Roadside parking data quality inspection method and system based on inspection vehicle
By collecting data on-site using patrol vehicles and processing it with data quality inspection algorithms, the problem of low-quality off-street parking data has been solved, achieving data accuracy and comprehensiveness, and optimizing parking management and user experience.
Patent Information
- Application Number
- CN202511736832.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-11-25
- Publication Date
- 2026-02-03
AI Technical Summary
Existing technologies make it difficult to accurately and completely identify off-street parking data, resulting in low-quality parking data and affecting the efficiency of the city's unified parking network platform.
By using a data quality inspection method based on patrol vehicles, we obtain the registration information and business data of off-street parking lots in the city. We then perform preprocessing and quality inspection using a data quality inspection algorithm module, collect vehicle data on-site, and transmit it back to the city's unified parking network platform.
It improved the accuracy and comprehensiveness of off-street parking data identification, optimized parking management, and enhanced the intelligent scheduling capabilities and user experience of the city's unified parking network platform.
Smart Images
Figure CN121459629A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present application relates to the technical field of intelligent parking management, and particularly relates to an off-street parking data quality inspection method and system based on an inspection vehicle. BACKGROUND
[0002] In order to manage the increasing parking lots and solve the problem of urban parking difficulty, the prior art integrates scattered parking resources such as business and community, uses Internet of Things, big data and cloud computing technology, realizes real-time sharing, unified scheduling and intelligent management of parking space information, and establishes a city parking network platform.
[0003] However, low parking data quality seriously hinders the effectiveness of the city parking network platform. First, inaccurate data leads users to navigate to parking difficulty areas, increasing the search time and reducing the platform's credibility. Second, incomplete data, such as the lack of access to some parking lots and missing information, causes resource loss, affecting global scheduling and resource utilization. Third, the accumulation of zombie vehicles during the operation of the parking lot negatively affects the city parking network platform's global perception and scheduling capabilities. In addition, parking operators intentionally omit some parking data to avoid supervision, which also seriously affects data quality.
[0004] The prior art has the technical problem of low parking data quality due to the difficulty in accurately and completely identifying off-street parking data. SUMMARY
[0005] The purpose of the present application is to provide an off-street parking data quality inspection method and system based on an inspection vehicle to solve the technical problem of low parking data quality due to the difficulty in accurately and completely identifying off-street parking data in the prior art.
[0006] In view of the above problems, the present application provides an off-street parking data quality inspection method and system based on an inspection vehicle.
[0007] The first aspect of the present application provides an off-street parking data quality inspection method based on an inspection vehicle, which comprises: calling a city parking network platform, obtaining the record information and business data of the city off-street parking lot through the parking lot record or parking facility registration business; obtaining inspection vehicle resource information, combining the business data to determine the parking lot entry and exit data trough period of the city off-street parking lot, and determining the inspection plan, wherein the inspection plan includes an inspection scheme and an inspection path; the inspection vehicle collects vehicle data according to the inspection path and the inspection scheme to obtain a vehicle data set; the vehicle data set is preprocessed and data quality inspected by a data quality inspection algorithm module in combination with the business data to obtain an inspection result, and the inspection result is returned to the city parking network platform.
[0008] Optionally, the record information includes parking lot name, parking lot code, longitude and latitude, number of parking spaces, internal map of the parking lot; the business data includes vehicle entry record, vehicle exit record, real-time number of vehicles present, real-time number of remaining parking spaces, and the like.
[0009] Optionally, the preprocessing includes deduplication of repeated license plates in the same parking lot and elimination of license plates that do not meet the rules.
[0010] Optionally, the vehicle data set is preprocessed by a data quality inspection algorithm module to obtain a preprocessed vehicle data set; the preprocessed vehicle data set is subjected to remaining parking space number quality inspection in combination with the business data to obtain a remaining parking space number quality inspection result; the preprocessed vehicle data set is subjected to parking record number quality inspection in combination with the business data to obtain a parking record number quality inspection result; the preprocessed vehicle data set is subjected to zombie vehicle quality inspection in combination with the business data to obtain a zombie vehicle quality inspection result; and the remaining parking space number quality inspection result, the parking record number quality inspection result, and the zombie vehicle quality inspection result are returned to the city parking one network platform as the quality inspection result.
[0011] Optionally, the number of vehicles in the parking lot collected during the inspection in the preprocessed vehicle data set is compared with the business data to execute a remaining parking space number quality inspection algorithm; if |RB QC -RB BD |>ρ*V*B total , the parking lot that fails the remaining parking space number quality inspection is taken as the remaining parking space number quality inspection result; wherein, RB QC is the remaining parking space number calculated based on the preprocessed vehicle data set during the inspection, B total is the total number of parking spaces in the parking lot, RB QC =B total -C CAR , C CAR is the number of vehicles in the parking lot collected during the inspection, R BBD is the remaining parking space number in the business data during the inspection, V is the percentage of error allowed by the system, and p is the correction coefficient, and p>1.
[0012] Optionally, the number of vehicles in the parking lot is calculated based on the business data to obtain a regular number of vehicles in the parking lot; the regular number of vehicles in the parking lot is compared with the number of vehicles in the parking lot collected during the inspection in the preprocessed vehicle data set to execute a parking record number quality inspection algorithm; if |C CAR -C B-CAR |>ρ*V*B total , the parking lot that fails the parking record number quality inspection is taken as the parking record number quality inspection result; wherein, C CAR is the number of vehicles in the parking lot collected during the inspection, C B-CAR is the regular number of vehicles in the parking lot.
[0013] Optionally, the off-site license plate statistics are performed on the business data to obtain off-site license plate details; the inspection license plate details collected by the inspection vehicle in the parking lot are obtained from the pre-processed vehicle data set; the off-site license plate details and the inspection license plate details are compared to obtain a list of pending zombie vehicles; a zombie vehicle inspection algorithm is executed based on the list of pending zombie vehicles to calculate the probability that each pending zombie vehicle is a zombie vehicle; wherein the zombie vehicle probability calculation formula is: P = lg (T) * n^(1 / 2) * 0.5; P is the probability that the pending zombie vehicle is a zombie vehicle, T is the length of time the vehicle has been parked in the parking lot, in days; n is the number of times the vehicle has been added to the list of pending zombie vehicles since the last record of the vehicle entering the business system; if the probability that the pending zombie vehicle is a zombie vehicle is greater than the determination threshold, it is added to the list of suspected zombie vehicles, and the list of suspected zombie vehicles is used as the zombie vehicle inspection result.
[0014] Optionally, when there is a parking lot that does not pass the remaining parking space number inspection in the inspection result, the city parking one map platform synchronizes the inaccurate remaining parking space number alarm and the remaining parking space number adjustment strategy to the third party parking system; when there is a parking lot that does not pass the parking record number inspection and the number of vehicles in the business data is less than the number of vehicles collected by the inspection vehicle in the inspection result, the data inspection algorithm module sends a business leakage alarm to the city parking one map platform; when there is a zombie vehicle that does not pass the zombie vehicle inspection in the inspection result, the data inspection algorithm module sends a suspected zombie vehicle alarm and a suspected zombie vehicle list to the city parking one map platform, and the city parking one map platform synchronizes the suspected zombie vehicle list to the third party parking system.
[0015] In a second aspect of the present application, a system for checking the quality of off-street parking data based on an inspection vehicle is provided, the system comprising: a data acquisition component configured to call a city parking one map platform to obtain the record information and business data of the city off-street parking lot through a parking lot record or parking facility registration business; a patrol plan determination component configured to obtain inspection vehicle resource information, determine the parking lot entry and exit data trough of the city off-street parking lot in combination with the business data, and determine an inspection plan, wherein the inspection plan comprises an inspection scheme and an inspection path; a vehicle data acquisition component configured to perform vehicle data acquisition by the inspection vehicle according to the inspection path and the inspection scheme to obtain a vehicle data set; and a quality inspection result obtaining component configured to preprocess and perform data quality inspection on the vehicle data set by a data inspection algorithm module in combination with the business data, obtain a quality inspection result, and return the quality inspection result to the city parking one map platform.
[0016] The one or more technical solutions provided in the present application have at least the following technical effects or advantages:
[0017] The method provided by the embodiment of the application obtains the record information and the business data of the urban off-street parking lot through the parking lot record or parking facility registration business by calling the urban parking one-stop network platform, obtains the inspection vehicle resource information, determines the parking lot access data valley period of the urban off-street parking lot in combination with the business data, and determines an inspection plan, wherein the inspection plan comprises an inspection scheme and an inspection path; the inspection vehicle collects vehicle data according to the inspection path and the inspection scheme, obtains a vehicle data set; the vehicle data set is pretreated and data quality inspected by a data quality inspection algorithm module in combination with the business data, an inspection result is obtained, and the inspection result is returned to the urban parking one-stop network platform. Vehicle information is collected on site by the inspection vehicle, and multi-dimensional data quality inspection is performed in combination with the business data, so that the technical effects of improving the accuracy and comprehensiveness of off-street parking data identification are achieved.
[0018] The above description is only a summary of the technical solutions of the application. In order to enable one skilled in the art to better understand the technical means of the application, the application can be implemented according to the content of the specification, and in order to enable the above and other purposes, characteristics and advantages of the application to be more obvious and easy to understand, the following specific embodiments of the application are described. It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the application, nor is it intended to limit the scope of the application. Other features of the application will become apparent from the following description. BRIEF DESCRIPTION OF DRAWINGS
[0019] In order to more clearly illustrate the technical solutions in the application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or the prior art description. Obviously, the drawings in the following description are only exemplary, and other drawings can be obtained by those skilled in the art without creating labor on the basis of the provided drawings.
[0020] Figure 1 A flowchart of a method for checking off-street parking data quality based on an inspection vehicle provided by the application.
[0021] Figure 2 A structure diagram of a system for checking off-street parking data quality based on an inspection vehicle provided by the application.
[0022] Explanation of reference signs: data acquisition component 11, inspection plan determination component 12, vehicle data acquisition component 13, and inspection result obtaining component 14. DETAILED DESCRIPTION
[0023] This application provides a method and system for quality inspection of off-street parking data based on patrol vehicles, addressing the technical problem of low-quality parking data due to the difficulty in accurately and completely identifying off-street parking data in existing technologies. By collecting vehicle information on-site using patrol vehicles and combining it with business data for multi-dimensional data quality inspection, the method achieves the technical effect of improving the accuracy and comprehensiveness of off-street parking data identification.
[0024] The technical solutions of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. It should be understood that the present invention is not limited to the exemplary embodiments described herein. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention. It should also be noted that, for ease of description, only the parts related to the present invention are shown in the accompanying drawings, not all of them.
[0025] Example 1, as Figure 1 As shown, this application provides a method for quality inspection of off-street parking data based on patrol vehicles. The method includes:
[0026] By calling the city's unified parking network platform, and through the parking lot registration or parking facility registration business, the registration information and business data of off-street parking lots in the city can be obtained.
[0027] Furthermore, the registration information includes the parking lot name, parking lot code, latitude and longitude, number of parking spaces, and internal map of the parking lot; the business data includes vehicle entry records, vehicle exit records, real-time number of vehicles on site, and real-time remaining number of parking spaces.
[0028] Specifically, data access is achieved by calling the city's unified parking network platform. This platform integrates all parking resources within the city, including public parking lots, off-street parking lots, commercial parking lots, and residential parking lots, forming a unified intelligent parking management platform capable of centrally managing the registration and operational information of various parking lots. Based on the parking lot registration or parking facility registration services provided by the city's unified parking network platform, registration information for all off-street parking lots within the city is obtained. This registration information includes parking lot name, parking lot code, latitude and longitude, number of parking spaces, and internal map of the parking lot. Simultaneously, through the parking lot registration or parking facility registration services, business data for off-street parking lots in the city is obtained. This business data includes vehicle entry records, vehicle exit records, real-time number of vehicles present, real-time remaining number of parking spaces, and total number of parking spaces in the parking lot.
[0029] By obtaining accurate filing information, the parking lot can be accurately positioned and identified, and by obtaining complete business information, the reliability and effectiveness of the entire off-street parking data quality inspection method can be ensured.
[0030] Obtain inspection vehicle resource information, and determine a parking lot access data trough period of a city off-street parking lot in combination with the business data to determine an inspection plan, wherein the inspection plan comprises an inspection scheme and an inspection path.
[0031] Specifically, the inspection vehicle resource information is obtained through a dispatch management system or an inspection vehicle management platform, and the inspection vehicle resource information comprises the number of available inspection vehicles, the endurance capability, and the vehicle dispatchable time period, etc., which are used to evaluate the inspection capability and develop a reasonable inspection plan. In combination with the vehicle access record, the parking lot access record, the real-time number of vehicles present, the real-time number of remaining parking spaces, etc. in the business data, the data is analyzed and counted to identify the parking lot access data trough period of each off-street parking lot. The trough period refers to a time period in which the vehicle flow in the parking lot is less and the number of access vehicles is the lowest, so as to minimize the interference with the normal parking vehicles and ensure the accuracy of the inspection collected data.
[0032] After the parking lot access data trough period is determined, the inspection plan of the inspection vehicle is generated according to the inspection vehicle resource information. The inspection plan comprises an inspection scheme and an inspection path. The inspection scheme is used to clearly define the range of the parking lot that each inspection vehicle is responsible for inspecting in a specific time period. While determining the inspection scheme, the load balancing and the optimal scheduling strategy of the vehicle resources are considered. The inspection path refers to the inspection path of the inspection vehicle inside each parking lot. The goal of the inspection path is to traverse each internal road of the parking lot under the premise of walking as few repeated routes as possible, to collect the vehicle data in every corner of the parking lot under the premise of collecting as few repeated data as possible, and to achieve full coverage data collection. For example, a graph theory algorithm such as the shortest path is used in combination with the internal map of each off-street parking lot to automatically generate a planned route, so as to achieve accurate path planning and maximize the inspection efficiency.
[0033] The inspection vehicle collects vehicle data according to the inspection path and the inspection scheme to obtain a vehicle data set.
[0034] Specifically, the inspection calculation is issued to the corresponding inspection vehicle, the inspection vehicle goes to the target parking lot according to the inspection plan, and traverses the internal roads of the off-street parking lot according to the inspection path and the inspection scheme. The vehicle data set is obtained by collecting all the vehicle data in the off-street parking lot by using the vehicle-mounted equipment such as a high-definition camera, an automatic number plate recognition (ANPR) system, and a global positioning system (GPS). The vehicle data set comprises the license plate number, the parking position, and the parking state, and the data collection time is recorded in combination with the time stamp.
[0035] The vehicle data set is preprocessed and data quality inspection is performed by a data quality inspection algorithm module in combination with business data, a quality inspection result is obtained, and the quality inspection result is returned to the urban parking one-net platform.
[0036] Further, the preprocessing includes deduplication of repeated license plates in the same parking lot and elimination of license plates that do not meet the rules.
[0037] Specifically, the quality inspection data is preprocessed by the data quality inspection algorithm module, the preprocessing includes deduplication of repeated license plates in the same parking lot, and elimination of license plates that do not meet the rules. After completing the preprocessing, the data quality inspection algorithm module performs data quality inspection on the preprocessed vehicle data in combination with business data, the data quality inspection includes remaining parking space number inspection, parking record number inspection, and zombie car inspection, the multi-dimensional data quality inspection results are integrated to obtain the final quality inspection result, the quality inspection result is returned through the data interface of the urban parking one-net platform, high-precision, standardized quality inspection and intelligent management of off-street parking data are achieved, and more accurate data and better user experience are provided for urban parking users.
[0038] Further, the vehicle data set is preprocessed and data quality inspection is performed by a data quality inspection algorithm module in combination with business data, a quality inspection result is obtained, and the quality inspection result is returned to the urban parking one-net platform, including: the vehicle data set is preprocessed by a data quality inspection algorithm module to obtain a preprocessed vehicle data set; the preprocessed vehicle data set is inspected for remaining parking space number in combination with business data to obtain a remaining parking space number inspection result; the preprocessed vehicle data set is inspected for parking record number in combination with business data to obtain a parking record number inspection result; the preprocessed vehicle data set is inspected for zombie car in combination with business data to obtain a zombie car inspection result; and the remaining parking space number inspection result, the parking record number inspection result, and the zombie car inspection result are returned as the quality inspection result to the urban parking one-net platform.
[0039] Specifically, the vehicle data set is preprocessed by a data quality inspection algorithm module, repeated license plates in the same parking lot are deduplicated, and license plates that do not meet the rules are eliminated to obtain a preprocessed vehicle data set. Taking the underground parking lot P m of a large supermarket in City C as an example, the parking lot has a total of 2000 parking spaces, the vehicle data in the parking lot collected by the inspection vehicle is 1750, 92 repeated license plate data are eliminated, 28 license plate data that do not meet the standard are eliminated, and finally the number of vehicles in the parking lot collected during the inspection period C car = 1630.
[0040] In combination with the business data, the remaining parking space number quality inspection is performed on the preprocessed vehicle data set, that is, by comparing the actual number of vehicles in the parking lot collected by the inspection vehicle with the remaining parking space number recorded in the business data, the remaining parking space number quality inspection result is obtained. At the same time, by comparing the number of vehicles collected during the inspection with the in-lot vehicle data counted in the business data, the parking record number quality inspection result is obtained. For the long-term non-out-of-lot vehicle, the zombie vehicle quality inspection is performed by comparing the non-out-of-lot license plate in the business data with the vehicle information collected by the inspection vehicle, and the zombie vehicle quality inspection result is obtained.
[0041] The remaining parking space number quality inspection result, the parking record number quality inspection result and the zombie vehicle quality inspection result are integrated into a unified quality inspection result, and are returned to the city parking network platform through a data interface, so that the synchronization of the quality inspection result and the support for strategy adjustment are realized.
[0042] By cross-comparing the real-time collected data of the inspection vehicle and the business data, not only the comprehensive and accurate inspection of the off-street parking data quality is realized, but also the standardized and returnable quality inspection result is formed. The combination of the data quality improvement of individual parking lot and the overall parking business data supervision optimizes the parking management, realizes the efficient cooperation of the city parking network, and improves the platform intelligent scheduling capability and user experience.
[0043] Further, the remaining parking space number quality inspection is performed on the preprocessed vehicle data set in combination with the business data, and a remaining parking space number quality inspection result is obtained, including: obtaining the number of in-lot vehicles collected during the inspection in the preprocessed vehicle data set, comparing with the business data, and executing a remaining parking space number quality inspection algorithm; if |RB QC -RB BD |>ρ*V*B total , the parking lot that does not pass the remaining parking space number quality inspection is taken as the remaining parking space number quality inspection result; wherein, RB QC is the remaining parking space number during the inspection calculated based on the preprocessed vehicle data set, B total is the total number of parking spaces of the parking lot, RB QC =B total -C CAR , C CAR is the number of in-lot vehicles collected during the inspection, RB BD is the remaining parking space number in the business data during the inspection, V is the percentage of error allowed by the system, and p is the correction coefficient, and p≥1.
[0044] Specifically, the number of in-lot vehicles actually collected during the inspection in the preprocessed vehicle data set is obtained, and the real-time remaining parking space number and the total number of parking spaces of the parking lot in the business data are compared, and a remaining parking space number quality inspection algorithm is executed. Automatic judgment is performed by using the remaining parking space number quality inspection algorithm, and the judgment condition is |RB QC -RB BD |>ρ*V*B total, when the parking lot |RB QC -RB BD |ρ*V*B total , the parking lot fails the remaining parking space number quality inspection, and the remaining parking space number quality inspection result is obtained. Wherein, RB QC is the remaining parking space number during the inspection calculated based on the pre-processed vehicle data set, B total is the total number of parking spaces of the parking lot, RB QC =B total -C CAR , C CAR is the number of vehicles in the parking lot during the inspection, RB BD is the remaining parking space number in the business data during the inspection, V is the percentage of the error allowed by the system, which is optionally 10%, and ρ is the correction coefficient, and ρ≥1. The determination condition is relaxed moderately for the false negatives, false positives, and changes of vehicles in the parking lot during the inspection, for example, ρ=1.1. BD is the remaining parking space number in the business data during the inspection, and the median of the multiple reported values is taken as the remaining parking space number.
[0045] For example, during the inspection, the city parking network platform receives 3 times of the remaining parking space data reported by the target parking lot, which are 819, 845, and 798 respectively. The median is taken, and the remaining parking space number RB BD of the target parking lot during the inspection is 819. The remaining parking space number RB QC calculated based on the data collected by the inspection vehicle is 2000-1630=370. The absolute value of the difference between the two is |RB QC -RB BD |=449. Compared with the determination threshold ρ*V*B total =1.1*0.1*2000=220. Since the remaining parking space number in the business data is significantly different from the remaining parking space number collected by the inspection vehicle, which exceeds the determination threshold, the target parking lot fails the remaining parking space number quality inspection, and the remaining parking space number quality inspection result is obtained.
[0046] The accurate comparison between the number of vehicles collected by the inspection vehicle in real time and the business data can timely find the abnormal problem of the remaining parking space number, provide reliable basis for parking resource management and data correction, and ensure the accuracy of off-street parking data.
[0047] Further, the pre-processed vehicle data set is subjected to parking record number quality inspection in combination with the business data to obtain a parking record number quality inspection result, including: periodically calculating the number of vehicles in the parking lot based on the business data to obtain a periodic number of vehicles in the parking lot; comparing the periodic number of vehicles in the parking lot and the number of vehicles in the parking lot collected during the inspection in the pre-processed vehicle data set to execute a parking record number quality inspection algorithm; and if |C CAR -CB-CAR | p * V * B total , then the parking lot fails the parking record number quality inspection, as the parking record number quality inspection result; wherein, C CAR is the number of vehicles in the parking lot collected during the inspection, and C B-CAR is the periodic number of vehicles in the parking lot.
[0048] Specifically, the periodic number of vehicles in the parking lot is calculated according to the business data, and the number of vehicles in the periodic parking lot is obtained, that is, the periodic number of vehicles in the parking lot is C B-CAR , if the number of C B-CAR calculated during the inspection is more than one, the median is taken as the periodic number of vehicles in the parking lot. The periodic number of vehicles in the parking lot is compared with the number of vehicles in the parking lot collected in the preprocessing vehicle data set during the inspection, and the parking record number quality inspection algorithm is executed. The parking record number quality inspection algorithm determines whether the number of vehicles in the parking lot is abnormal by difference, and the determination rule is: if | C CAR -C B-CAR | > p * V * B total , then the parking lot fails the parking record number quality inspection, as the parking record number quality inspection result, wherein, C CAR is the number of vehicles in the parking lot collected during the inspection, and C B-CAR is the periodic number of vehicles in the parking lot.
[0049] Exemplarily, during the inspection, the city parking network platform calculates the number of vehicles in the parking lot based on the entry and exit data reported by the parking lot Pm every 10 minutes, a total of 3 times, the results are 920, 1050, 1080 vehicles respectively. Take the median, consider that the number of vehicles in the parking lot during the inspection is 1050. Then calculate the absolute value of the difference between the third party business system data and the inspection vehicle collection data | C CAR -C B-CAR | = 580 vehicles, which is much larger than the determination threshold p * V * B total = 1.1 * 0.1 * 2000 = 220, so it fails the parking record number quality inspection.
[0050] For the parking lot that fails the parking record number quality inspection, there are two cases: when C CAR is obviously more than C B-CAR , that is, the number of vehicles in the parking lot collected by the inspection vehicle is much more than the number of vehicles reported by the business data, which reflects that the parking lot has the problem of subjectively concealing part of the parking data not reported, then the city parking supervision department will arrange follow-up investigation and collect evidence; when C B-CAR is obviously more than C CAR , that is, the number of vehicles in the parking lot reported by the business data is much more than the number of vehicles collected by the inspection vehicle, which reflects that there are a large number of zombie car data in the parking lot, which is further cleaned up in combination with the zombie car quality inspection algorithm.
[0051] By cross-verification of the real vehicle quantity collected by the inspection vehicle and the business data, abnormal situations such as non-registered entry, non-canceled exit and zombie vehicle can be effectively found, so as to improve the accuracy and reliability of the parking data record.
[0052] Further, the zombie vehicle quality inspection is performed on the preprocessed vehicle data set in combination with the business data to obtain a zombie vehicle quality inspection result, including: performing license plate statistics of vehicles that have not exited on the business data to obtain a non-exit license plate detail; obtaining an inspection license plate detail collected by the inspection vehicle in the parking lot in the preprocessed vehicle data set; comparing the non-exit license plate detail and the inspection license plate detail to obtain a pending zombie vehicle list; and performing a zombie vehicle quality inspection algorithm based on the pending zombie vehicle list to calculate a probability that each pending zombie vehicle is a zombie vehicle, wherein the zombie vehicle probability calculation formula is P = lg(T) * n^(1 / 2) * 0.5, P is the probability that the pending zombie vehicle is a zombie vehicle, T is the length of time that the vehicle has been parked in the parking lot, and n is the number of times that the vehicle has been added to the pending zombie vehicle list since the business system last recorded the entry of the vehicle.
[0053] Specifically, the entry and exit records of the vehicles in the business data are analyzed, the vehicle information in the current state of not having exited is counted, and a non-exit license plate detail is formed, which reflects the vehicle list considered to be still in the parking lot in the business system. The license plate detail collected by the inspection vehicle in the parking lot in the preprocessed vehicle data set is cross-compared with the non-exit license plate detail, and the license plate data that satisfies both the business non-exit and still being in the parking lot but not collected by the inspection vehicle is screened out to form a pending zombie vehicle list, i.e., a candidate vehicle set that exists a long-term occupation risk.
[0054] Based on the zombie vehicle quality inspection algorithm, the probability P that each pending zombie vehicle in the pending zombie vehicle list is a zombie vehicle is calculated, and the zombie vehicle probability calculation formula is P = lg(T) * n^(1 / 2) * 0.5, wherein P is the probability that the pending zombie vehicle is a zombie vehicle, which is not more than 1, T is the length of time that the vehicle has been parked in the parking lot, and n is the number of times that the vehicle has been added to the pending zombie vehicle list since the business system last recorded the entry of the vehicle, which is used to measure the stability degree of its abnormal behavior. Through the zombie vehicle probability calculation formula, the parking time and the abnormal frequency are considered comprehensively to quantitatively evaluate the vehicle, and the probability that the pending vehicle is a zombie vehicle is calculated. The probability that the pending zombie vehicle is a zombie vehicle is compared with a determination threshold, for example, the determination threshold is set to 0.8, when the probability that the pending zombie vehicle is a zombie vehicle is greater than the determination threshold, the vehicle is marked as a suspected zombie vehicle and added to a suspected zombie vehicle list as a zombie vehicle quality inspection result.
[0055] Exemplarily, the license plate details collected by the inspection vehicle in the parking lot are compared with the license plate details still not out of the parking lot calculated based on the reporting data of the parking business system. The license plate data still in the parking lot but not collected by the inspection vehicle in the business system is added to the pending zombie car list. In the pending zombie car list, a total of 22 cars are obtained. Through the zombie car quality inspection algorithm, the probability P of each vehicle in the list being a zombie car is further calculated. Taking vehicle Ca as an example, it has been parked in the parking lot for 25 days and has been in the pending zombie car list for the second time. The probability P of it being a zombie car is P = lg(T) * n^(1 / 2) * 0.5 = 0.99, which is greater than the determination threshold 0.8, and the vehicle is added to the suspected zombie car list.
[0056] By automatically identifying zombie cars in off-road parking, long-term abnormal occupying vehicles can be accurately identified, the utilization rate of parking space resources can be optimized, and abnormal phenomena such as stopping without leaving and occupying to evade fees can be effectively reduced, thereby improving the level of urban parking management.
[0057] Further, when there is a parking lot that does not pass the remaining parking space number quality inspection in the quality inspection result, the city parking one net platform synchronizes the inaccurate remaining parking space number alarm and the remaining parking space number adjustment strategy suggestion to the third party parking system; when there is a parking lot that does not pass the parking record number quality inspection and the number of vehicles in the business data in the parking lot is less than the number of vehicles collected by the inspection vehicle in the quality inspection result, the data quality inspection algorithm module sends a business data missing alarm to the city parking one net platform; when there is a parking lot that does not pass the zombie car quality inspection in the quality inspection result, the data quality inspection algorithm module sends a suspected zombie car alarm and a suspected zombie car list to the city parking one net platform, and the city parking one net platform synchronizes the suspected zombie car list to the third party parking system.
[0058] Specifically, when there is a parking lot that does not pass the remaining parking space number quality inspection in the quality inspection result, the city parking one net platform synchronizes the inaccurate remaining parking space number alarm to the third party parking system, and pushes the remaining parking space number adjustment strategy suggestion, which is a parking space number correction suggestion based on the inspection data, to guide the third party system to update the real-time parking space display and avoid showing incorrect information to users.
[0059] When there is a parking lot that does not pass the parking record number quality inspection and the number of vehicles in the business data in the parking lot is less than the number of vehicles collected by the inspection vehicle in the quality inspection result, it indicates that there is an abnormal situation such as entering without registering, and the data quality inspection algorithm module sends a business data missing alarm to the city parking one net platform, which is used to prompt the third party system to perform data supplement or confirm the abnormal source.
[0060] When there is a parking lot that does not pass the zombie car quality inspection in the quality inspection result, the data quality inspection algorithm module sends a suspected zombie car alarm and a suspected zombie car list to the city parking one net platform, and synchronizes the suspected zombie car list to the third party parking system, so that the third party parking system confirms and cleans up the zombie car data.
[0061] By comprehensively and accurately checking the off-street parking data quality, judging the checking results, and giving feasible strategies for data governance and improving data quality of the third-party parking lot when quality problems are found, the distribution and management of parking resources are optimized, and the accuracy and reliability of the off-street parking data quality are improved.
[0062] In the second embodiment, based on the same inventive concept as the off-street parking data quality checking method based on a patrol vehicle in the foregoing embodiments, as shown in FIG. 2, the present application provides an off-street parking data quality checking system based on a patrol vehicle, wherein the off-street parking data quality checking system based on a patrol vehicle comprises: Figure 2 The data acquisition component 11 is configured to call the city parking network platform, acquire the record information and business data of the city off-street parking lot through the parking lot record or parking facility registration service, acquire the patrol vehicle resource information, determine the parking lot access data valley period of the city off-street parking lot in combination with the business data, and determine the patrol plan, wherein the patrol plan comprises a patrol scheme and a patrol path.
[0063] Further, the patrol plan determination component 12 is further configured to: the record information comprises the parking lot name, parking lot code, longitude and latitude, number of parking spaces, and internal map of the parking lot; and the business data comprises vehicle access records, vehicle exit records, real-time number of vehicles, real-time number of remaining parking spaces, and the like.
[0064] Further, the quality inspection result acquisition component 14 is further configured to: the preprocessing comprises deduplication of the same parking lot and elimination of the parking lot that does not conform to the rules.
[0065] Further, the quality inspection result acquisition component 14 is further configured to: the preprocessing comprises deduplication of the same parking lot and elimination of the parking lot that does not conform to the rules.
[0066] Further, the quality inspection result obtaining component 14 is further configured to: pre-process the vehicle data set by a data quality inspection algorithm module to obtain a pre-processed vehicle data set; perform remaining parking space number quality inspection on the pre-processed vehicle data set in combination with business data to obtain a remaining parking space number quality inspection result; perform parking record number quality inspection on the pre-processed vehicle data set in combination with business data to obtain a parking record number quality inspection result; perform zombie vehicle quality inspection on the pre-processed vehicle data set in combination with business data to obtain a zombie vehicle quality inspection result; and return the remaining parking space number quality inspection result, the parking record number quality inspection result and the zombie vehicle quality inspection result to the urban parking one-stop network platform as quality inspection results.
[0067] Further, the quality inspection result obtaining component 14 is further configured to: acquire the number of in-park vehicles collected during the inspection period in the pre-processed vehicle data set, compare the number of in-park vehicles with business data, and perform a remaining parking space number quality inspection algorithm; if |RB QC -RB BD |>ρ*V*B total , the parking lot fails the remaining parking space number quality inspection, and the remaining parking space number quality inspection result is obtained; wherein, RB QC is the number of remaining parking spaces during the inspection period calculated based on the pre-processed vehicle data set, B total is the total number of parking spaces in the parking lot, RB QC =B total -C CAR , C CAR is the number of in-park vehicles collected during the inspection period, R BBD is the number of remaining parking spaces in the business data during the inspection period, V is the percentage of system-allowed error, and p is the correction coefficient, and p≥1.
[0068] Further, the quality inspection result obtaining component 14 is further configured to: calculate the number of in-park vehicles on a regular basis based on business data to obtain a regular number of in-park vehicles; compare the regular number of in-park vehicles with the number of in-park vehicles collected during the inspection period in the pre-processed vehicle data set, and perform a parking record number quality inspection algorithm; if |C CAR -C B-CAR |>ρ*V*Btotal, the parking lot fails the parking record number quality inspection, and the parking record number quality inspection result is obtained; wherein, C CAR is the number of in-park vehicles collected during the inspection period, C B-CAR is the regular number of in-park vehicles.
[0069] Further, the quality inspection result obtaining component 14 is further used for: performing off-site license plate statistics on the business data to obtain off-site license plate details; obtaining inspection license plate details collected by the inspection vehicle in the parking lot from the pre-processed vehicle data set; comparing the off-site license plate details and the inspection license plate details to obtain a list of pending zombie vehicles; executing a zombie vehicle quality inspection algorithm based on the list of pending zombie vehicles to calculate a probability that each pending zombie vehicle is a zombie vehicle; wherein the zombie vehicle probability calculation formula is: P = lg(T) * n^(1 / 2) * 0.5; P is the probability that the pending zombie vehicle is a zombie vehicle, T is the length of time that the vehicle has been parked in the parking lot, in days; n is the number of times since the last time the business system records the vehicle entering the parking lot that the vehicle has been added to the list of pending zombie vehicles; if the probability that the pending zombie vehicle is a zombie vehicle is greater than a determination threshold, then it is added to a suspected zombie vehicle list, and the suspected zombie vehicle list is used as the zombie vehicle quality inspection result.
[0070] Further, the quality inspection result obtaining component 14 is further used for: when there is a parking lot that does not pass the remaining parking space number quality inspection in the quality inspection result, the city parking one network platform synchronizes the inaccurate remaining parking space number alarm and the remaining parking space number adjustment strategy suggestion to the third-party parking system; when there is a parking lot that does not pass the parking record number quality inspection and the number of vehicles in the business data is less than the number of vehicles collected by the inspection vehicle in the quality inspection result, the data quality inspection algorithm module sends a business report missing alarm to the city parking one network platform; when there is a parking lot that does not pass the zombie vehicle quality inspection in the quality inspection result, the data quality inspection algorithm module sends a suspected zombie vehicle alarm and a suspected zombie vehicle list to the city parking one network platform, and the city parking one network platform synchronizes the suspected zombie vehicle list to the third-party parking system.
[0071] The embodiments in the specification are described in a progressive manner, and each embodiment focuses on the difference from other embodiments. The method and specific examples of the first embodiment are also applicable to the system of the second embodiment. Based on the detailed description of the method, those skilled in the art can clearly understand the system of the second embodiment. Therefore, for the sake of brevity of the specification, the system will not be described in detail.
[0072] The above description of the disclosed embodiments enables a person skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.
[0073] Obviously, for those skilled in the art, some improvements and modifications can be made to the present application without departing from the principles of the present application, and these improvements and modifications also fall within the protection scope of the present application.
Claims
1. A method for off-street parking data quality inspection based on a patrol vehicle, characterized in that, The method comprises: Calling the urban parking network platform, obtaining the record information and business data of the urban off-street parking lot through the parking lot record or parking facility registration business; Obtaining the inspection vehicle resource information, combining the business data to determine the low-traffic period of the urban off-street parking lot, and determining the inspection plan, wherein the inspection plan comprises an inspection scheme and an inspection path; The inspection vehicle collects vehicle data according to the inspection path and the inspection scheme to obtain a vehicle data set; Combining the business data, the vehicle data set is preprocessed and data quality inspected through a data quality inspection algorithm module to obtain a quality inspection result, and the quality inspection result is returned to the urban parking network platform.
2. A method of off-street parking data quality inspection based on a patrol vehicle according to claim 1, characterized in that, The record information comprises the parking lot name, parking lot code, longitude and latitude, number of parking spaces, and internal map of the parking lot. The business data comprises vehicle entry records, vehicle exit records, real-time number of vehicles present, and real-time number of remaining parking spaces.
3. The off-street parking data quality inspection method based on a patrol vehicle according to claim 1, characterized in that, The preprocessing comprises deduplication of repeated license plates in the same parking lot and elimination of license plates that do not meet the rules.
4. The off-street parking data quality inspection method based on a patrol vehicle of claim 1, wherein, Combining the business data, the vehicle data set is preprocessed and data quality inspected through a data quality inspection algorithm module to obtain a quality inspection result, and the quality inspection result is returned to the urban parking network platform, comprising: The vehicle data set is preprocessed through a data quality inspection algorithm module to obtain a preprocessed vehicle data set; The preprocessed vehicle data set is combined with the business data to perform remaining parking space number quality inspection to obtain a remaining parking space number quality inspection result; The preprocessed vehicle data set is combined with the business data to perform parking record number quality inspection to obtain a parking record number quality inspection result; The preprocessed vehicle data set is combined with the business data to perform zombie vehicle quality inspection to obtain a zombie vehicle quality inspection result; The remaining parking space number quality inspection result, the parking record number quality inspection result, and the zombie vehicle quality inspection result are returned to the urban parking network platform as the quality inspection result.
5. A method of off-street parking data quality inspection based on a patrol vehicle according to claim 4, characterized in that, Combining the business data, the vehicle data set is preprocessed and data quality inspected through a data quality inspection algorithm module to obtain a quality inspection result, and the quality inspection result is returned to the urban parking network platform, comprising: The number of vehicles in the parking lot collected during the inspection is obtained from the preprocessed vehicle data set, and compared with the business data to perform a remaining parking space number quality inspection algorithm; if |RB QC -RB BD |>ρ*V*B total then the number of remaining parking spaces that do not pass the quality inspection is taken as the result of the quality inspection of the number of remaining parking spaces; wherein RB QC is the number of remaining parking spaces during the inspection period calculated based on the pre-processed vehicle data set, B total is the total number of parking spaces in the parking lot, RB QC = B total - C CAR , C CAR is the number of vehicles in the parking lot collected during the inspection period, R BBD is the number of remaining parking spaces in the business data during the inspection period, V is the percentage of error allowed by the system, and p is the correction coefficient, and p ≥ 1.
6. A method of off-street parking data quality inspection based on a patrol vehicle according to claim 5, characterized in that, The number of vehicles in the parking lot collected during the inspection is obtained from the preprocessed vehicle data set, and compared with the business data to perform a remaining parking space number quality inspection algorithm; The preprocessed vehicle data set is combined with the business data to perform parking record number quality inspection to obtain a parking record number quality inspection result, comprising: The number of vehicles in the parking lot is calculated based on the business data to obtain a regular number of vehicles in the parking lot; If |C CAR -C B-CAR |>ρ*V*B total , the parking lot fails the parking record number quality inspection, and the parking record number quality inspection result is obtained. where C CAR is the number of vehicles in the field collected during the inspection, C B-CAR is the number of vehicles in the field periodically.
7. A method of off-street parking data quality inspection based on a patrol vehicle according to claim 6, characterized in that, The number of vehicles in the parking lot collected during the inspection is compared with the number of vehicles in the parking lot in the preprocessed vehicle data set to perform a parking record number quality inspection algorithm; The preprocessed vehicle data set is combined with the business data to perform zombie vehicle quality inspection to obtain a zombie vehicle quality inspection result, comprising: The number of non-exit license plates is counted based on the business data to obtain a non-exit license plate detail; The inspection license plate detail collected by the inspection vehicle in the parking lot is obtained from the preprocessed vehicle data set; The non-exit license plate detail and the inspection license plate detail are compared to obtain a pending zombie vehicle list; A zombie vehicle quality inspection algorithm is performed based on the pending zombie vehicle list to calculate the probability that each pending zombie vehicle is a zombie vehicle; The zombie vehicle probability calculation formula is: P = lg(T) * n^(1 / 2) * 0.5; P is the probability of the pending zombie car being a zombie car, T is the length of time the vehicle has been parked in the parking lot, in days; n is the number of times the vehicle has been added to the list of pending zombie cars since the business system last recorded the vehicle entering the parking lot; If the probability of the pending zombie car being a zombie car is greater than the determination threshold, it is added to the list of suspected zombie cars, and the list of suspected zombie cars is used as the zombie car quality inspection result.
8. The off-street parking data quality inspection method based on a patrol vehicle according to claim 1, characterized in that, Comprise: When there is a parking lot in the quality inspection result that does not pass the remaining parking space number quality inspection, the city parking one net platform synchronizes the inaccurate remaining parking space number alarm and the remaining parking space number adjustment strategy to the third party parking system; When there is a parking lot in the quality inspection result that does not pass the parking record number quality inspection and the number of vehicles in the business data in the parking lot is less than the number of vehicles collected by the inspection vehicle, the data quality inspection algorithm module sends a business data leakage alarm to the city parking one net platform; When there is a parking lot in the quality inspection result that does not pass the zombie car quality inspection, the data quality inspection algorithm module sends a suspected zombie car alarm and a suspected zombie car list to the city parking one net platform, and the city parking one net platform synchronizes the suspected zombie car list to the third party parking system.
9. An off-street parking data quality inspection system based on a patrol vehicle, characterized by, For implementing the steps of the off-street parking data quality inspection method based on the inspection vehicle according to any one of claims 1-8, comprising: A data acquisition component for calling the city parking one net platform to obtain the record information and business data of the city off-street parking lot through the parking lot record or parking facility registration business; An inspection plan determination component for obtaining inspection vehicle resource information and determining the vehicle entry and exit data valley period of the city off-street parking lot in combination with the business data to determine an inspection plan, wherein the inspection plan comprises an inspection scheme and an inspection path; A vehicle data acquisition component for the inspection vehicle to collect vehicle data according to the inspection path and the inspection scheme to obtain a vehicle data set; A quality inspection result obtaining component for pre-processing and data quality inspection of the vehicle data set by a data quality inspection algorithm module in combination with the business data to obtain a quality inspection result and return the quality inspection result to the city parking one net platform.