external verification are solidified into a reliable and persistent real data source for business systems. By covering the initial geographical range data with the estimated geographical range, the problem of outdated geographical range information caused by changes in the physical boundaries of parking lots (such as expansion or entrance renovation) or inaccurate original data entry is fundamentally solved. This ensures that downstream services that rely on this range data (such as parking space status mapping, entry and exit electronic fence judgment, and precise navigation guidance) use the most accurate spatial information that is logically consistent with the central location. This directly improves the service accuracy, user satisfaction, and operational automation level of the entire parking management system, and constructs a dynamic, accurate, and closed-loop spatial data maintenance system.

[0135] The parking lot location calibration system in the embodiments of this invention is described below from the perspective of hardware processing. Please refer to [link / reference needed]. Figure 3 , Figure 3 This is a schematic diagram of the structure of a parking lot location calibration system provided in an embodiment of this application.

[0136] It should be noted that, Figure 3 The structure of the parking lot location calibration system shown is merely an example and should not impose any limitations on the functionality and scope of application of the embodiments of the present invention.

[0137] like Figure 3 As shown, the parking lot location calibration system includes a central processing unit 601, which can perform various appropriate actions and processes according to a program stored in a read-only memory 602 or a program loaded from a storage section 608 into a random access memory 603, such as performing the methods described in the above embodiments. The random access memory 603 also stores various programs and data required for system operation. The central processing unit 601, the read-only memory 602, and the random access memory 603 are interconnected via a bus 604. An input / output interface 605 is also connected to the bus 604.

[0138] The following components are connected to the input / output interface 605: an input section 606 including audio input devices, push-button switches, etc.; an output section 607 including a liquid crystal display (LCD) and audio output devices, indicator lights, etc.; a storage section 608 including a hard disk, etc.; and a communication section 609 including a network interface card such as a LAN (Local Area Network) card, modem, etc. The communication section 609 performs communication processing via a network such as the Internet. A drive 610 is also connected to the input / output interface 605 as needed. A removable medium 611, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., is installed on the drive 610 as needed so that computer programs read from it can be installed into the storage section 608 as needed.

[0139] In particular, according to embodiments of the present invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of the present invention include a computer program product comprising a computer program carried on a computer-readable medium, the computer program including a computer program for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via communication section 609, and / or installed from removable medium 611. When the computer program is executed by central processing unit 601, it performs the various functions defined in the present invention. It should be noted that specific examples of computer-readable storage media may include, but are not limited to: electrical connections having one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), flash memory, optical fiber, portable compact disc read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this invention, a computer-readable storage medium can be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0140] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present invention. Each block in a flowchart or block diagram may represent a module, program segment, or portion of code, which contains one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those shown in the drawings.

[0141] Specifically, the parking lot location calibration system of this embodiment includes a processor and a memory. The memory stores a computer program, and when the computer program is executed by the processor, it implements the parking lot location calibration method provided in the above embodiment.

[0142] In another aspect, the present invention also provides a computer-readable storage medium, which may be included in the parking lot location calibration system described in the above embodiments; or it may exist independently and not assembled into the parking lot location calibration system. The storage medium carries one or more computer programs that, when executed by a processor of the parking lot location calibration system, cause the parking lot location calibration system to implement the parking lot location calibration method provided in the above embodiments.

[0143] The above description is merely an exemplary embodiment of this disclosure and should not be construed as limiting the scope of this disclosure. Any equivalent changes and modifications made in accordance with the teachings of this disclosure shall still fall within the scope of this disclosure. This application is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not described in this disclosure. The specification and embodiments are considered exemplary only, and the scope and spirit of this disclosure are defined by the claims.

Claims

1. A parking lot location calibration method, characterized in that, include: At the end of the current preset calibration period, acquire the parking payment dataset, the initial location coordinates of the target parking lot, and the physical attribute data when the user makes a payment within the preset coordinate range during the preset data collection period; Extract the set of valid payment location coordinates from the parking payment dataset, wherein the set of valid payment location coordinates includes multiple valid payment location coordinates; Based on the physical attribute data, multiple location clusters are determined in the set of valid payment location coordinates; Calculate the average distance between all valid payment location coordinates of the target location cluster and the center coordinates corresponding to the target location cluster, wherein the target location cluster is any of the location clusters; The target location clusters whose average distance is less than or equal to the first preset maximum distance threshold are taken as effective location clusters, the number of effective location clusters is taken as the first number, and the center coordinates of the effective location clusters are taken as the center coordinates of the effective locations. The number of effective position center coordinates whose distance from the initial position coordinates to each effective position center coordinate is less than a second preset maximum distance threshold is counted as the second number. The ratio of the second number to the first number is used as the first coordinate confidence of the initial position coordinates. The second preset maximum distance threshold is greater than the first preset maximum distance threshold. When the confidence level of the first coordinate is less than the preset confidence threshold, the initial position coordinates are calibrated based on the center coordinates of each valid position to obtain the calibrated position coordinates of the target parking lot in the current preset calibration period.

2. The method according to claim 1, characterized in that, Before acquiring the parking payment dataset, the initial location coordinates of the target parking lot, and the physical attribute data when the user makes payments within a preset coordinate range during the preset data collection period at the end of the current preset calibration period, the method further includes: Obtain the second coordinate confidence level of the calibration position coordinates of the previous preset calibration cycle, and use the calibration position coordinates of the previous preset calibration cycle as the initial position coordinates of the target parking lot; The previous preset calibration cycle is adjusted based on the second coordinate confidence level to obtain the current preset calibration cycle; Obtain the average daily traffic flow of the target parking lot during the current preset calibration period; When the average daily traffic flow is less than the preset traffic flow threshold, the first data collection period will be used as the preset data collection period. When the average daily traffic flow is greater than or equal to the preset traffic flow threshold, the second data collection period is used as the preset data collection period, and the first data collection period is greater than the second data collection period.

3. The method according to claim 1, characterized in that, The parking payment dataset includes multiple payment data entries, each of which includes at least payment location coordinates and a positioning accuracy identifier. Extracting the set of valid payment location coordinates from the parking payment dataset includes: Based on the positioning accuracy identifier, the positioning error of each payment data is determined; Payment data whose payment location coordinates are within the preset coordinate range and whose positioning error is less than or equal to the preset error threshold are considered valid payment data. The payment location coordinates corresponding to each of the valid payment data are used as the valid payment location coordinates; By combining the coordinates of each valid payment location, the set of valid payment location coordinates is obtained; When the ratio of the number of valid payment data to the number of payment data is less than or equal to a preset ratio threshold, the preset data collection period is adjusted so that the ratio is greater than the preset ratio threshold; When the number of valid payment location coordinates is less than a preset threshold, the preset data collection period is adjusted so that the number of valid payment location coordinates is greater than or equal to the preset threshold.

4. The method according to claim 1, characterized in that, The physical attribute data also includes the area of ​​the target parking lot. Before counting the number of effective location center coordinates whose distance from the initial location coordinates to each effective location center coordinate is less than a second preset maximum distance threshold, the method further includes: When the area occupied is less than or equal to the first preset area threshold, the first preset distance is used as the second preset maximum distance threshold; When the area occupied is greater than the first preset area threshold and less than or equal to the second preset area threshold, the second preset distance is used as the second preset maximum distance threshold, and the second preset distance is greater than the first preset distance. When the area occupied is greater than the second preset area threshold, the third preset distance is used as the second preset maximum distance threshold, and the third preset distance is greater than the second preset distance.

5. The method according to claim 1, characterized in that, The initial position coordinates include initial longitude and initial latitude. Before counting the number of effective position center coordinates whose distance from the initial position coordinates to each effective position center coordinate is less than a second preset maximum distance threshold, the method further includes: The distance between the initial position coordinates and the target effective position center coordinates is calculated as follows: the target effective position center coordinates can be any of the effective position center coordinates, and the target effective position center coordinates include the target center longitude and the target center latitude: The distance = Earth radius × arccos[sin(initial latitude) × sin(target center latitude) + cos(initial latitude) × cos(target center latitude) × cos(initial longitude - target center longitude)].

6. The method according to claim 1, characterized in that, The calibration location coordinates include calibration latitude and calibration longitude. When the confidence level of the first coordinate is less than a preset confidence threshold, the initial location coordinates are calibrated based on the center coordinates of each valid location to obtain the calibration location coordinates of the target parking lot, including: The number of valid payment location coordinates for all the valid location clusters is taken as the third quantity; The number of all valid payment location coordinates in the valid location cluster corresponding to the center coordinates of the target valid location is taken as the fourth quantity. The center coordinates of the target valid location are any valid location center coordinates, and the center coordinates of the target valid location include the longitude and latitude of the target center. The ratio of the fourth quantity to the third quantity is used as the target weight of the effective location center coordinates of the target; The product of the target center latitude and the target weight is used as the calibration sub-latitude of the target effective position center coordinates, and the product of the target center longitude and the target weight is used as the calibration sub-longitude of the target effective position center coordinates. The sum of all the said calibrating sub-latitudes is taken as the calibrated latitude, and the sum of all the said calibrating sub-longitudes is taken as the calibrated longitude.

7. The method according to claim 1, characterized in that, The physical attribute data also includes the initial geographical extent data of the target parking lot, and the method further includes: Calculate the deviation distance between the calibration position coordinates and the initial position coordinates; When the deviation distance is greater than the preset coordinate deviation threshold, the preset API interface is called to query and obtain the preset geographical range parameters of the target parking lot based on the calibration location coordinates; Based on the calibration location coordinates and the preset geographical range parameters, the estimated geographical range of the target parking lot is calculated; The estimated geographic range is used to update the physical attribute data of the target parking lot, thereby overwriting the initial geographic range data.

8. A parking lot location calibration system, characterized in that, The parking lot location calibration system includes: one or more processors and a memory; the memory is coupled to the one or more processors, the memory is used to store computer program code, the computer program code including computer instructions, and the one or more processors call the computer instructions to cause the parking lot location calibration system to perform the method as described in any one of claims 1-7.

9. A computer-readable storage medium comprising instructions, characterized in that, When the instruction is executed on the parking lot location calibration system, the parking lot location calibration system performs the method as described in any one of claims 1-7.

10. A computer program product, characterized in that, When the computer program product is run on the parking lot location calibration system, the parking lot location calibration system performs the method as described in any one of claims 1-7.

Images