Parking Point Identification Method, Device, Computer Equipment and Storage Medium

Abstract

The present application relates to a parking point recognition method, apparatus, computer device, and storage medium. The method includes: obtaining position data of multiple vehicles at a parking point, grouping the position data according to the position characteristics of each position data to obtain multiple combinations, clustering the position data in each combination to obtain multiple sets, and obtaining the position information of the center point of each set. By using this method, the recognition of the vehicle parking point is realized. Since before clustering the position data, the position data is grouped based on the position characteristics, and then the position data is clustered in units of the combinations obtained after grouping, it avoids directly clustering and recognizing the position data with a large amount of data, reduces the performance requirements of the hardware for processing data, and improves the recognition efficiency of the vehicle parking point. In addition, the position data in each combination has similar position characteristics, and clustering and recognizing the position data based on this grouping improves the accuracy of the parking point recognition.

Description

Technical Field

[0001] This application relates to the technical field of data processing, and particularly to a method, apparatus, computer device, and storage medium for identifying parking spots. Background Art

[0002] With the development of automotive technology, the utilization rate of automobiles has been increasing year by year, and users' requirements for the use of automobiles have also been upgraded accordingly. Currently, various data of vehicles can be analyzed by means of technologies such as the Internet of Vehicles and big data to optimize and upgrade the performance of various parts of the automobile.

[0003] However, for the identification of parking gathering places, although there are available methods, there is a lack of attention to improving data processing efficiency, which cannot meet the current application scenarios, may lead to difficulty in achieving ideal application effects or even difficulty in actual application. Especially for the Internet of Vehicles data characterized by a large volume, a reasonable method for improving data processing efficiency is needed to realize the identification of parking gathering places. Summary of the Invention

[0004] Based on this, in view of the above technical problems, it is necessary to provide a method, apparatus, computer device, computer-readable storage medium, and computer program product for identifying parking spots, which improves the efficiency of parking spot identification.

[0005] In a first aspect, this application provides a method for identifying parking spots. The method includes:

[0006] Obtain the position data of multiple vehicles at parking spots;

[0007] Group the position data according to the position characteristics of each position data to obtain multiple combinations; where the number of position data included in each combination does not exceed a preset threshold;

[0008] Cluster the position data in each combination to obtain multiple sets;

[0009] Obtain the position information of the center point of each set; where the center point is the gathering point where the vehicles park.

[0010] In one embodiment, the position characteristic includes a geographical location characteristic, and the grouping of the position data according to the position characteristics of each position data to obtain multiple combinations includes:

[0011] Group the position data according to the geographical location characteristics of each position data to obtain multiple initial combinations;

[0012] Group the location data in the target combination according to the longitude and latitude parameters of each location data in the target combination to obtain a plurality of candidate combinations; wherein, the target combination is a combination in which the number of location data included in a plurality of the initial combinations exceeds the preset threshold.

[0013] When the number of location data included in the candidate combination exceeds the preset threshold, use the candidate combination as the target combination for re-grouping until the number of location data in each combination after grouping is less than or equal to the preset threshold.

[0014] In one embodiment, the grouping the location data in the target combination according to the longitude and latitude parameters of each location data in the target combination to obtain a plurality of candidate combinations includes:

[0015] Obtain the target mean of each location data in the target combination; wherein, the target mean includes one of a target longitude mean and a target latitude mean.

[0016] Group the location data in the target combination according to the target direction according to the target mean to obtain two candidate combinations; wherein, the target direction includes one of a longitude direction and a latitude direction, and the longitude and latitude identifiers in the target mean and the target direction are the same.

[0017] In one embodiment, before obtaining the target mean of each location data in the target combination, the parking point recognition method further includes:

[0018] Perform truncation processing on the longitude parameter and the latitude parameter of each location data in the target combination to obtain the truncated first longitude data and first latitude data; wherein, the number of decimal places of the first longitude data and the first latitude data does not exceed a preset length.

[0019] Perform deduplication processing on the first longitude data and the first latitude data respectively to obtain the deduplicated second longitude data and second latitude data.

[0020] Obtain the number of longitudes of the second longitude data and the number of latitudes of the second latitude data respectively.

[0021] Determine the target direction based on the direction corresponding to the target number; wherein, the target number is the decimal of the number of longitudes and the number of latitudes.

[0022] In one embodiment, the obtaining the target mean of each location data in the target combination includes:

[0023] Obtain three means corresponding to three types of data in the target set respectively; wherein, the target set is a longitude set or a latitude set, the target set is determined according to the target direction, the longitude set includes the first longitude data, the second longitude data and the longitude parameters of each position data in the target combination, and the latitude set includes the first latitude data, the second latitude data and the latitude parameters of each position data in the target combination;

[0024] Obtain the target mean of each position data in the target combination according to the three means corresponding to the three types of data in the target set.

[0025] In one embodiment, the parking point recognition method further includes:

[0026] Group the position data in the target combination according to the target direction respectively according to the three means of the three types of data in the target set, and obtain the corresponding first candidate combination and second candidate combination respectively;

[0027] Obtain the first quantity of the position data included in the first candidate combination and the second quantity of the position data included in the second candidate combination;

[0028] The step of obtaining the target mean of each position data in the target combination according to the three means corresponding to the three types of data in the target set includes:

[0029] Determine the mean corresponding to the target ratio in the target set as the target mean according to the ratio of the first quantity and the second quantity corresponding to each of the three means; wherein, the difference between the target ratio and 1 is the smallest.

[0030] In one embodiment, the step of clustering the position data in each combination to obtain multiple sets includes:

[0031] Cluster the position data in each combination by using a preset density-based spatial clustering of applications with noise (DBSCAN) algorithm to obtain multiple sets.

[0032] In a second aspect, the present application further provides a parking point recognition device. The device includes:

[0033] A position module, configured to obtain the position data of multiple vehicles at the parking point;

[0034] A grouping module, configured to group each of the position data according to the position characteristics of each of the position data to obtain multiple combinations; wherein, the number of position data included in each combination does not exceed a preset threshold;

[0035] A clustering module, configured to cluster the position data in each combination to obtain multiple sets;

[0036] A central module for obtaining the position information of the center points of the sets; wherein, the center point is the gathering point where the vehicle parks.

[0037] In a third aspect, the present application also provides a computer device. The computer device includes a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, the steps of the parking point recognition method provided in any of the above embodiments are implemented.

[0038] In a fourth aspect, the present application also provides a computer-readable storage medium. The computer-readable storage medium stores a computer program thereon, and when the computer program is executed by a processor, the steps of the parking point recognition method provided in any of the above embodiments are implemented.

[0039] In a fifth aspect, the present application also provides a computer program product. The computer program product includes a computer program, and when the computer program is executed by a processor, the steps of the parking point recognition method provided in any of the above embodiments are implemented.

[0040] For the above parking point recognition method, device, computer device, storage medium, and computer program product, first obtain the position data of multiple vehicles at the parking point, then group the position data according to the position characteristics of each position data to obtain multiple combinations, and then cluster the position data in each combination to obtain multiple sets, and then obtain the position information of the center points of each set, realizing the recognition of the vehicle parking point. Since before clustering the position data, the position data is grouped based on the position characteristics, and then the position data is clustered and recognized in units of the combinations obtained after grouping, it avoids directly clustering and recognizing the position data with a large amount of data, reduces the performance requirements for hardware to process data, and also improves the recognition efficiency of the vehicle parking point. In addition, the position data in each combination obtained by grouping has similar position characteristics, and clustering and recognizing the position data based on this grouping improves the accuracy of parking point recognition. Description of the Drawings

[0041] Figure 1 It is a schematic flowchart of the parking point recognition method in an embodiment;

[0042] Figure 2 It is a schematic flowchart of the parking point recognition method in an embodiment;

[0043] Figure 3 It is a schematic flowchart of the parking point recognition method in an embodiment;

[0044] Figure 4 It is a schematic flowchart of the parking point recognition method in an embodiment;

[0045] Figure 5 It is a schematic flowchart of a parking point recognition method in an embodiment;

[0046] Figure 6 It is a schematic flowchart of a parking point recognition method in an embodiment;

[0047] Figure 7 It is a schematic flowchart of a parking point recognition method in an embodiment;

[0048] Figure 8 It is a schematic flowchart of a parking point recognition method in an embodiment;

[0049] Figure 9 It is a structural block diagram of a parking point recognition device in an embodiment;

[0050] Figure 10 It is an internal structure diagram of a computer device in an embodiment. Detailed implementation manners

[0051] In order to make the objectives, technical solutions and advantages of the present application clearer, the present application will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0052] As described in the background art, the recognition efficiency of vehicle parking points is relatively low. Therefore, the present application provides a parking point recognition method, device, computer device, storage medium and computer program product to improve the parking point recognition efficiency.

[0053] In one embodiment, as Figure 1 shown, a parking point recognition method is provided. In this embodiment, it is exemplified that the method is applied to a terminal. It can be understood that the method can also be applied to a server, and can also be applied to a system including a terminal and a server, and is implemented through the interaction between the terminal and the server. In this embodiment, the parking point recognition method may include the following steps S10 to step S40.

[0054] S10: Obtain the position data of multiple vehicles at the parking point.

[0055] The position data is used to represent the position where the vehicle is in a stationary state, and each parking point corresponds to one position data. Exemplarily, the position data may include longitude parameters, latitude parameters, altitude, etc. Exemplarily, based on vehicle networking data, the parking point recognition method can be used to obtain the position data of multiple vehicles at the parking point, where the vehicle networking data may include time stamps, vehicle speeds, longitude parameters, latitude parameters, etc. In the embodiments of the present application, there are no limitations on the manner of obtaining the position data, the specific content included in the position data, etc.

[0056] S20: Group the location data according to the location characteristics of each location data to obtain multiple combinations.

[0057] The location characteristics are used to reflect the location characteristics of the vehicle at the parking point. Among them, the number of location data included in each combination does not exceed a preset threshold. The preset threshold is a fixed value set in advance, which can be set according to the data processing ability of the terminal, multiple experiments, etc., and is not limited here. Exemplarily, the preset threshold can be set to 10000, 60000, 150000, etc. Suppose n combinations are obtained after grouping and are denoted as Group1, Group 2, ……, Group n, where n is a positive integer greater than or equal to 1.

[0058] S30: Cluster the location data in each combination to obtain multiple sets.

[0059] Cluster the n combinations Group 1, Group 2, …, Group n obtained in step S30 to obtain N sets, denoted as Class 1, Class 2, ……, Class N, where N is a positive integer greater than or equal to 1. In this application, any suitable clustering algorithm can be used to cluster the location data in each combination. For example, partition-based methods, density-based methods, hierarchical methods. The specific way of clustering the location data is not limited here. After clustering, the category to which each parking point belongs, the clustering center and the number of clusters of its category can be obtained, and at the same time, the identification of abnormal data points is realized.

[0060] S40: Obtain the location information of the center point of each set.

[0061] The center point of each set is the gathering point of vehicle parking, that is, the center point of each set is the location with the highest vehicle parking density. Based on the N sets: Class 1, Class 2, …, Class N obtained in the above step S30, the center points of these N sets: Center 1, Center 2, ……, Center N, then the location information corresponding to these N center points can be obtained.

[0062] The parking spot recognition method provided by the above embodiments first obtains the position data of multiple vehicles at the parking spots, then groups the position data according to the position characteristics of each position data to obtain multiple combinations, then clusters the position data in each combination to obtain multiple sets, and then obtains the position information of the center points of each set, realizing the recognition of vehicle parking spots. Since before clustering the position data, the position data is grouped based on the position characteristics, and then the position data is clustered and recognized in units of the combinations obtained after grouping, it avoids directly clustering and recognizing the position data with a large amount of data, reduces the performance requirements for hardware to process data, and also improves the recognition efficiency of vehicle parking spots. In addition, the position data in each combination obtained by grouping has similar position characteristics, and clustering and recognizing the position data based on this grouping improves the accuracy of parking spot recognition.

[0063] In one embodiment, the above position characteristics may include geographical location characteristics. Then, step S20, grouping the position data according to the position characteristics of each position data to obtain multiple combinations, may include the following steps S210 to S230.

[0064] S210: Group the position data according to the geographical location characteristics of each position data to obtain multiple initial combinations.

[0065] The geographical location characteristics refer to the spatio-temporal characteristics possessed by the position data, which may include natural geographical location characteristics, human geographical location characteristics, etc. For example, sea-land location characteristics, longitude and latitude location characteristics, economic geographical location characteristics, political geographical location characteristics, cultural geographical location characteristics, etc., which are not limited here. Exemplarily, the position data can be grouped according to the longitude and latitude location characteristics of the position data and the map data by prefecture-level administrative regions. Exemplarily, the position data can be grouped according to the economic geographical location characteristics of the position data into multiple preset economic regions. Based on the geographical location characteristics, m initial combinations are obtained after grouping the position data, denoted as: G1, G2, ……, Gm.

[0066] S220: Group the position data in the target combination according to the longitude and latitude parameters of the position data in the target combination to obtain multiple candidate combinations.

[0067] The target combination (denoted as Gi) is a combination in which the number of location data included in multiple initial combinations exceeds a preset threshold. Here, the preset threshold is a certain fixed value set in advance. Among the m initial combinations obtained after grouping based on step S210, if there is one or more initial combinations in which the number of location data included exceeds the preset threshold, it indicates that the number of location data included in this initial combination is large. Therefore, this initial combination is taken as the target combination, and the target combination is regrouped and further segmented. Among them, regrouping the target combination is based on the longitude and latitude parameters of the target combination. The longitude and latitude parameters include the longitude parameter (longitude) and the latitude parameter (latitude). Exemplarily, the preset threshold is set to 60000. If the number of location data included in the initial combinations G3 and G9 both exceeds 60000, then the initial combinations G3 and G9 are taken as the target combinations, and based on the longitude and latitude parameters of each location data in G3 and G9, the initial combinations G3 and G9 are grouped again to obtain the candidate combinations G31, G32, G91, and G92.

[0068] S230: In the case where the number of location data included in the candidate combination exceeds the preset threshold, the candidate combination is taken as the target combination for regrouping until the number of location data in each combination after grouping is less than or equal to the preset threshold.

[0069] For the candidate combinations obtained after the secondary segmentation based on step S220, if the number of location data included in the candidate combination still exceeds the preset threshold, it indicates that the number of location data included in this candidate combination is large. It is necessary to take this candidate combination as the target combination and, based on step S220, regroup this target combination for further segmentation until the number of location data in each combination obtained after grouping is less than or equal to the preset threshold to meet the requirements of location data clustering. Exemplarily, for the above candidate combinations G31, G32, G91, and G92, if the number of location data included in the candidate combination G91 still exceeds 60000, then this candidate combination G91 is taken as the target combination, and based on the longitude and latitude parameters of each location data in G91, G91 is grouped again until the number of location data in each combination obtained after grouping is less than or equal to 60000.

[0070] The parking point recognition method provided by the above embodiments performs a primary grouping on each piece of location data according to the geographical location characteristics of the location data to obtain an initial combination, and then re-groups the target combinations including the number of location data exceeding a preset threshold until the number of location data included in each combination obtained after grouping does not exceed the preset threshold, that is, the location data is segmented multiple times so that each combination after grouping meets the preset threshold requirements. Based on this grouping result, clustering is performed on each combination, which can improve the clustering recognition accuracy and processing efficiency of the location data, thereby facilitating the improvement of the parking point recognition efficiency and accuracy.

[0071] In one embodiment, as Figure 3 shown, in step S220, according to the longitude and latitude parameters of each piece of location data in the target combination, the location data in the target combination is grouped to obtain multiple candidate combinations, which may include the following steps S221 and step S222.

[0072] S221: Obtain the target mean value of each piece of location data in the target combination.

[0073] The target mean value includes one of the target longitude mean value and the target latitude mean value. Exemplarily, the target longitude mean value of each piece of location data in the target combination can be obtained according to the longitude parameters of each piece of location data in the target combination; alternatively, the target latitude mean value of each piece of location data in the target combination can be obtained according to the latitude parameters of each piece of location data in the target combination. In practical applications, it can be determined whether to obtain the target longitude mean value or the target latitude mean value according to the specific application scenario, which is not limited herein.

[0074] S222: According to the target mean value, group the location data in the target combination in the target direction to obtain two candidate combinations.

[0075] The target direction includes one of the longitude direction and the latitude direction, and the longitude and latitude identifiers in the target mean value and the target direction are the same. Exemplarily, when the target mean value is the target longitude mean value, the target direction is the longitude direction, then according to the target longitude mean value, the location data in the target combination is grouped in the longitude direction to obtain two candidate combinations; when the target mean value is the target latitude mean value, the target direction is the latitude direction, then according to the target latitude mean value, the location data in the target combination is grouped in the latitude direction to obtain two candidate combinations.

[0076] The parking point recognition method provided by the above embodiments realizes the re-grouping of the target combination by obtaining the target longitude mean value or the target latitude mean value of each piece of location data in the target combination and grouping the location data in the target combination in the longitude direction or the latitude direction to obtain two candidate combinations, so as to meet the requirements of combination clustering recognition, which is beneficial to improving the clustering recognition efficiency and accuracy, thereby improving the parking point recognition efficiency and accuracy.

[0077] In one embodiment, as Figure 4 shown, before performing step S221 of obtaining the target mean of each position data in the target combination, the parking point recognition method may further include the following steps S50 to S80.

[0078] S50: Truncate the longitude parameter and latitude parameter of each position data in the target combination respectively to obtain the truncated first longitude data and first latitude data.

[0079] The number of decimal places of the first longitude data (lons) and the first latitude data (lats) does not exceed a preset length. Wherein, the preset length is a certain positive integer set in advance. For example, the preset length can be set to 2, 3, 4, etc., and can be set according to factors such as positioning accuracy, data processing performance of the terminal, and the number of position data, and is not limited herein. Exemplarily, when the preset length is set to 2, and the longitude and latitude parameters of a position data A are (116.41667, 39.91667), after truncating the longitude and latitude parameters of position data A, the first longitude data obtained by retaining 2 decimal places is 116.41 and the first latitude data is 39.91.

[0080] S60: Remove duplicates from the first longitude data and the first latitude data respectively to obtain the deduplicated second longitude data and second latitude data.

[0081] The second longitude data is denoted as lon, and the data in the second longitude data are all different from each other. The second latitude data is denoted as lat, and the data in the second latitude data are also all different from each other.

[0082] S70: Obtain the number of longitudes of the second longitude data and the number of latitudes of the second latitude data respectively.

[0083] Statistical analysis is performed on the second longitude data and the second latitude data respectively to obtain the number of longitudes of the second longitude data and the number of latitudes of the second latitude data. Among them, the number of longitudes of the second longitude data is denoted as sum_lon, and the number of latitudes of the second latitude data is denoted as sum_lat.

[0084] S80: Determine the target direction according to the direction corresponding to the target number.

[0085] The number of targets is the smaller value between the number of longitudes and the number of latitudes. If the number of longitudes in the second longitude data is less than or equal to the number of latitudes in the second latitude data, i.e., sum_lon < sum_lat, then the number of targets is the number of longitudes in the second longitude data. Therefore, the target direction is determined as the longitude direction, and thus the target longitude mean is obtained, and the target combinations are grouped according to the longitude direction. If the number of longitudes in the second longitude data is greater than the number of latitudes in the second latitude data, i.e., sum_lot ≥ sum_lat, then the number of targets is the number of latitudes in the second latitude data. Therefore, the target direction is determined as the latitude direction, and thus the target latitude mean is obtained, and the target combinations are grouped according to the latitude direction.

[0086] The parking point recognition method provided in the above embodiment truncates and de-duplicates the position data in the target combination, and determines the target direction according to the number of longitudes and the number of latitudes. Thus, the corresponding target mean is obtained according to the target direction, and the position data in the target combination are grouped again according to the target direction, improving the position data processing efficiency, which is beneficial to improving the subsequent clustering recognition efficiency and accuracy, so as to improve the parking point recognition efficiency and accuracy.

[0087] In one embodiment, as Figure 5 shown, step S221 of obtaining the target mean of the position data in the target combination may include the following steps S221a and S221b.

[0088] S221a: Respectively obtain the three means corresponding to the three types of data in the target set.

[0089] The target set is one of the longitude set or the latitude set, and whether the target set is the longitude set or the latitude set can be determined according to the target direction. Among them, the longitude set includes the first longitude data, the second longitude data, and the longitude parameters of the position data in the target combination, and the latitude set includes the first latitude data, the second latitude data, and the latitude parameters of the position data in the target combination.

[0090] Exemplarily, if the target direction is the longitude direction, then the target set is the longitude set. Therefore, the first mean of the first longitude data (denoted as avg_lons), the second mean of the second longitude data (denoted as avg_lon), and the third mean of the longitude parameters of the position data in the target combination (denoted as avg_longitude) are respectively obtained. If the target direction is the latitude direction, then the target set is the latitude set. Therefore, the fourth mean of the first latitude data (denoted as avg_lats), the fifth mean of the second latitude data (avg_lat), and the sixth mean of the latitude parameters of the position data in the target combination (denoted as avg_latitude) are respectively obtained.

[0091] S221b: Obtain the target mean of the data at each position in the target combination according to the three means corresponding to the three types of data in the target set.

[0092] Exemplarily, if the three means obtained in step S221a are the three means of the three types of data in the longitude set, that is, the first mean, the second mean, and the third mean, then obtain the target longitude mean of the data at each position in the target combination according to the first mean, the second mean, and the third mean. If the three means obtained in step S221a are the three means of the three types of data in the latitude set, that is, the fourth mean, the fifth mean, and the sixth mean, then obtain the target latitude mean of the data at each position in the target combination according to the fourth mean, the fifth mean, and the sixth mean.

[0093] The parking point recognition method provided in the above embodiment determines the target mean of the data at each position in the target combination by obtaining the three means of the three types of data in the target set, and thus groups the target combination in the target direction according to the target mean, making the segmentation method of each combination more reasonable, and thus the clustering recognition based on the grouped combination is more accurate.

[0094] In one embodiment, as Figure 6 shown, based on the parking point recognition method provided in the above embodiment, the following steps S90 to step S100 may further be included.

[0095] S90: Respectively group the position data in the target combination in the target direction according to the three means of the three types of data in the target set, and obtain the corresponding first candidate combination and second candidate combination respectively.

[0096] For the target combination Gi, the first candidate combination obtained by grouping the position data therein in the target direction is denoted as Ni1, and the second candidate combination is denoted as Ni2.

[0097] Exemplarily, in the case where the target set is the longitude set, the three means of the three types of data in the longitude set obtained based on step S221a are the first mean, the second mean, and the third mean respectively. Based on this, according to the first mean, the second mean, and the third mean, group the position data in the target combination in the longitude direction three times respectively, and obtain the first candidate combination (denoted as N11) and the second candidate combination (denoted as N12) corresponding to the first mean, the first candidate combination (denoted as N21) and the second candidate combination (denoted as N22) corresponding to the second mean, and the first candidate combination (denoted as N31) and the second candidate combination (denoted as N32) corresponding to the third mean.

[0098] Exemplarily, when the target set is the latitude set, the three means of the three types of data in the latitude set obtained based on step S221a are the fourth mean, the fifth mean, and the sixth mean respectively. Based on this, according to the fourth mean, the fifth mean, and the sixth mean, the position data in the target combination are grouped three times in the latitude direction respectively to obtain the first candidate combination (denoted as N41) and the second candidate combination (denoted as N42) corresponding to the fourth mean, the first candidate combination (denoted as N51) and the second candidate combination (denoted as N52) corresponding to the fifth mean, and the first candidate combination (denoted as N61) and the second candidate combination (denoted as N62) corresponding to the sixth mean.

[0099] S100: Obtain the first quantity of the position data included in the first candidate combination and the second quantity of the position data included in the second candidate combination.

[0100] For the first candidate combination Ni1 and the second candidate combination Ni2, respectively obtain the first quantity corresponding to the first candidate combination Ni1 denoted as ni1 and the second quantity corresponding to the second candidate combination Ni2 denoted as ni2.

[0101] Exemplarily, when the target set is the longitude set, respectively obtain the first quantity (denoted as n11) of the first candidate combination corresponding to the first mean and the second quantity (denoted as n12) of the second candidate combination, the first quantity (denoted as n21) of the first candidate combination corresponding to the second mean and the second quantity (denoted as n22) of the second candidate combination, and the first quantity (denoted as n31) of the first candidate combination corresponding to the third mean and the second quantity (denoted as n32) of the second candidate combination.

[0102] Exemplarily, when the target set is the latitude set, respectively obtain the first quantity (denoted as n41) of the first candidate combination corresponding to the fourth mean and the second quantity (denoted as n42) of the second candidate combination, the first quantity (denoted as n51) of the first candidate combination corresponding to the fifth mean and the second quantity (denoted as n52) of the second candidate combination, and the first quantity (denoted as n61) of the first candidate combination corresponding to the sixth mean and the second quantity (denoted as n62) of the second candidate combination.

[0103] Based on the above, step S221b: Obtain the target mean of each position data in the target combination according to the three means corresponding to the three types of data in the target set, may include: the step of determining the mean corresponding to the target ratio as the target mean in the target set according to the ratio of the first quantity and the second quantity corresponding to each of the three means. Wherein, the difference between the target ratio and 1 is the smallest. The closer the ratio between the first quantity and the second quantity is to 1, the closer the number of position data included in the first candidate combination and the second candidate combination is, and the more reasonable the segmentation method is.

[0104] Exemplarily, the absolute value of the difference between the ratio of the first quantity ni1 and the second quantity ni2 and 1 can be calculated and denoted as p, which can be expressed by the formula: p = |ni1 / ni2 - 1|. In the case where the target set is the longitude set, p1 = |n11 / n12 - 1| corresponding to the first mean value, p2 = |n21 / n22 - 1| corresponding to the second mean value, and p3 = |n31 / n32 - 1| corresponding to the third mean value can be calculated respectively, and then the mean value corresponding to the one closest to 0 among p1, p2, and p3 is determined as the target longitude mean value. Similarly, in the case where the target set is the latitude set, p4 = |n41 / n42 - 1| corresponding to the fourth mean value, p5 = |n51 / n52 - 1| corresponding to the fifth mean value, and p6 = |n61 / n62 - 1| corresponding to the sixth mean value can be calculated respectively, and then the mean value corresponding to the one closest to 0 among p4, p5, and p6 is determined as the target latitude mean value.

[0105] The parking point recognition method provided by the above embodiment calculates three mean values of three types of data, then groups the target combination according to the three mean values to obtain a first candidate combination and a second candidate combination, and determines the mean value corresponding to the ratio of the number of position data included in the first candidate combination and the second candidate combination being close to 1 as the target mean value, improving the balance of the re-segmentation of the target combination, keeping the number scales of the combinations for clustering recognition at a similar level, and being beneficial to the processing efficiency of clustering recognition.

[0106] In one embodiment, step S30 of clustering the position data in each combination to obtain multiple sets may include: the step of clustering the position data in each combination using a preset density-based spatial clustering of applications with noise (DBSCAN) algorithm to obtain multiple sets. Exemplarily, possible values of the parameters of the DBSCAN algorithm (such as the minimum number of samples, neighborhood radius) can be set based on the actual scenario and business experience, and then sample data is selected to select the parameter values under the optimal silhouette coefficient using the cross-validation method, so as to determine the algorithm parameters; at the same time, since the position data includes longitude parameters and latitude parameters, the Haversine distance, which is used to measure the actual distance between geographical locations, can be selected as the distance metric method. Based on this, the DBSCAN algorithm is used to perform clustering recognition on each combination, improving the parking point recognition efficiency and recognition accuracy.

[0107] For better understanding, as Figure 7 shown, another parking point recognition method is provided, and this parking point recognition method may include the following steps S71 to S75.

[0108] S71: Based on the timestamp, vehicle speed, longitude and latitude data in the vehicle networking data, use the preset parking point recognition method to obtain the location data of multiple vehicle parking points.

[0109] S72: Match the location data with the map data, divide each location data into each prefecture-level city, and obtain m initial combinations: G1, G2, ……, Gm.

[0110] S73: Set a preset threshold, count the number of location data in each initial combination and compare it with the preset threshold. If the number of location data in the target combination (Gi) is greater than the preset threshold (for example, 60000), then continue to divide the target combination until the number of location data in each combination is less than or equal to the preset threshold. As Figure 8 shown, the steps of dividing the target combination include the following S731 to S738:

[0111] S731: Obtain the longitude parameter (longitude) and latitude parameter (latitude) of each location data in the target combination.

[0112] S732: Truncate the longitude parameter and latitude parameter of each location data in the target combination, retain 2 decimal places after the decimal point, and obtain the truncated first longitude data (lons) and first latitude data (lats).

[0113] S733: Remove duplicates from the truncated first longitude data and first latitude data respectively, obtain the deduplicated second longitude data (lon) and second latitude data (lat), and calculate the number of longitudes (sum_lon) including longitude parameters in the second longitude data and the number of latitudes (sum_lat) including latitude parameters in the second latitude data.

[0114] S734: If sum_lon < sum_lat, determine the target direction as the longitude direction, that is, group the target combination along the longitude direction; otherwise, determine the target direction as the latitude direction, that is, group the target combination along the latitude direction.

[0115] S735: In the case where the target direction is the longitude direction, obtain the first mean (avg_lons) of the first longitude data, the second mean (avg_lon) of the second longitude data, and the third mean (avg_longitude) of the longitude parameters of each location data in the target combination; in the case where the target direction is the latitude direction, obtain the fourth mean (avg_lats) of the first latitude data, the fifth mean (avg_lat) of the second latitude data, and the sixth mean (avg_latitude) of the latitude parameters of each location data in the target combination.

[0116] S736: When the target direction is the longitude direction, group the target combination according to the first mean, the second mean, and the third mean, and respectively obtain the first candidate combination (N11) and the second candidate combination (N12) of the first mean, the first candidate combination (N21) and the second candidate combination (N22) of the second mean, and the first candidate combination (N31) and the second candidate combination (N32) of the third mean; when the target direction is the latitude direction, group the target combination according to the fourth mean, the fifth mean, and the sixth mean, and respectively obtain the first candidate combination (N41) and the second candidate combination (N42) of the fourth mean, the first candidate combination (N51) and the second candidate combination (N52) of the fifth mean, and the first candidate combination (N61) and the second candidate combination (N62) of the sixth mean.

[0117] S737: When the target direction is the longitude direction, respectively calculate the absolute value p1 of the difference between the ratio of the first quantity (n11) of the first candidate combination of the first mean and the second quantity (n12) of the second candidate combination and 1, the absolute value p2 of the difference between the ratio of the first quantity (n21) of the first candidate combination of the second mean and the second quantity (n22) of the second candidate combination and 1, and the absolute value p3 of the difference between the ratio of the first quantity (n31) of the first candidate combination of the third mean and the second quantity (n32) of the second candidate combination and 1; when the target direction is the latitude direction, respectively calculate the absolute value p4 of the difference between the ratio of the first quantity (n41) of the first candidate combination of the fourth mean and the second quantity (n42) of the second candidate combination and 1, the absolute value p5 of the difference between the ratio of the first quantity (n51) of the first candidate combination of the fifth mean and the second quantity (n52) of the second candidate combination and 1, and the absolute value p6 of the difference between the ratio of the first quantity (n61) of the first candidate combination of the sixth mean and the second quantity (n62) of the second candidate combination and 1.

[0118] S738: When the target direction is the longitude direction, select the mean corresponding to the value closest to 0 from p1, p2, and p3 as the target longitude mean, and use the target longitude mean as the grouping position to group the target combination along the longitude direction; when the target direction is the latitude direction, select the mean corresponding to the value closest to 0 from p4, p5, and p6 as the target latitude mean, and use the target latitude mean as the grouping position to group the target combination along the latitude direction.

[0119] S74: For the n combinations obtained after grouping: Group 1, Group 2, ……, Group n, use the preset DBSCAN algorithm to cluster each of the n combinations to obtain N sets: Class 1, Class 2, ……, Class N.

[0120] S75: For the N sets obtained by clustering, use reverse geocoding to obtain the location information of the centers Center1, Center2, ……, CenterN of the N sets. These N set centers are the clustering points of the vehicle parking points, thus completing the identification of the parking clustering areas.

[0121] It should be understood that although the steps in the flowcharts involved in the above-described embodiments are shown in sequence according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise clearly stated in this article, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders. Moreover, at least some of the steps in the flowcharts involved in the above-described embodiments may include multiple steps or multiple stages. These steps or stages are not necessarily executed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be executed alternately or in turn with at least some of the steps or stages in other steps or other steps.

[0122] Based on the same inventive concept, the embodiments of the present application also provide a parking point identification device for implementing the above-mentioned parking point identification method. The solution provided by this device to solve the problem is similar to the solution described in the above method. Therefore, the specific limitations in one or more of the following embodiments of the parking point identification device can refer to the limitations on the parking point identification method in the above text and will not be repeated here.

[0123] In one embodiment, as Figure 9 shown, a parking point identification device 900 is provided, including: a location module 901, a grouping module 902, a clustering module 903, and a center module 904. Among them, the location module 901 is used to obtain the location data of multiple vehicles at the parking points. The grouping module 902 is used to group the location data according to the location characteristics of each location data to obtain multiple combinations, where the number of location data included in each combination does not exceed a preset threshold. The clustering module 903 is used to cluster the location data in each combination to obtain multiple sets. The center module 904 is used to obtain the location information of the center of each set, where the center is the clustering point of vehicle parking.

[0124] The parking point recognition device provided by the above embodiments realizes the recognition of the vehicle parking point. Moreover, before clustering the location data, each location data is grouped based on the location features, and then the location data is clustered and recognized in units of the combinations obtained after grouping, avoiding directly clustering and recognizing the location data with a large amount of data, reducing the performance requirements for the hardware to process data, and also improving the recognition efficiency of the vehicle parking point. In addition, the location data in each combination obtained by grouping has similar location features, and clustering and recognizing the location data based on this grouping improves the accuracy of parking point recognition.

[0125] In one embodiment, the grouping module 902 is further configured to group each of the location data according to the geographical location features of each of the location data to obtain a plurality of initial combinations; group the location data in the target combination according to the longitude and latitude parameters of the location data in the target combination to obtain a plurality of candidate combinations; wherein, the target combination is a combination in which the number of location data included in a plurality of the initial combinations exceeds the preset threshold; when the number of location data included in the candidate combination exceeds the preset threshold, the candidate combination is used as the target combination for re-grouping until the number of location data in each combination after grouping is less than or equal to the preset threshold.

[0126] In one embodiment, the grouping module 902 is further configured to obtain the target mean of each location data in the target combination; wherein, the target mean includes one of the target longitude mean and the target latitude mean; group the location data in the target combination in the target direction according to the target mean to obtain two candidate combinations; wherein, the target direction includes one of the longitude direction and the latitude direction, and the longitude and latitude identifiers in the target mean and the target direction are the same.

[0127] In one embodiment, the parking point recognition device 900 may further include a truncation module, a duplicate removal module, a statistics module, and a determination module. Among them, the truncation module is configured to perform truncation processing on the longitude parameter and the latitude parameter of each location data in the target combination respectively to obtain the truncated first longitude data and the first latitude data; wherein, the number of decimal places of the first longitude data and the first latitude data does not exceed a preset length. The duplicate removal module is configured to perform duplicate removal processing on the first longitude data and the first latitude data respectively to obtain the duplicate-removed second longitude data and the second latitude data. The statistics module is configured to obtain the longitude number of the second longitude data and the latitude number of the second latitude data respectively. The determination module is configured to determine the target direction based on the direction corresponding to the target number; wherein, the target number is the smaller of the longitude number and the latitude number.

[0128] In one embodiment, the grouping module 902 is further configured to respectively obtain three means corresponding to three types of data in the target set; wherein, the target set is a longitude set or a latitude set, the target set is determined according to the target direction, the longitude set includes the first longitude data, the second longitude data, and the longitude parameters of each position data in the target combination, and the latitude set includes the first latitude data, the second latitude data, and the latitude parameters of each position data in the target combination; according to the three means corresponding to the three types of data in the target set, obtain the target mean of each position data in the target combination.

[0129] In one embodiment, the determining module is further configured to respectively group the position data in the target combination according to the target direction according to the three means of the three types of data in the target set, and obtain the corresponding first candidate combination and second candidate combination respectively; obtain the first quantity of the position data included in the first candidate combination and the second quantity of the position data included in the second candidate combination. The grouping module 902 is further configured to determine the mean corresponding to the target ratio as the target mean in the target set according to the ratio of the first quantity and the second quantity corresponding to each of the three means; wherein, the difference between the target ratio and 1 is the smallest.

[0130] In one embodiment, the clustering module 903 is further configured to cluster the position data in each combination by using a preset density-based spatial clustering of applications with noise (DBSCAN) algorithm to obtain a plurality of the sets.

[0131] Each module in the above parking point recognition device can be implemented in whole or in part by software, hardware, and their combination. The above modules can be embedded in the processor in the computer device in hardware form or independent of the processor, or stored in the memory in the computer device in software form, so that the processor can call and execute the operations corresponding to the above modules.

[0132] In one embodiment, a computer device is provided. The computer device can be a server, and its internal structure diagram can be as Figure 10 shown. The computer device includes a processor, a memory, and a network interface connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store position data. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, it implements a parking point recognition method.

[0133] Those skilled in the art can understand that Figure 10 The structure shown in Figure 10 is only a block diagram of some structures related to the solution of this application, and does not constitute a limitation on the computer device to which the solution of this application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine some components, or have different component arrangements.

[0134] In one embodiment, a computer device is provided, including a memory and a processor. A computer program is stored in the memory. When the processor executes the computer program, the following steps are implemented:

[0135] Obtain the position data of multiple vehicles at the parking point;

[0136] Group the position data according to the position characteristics of each position data to obtain multiple combinations; wherein, the number of position data included in each combination does not exceed a preset threshold;

[0137] Cluster the position data in each combination to obtain multiple sets;

[0138] Obtain the position information of the center point of each set; wherein, the center point is the aggregation point where the vehicle parks.

[0139] In one embodiment, when the processor executes the computer program, the following steps are further implemented: Group the position data according to the geographical location characteristics of each position data to obtain multiple initial combinations; Group the position data in the target combination according to the longitude and latitude parameters of each position data in the target combination to obtain multiple candidate combinations; wherein, the target combination is a combination in which the number of position data included in multiple initial combinations exceeds the preset threshold; In the case where the number of position data included in the candidate combination exceeds the preset threshold, use the candidate combination as the target combination for re-grouping until the number of position data in each combination after grouping is less than or equal to the preset threshold.

[0140] In one embodiment, when the processor executes the computer program, the following steps are further implemented: Obtain the target mean of each position data in the target combination; wherein, the target mean includes one of the target longitude mean and the target latitude mean; Group the position data in the target combination according to the target direction according to the target mean to obtain two candidate combinations; wherein, the target direction includes one of the longitude direction and the latitude direction, and the longitude and latitude identifiers in the target mean and the target direction are the same.

[0141] In one embodiment, when the processor executes the computer program, the following steps are further implemented: truncate the longitude parameter and the latitude parameter of each position data in the target combination respectively to obtain the truncated first longitude data and the first latitude data; wherein, the number of digits after the decimal point of the first longitude data and the first latitude data does not exceed a preset length; perform a deduplication process on the first longitude data and the first latitude data respectively to obtain the deduplicated second longitude data and the second latitude data; obtain the number of longitudes of the second longitude data and the number of latitudes of the second latitude data respectively; determine the target direction based on the smaller number of the number of longitudes and the number of latitudes; wherein, the target number is the smaller number of the number of longitudes and the number of latitudes.

[0142] In one embodiment, when the processor executes the computer program, the following steps are further implemented: obtain three means corresponding to three types of data in the target set respectively; wherein, the target set is a longitude set or a latitude set, the target set is determined according to the target direction, the longitude set includes the first longitude data, the second longitude data and the longitude parameters of each position data in the target combination, and the latitude set includes the first latitude data, the second latitude data and the latitude parameters of each position data in the target combination; obtain the target mean of each position data in the target combination according to the three means corresponding to the three types of data in the target set.

[0143] In one embodiment, when the processor executes the computer program, the following steps are further implemented: group the position data in the target combination according to the target direction respectively according to the three means of the three types of data in the target set to obtain the corresponding first candidate combination and second candidate combination respectively; obtain the first quantity of the position data included in the first candidate combination and the second quantity of the position data included in the second candidate combination; the step of obtaining the target mean of each position data in the target combination according to the three means corresponding to the three types of data in the target set includes: determining the mean corresponding to the target ratio in the target set as the target mean according to the ratio of the first quantity and the second quantity corresponding to each of the three means; wherein, the difference between the target ratio and 1 is the smallest.

[0144] In one embodiment, when the processor executes the computer program, the following steps are further implemented: cluster the position data in each combination by using a preset density clustering algorithm with noise to obtain a plurality of sets.

[0145] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the following steps are implemented:

[0146] Obtain the position data of multiple vehicles at the parking point;

[0147] Group the position data according to the position characteristics of each piece of position data to obtain multiple combinations; wherein, the number of position data included in each combination does not exceed a preset threshold;

[0148] Cluster the position data in each combination to obtain multiple sets;

[0149] Obtain the position information of the center point of each set; wherein, the center point is the aggregation point where the vehicle parks.

[0150] In one embodiment, when the computer program is executed by the processor, the following steps are further implemented: Group the position data according to the geographical location characteristics of each piece of position data to obtain multiple initial combinations; Group the position data in the target combination according to the longitude and latitude parameters of each piece of position data in the target combination to obtain multiple candidate combinations; wherein, the target combination is a combination in which the number of position data included in multiple initial combinations exceeds the preset threshold; In the case where the number of position data included in the candidate combination exceeds the preset threshold, use the candidate combination as the target combination for re-grouping until the number of position data in each combination after grouping is less than or equal to the preset threshold.

[0151] In one embodiment, when the computer program is executed by the processor, the following steps are further implemented: Obtain the target mean value of each piece of position data in the target combination; wherein, the target mean value includes one of the target longitude mean value and the target latitude mean value; Group the position data in the target combination according to the target direction according to the target mean value to obtain two candidate combinations; wherein, the target direction includes one of the longitude direction and the latitude direction, and the longitude and latitude identifiers in the target mean value and the target direction are the same.

[0152] In one embodiment, when the computer program is executed by the processor, the following steps are further implemented: Perform truncation processing on the longitude parameter and the latitude parameter of each piece of position data in the target combination to obtain the truncated first longitude data and the first latitude data; wherein, the number of digits after the decimal point of the first longitude data and the first latitude data does not exceed a preset length; Perform duplicate removal processing on the first longitude data and the first latitude data respectively to obtain the duplicate-removed second longitude data and the second latitude data; respectively obtain the longitude number of the second longitude data and the latitude number of the second latitude data; Determine the target direction according to the direction corresponding to the target number; wherein, the target number is the smaller of the longitude number and the latitude number.

[0153] In one embodiment, when the computer program is executed by a processor, the following steps are further implemented: respectively obtaining three means corresponding to three types of data in the target set; wherein, the target set is a longitude set or a latitude set, the target set is determined according to the target direction, the longitude set includes the first longitude data, the second longitude data, and the longitude parameters of each position data in the target combination, and the latitude set includes the first latitude data, the second latitude data, and the latitude parameters of each position data in the target combination; obtaining the target mean of each position data in the target combination according to the three means corresponding to the three types of data in the target set.

[0154] In one embodiment, when the computer program is executed by a processor, the following steps are further implemented: respectively grouping the position data in the target combination according to the target direction according to the three means of the three types of data in the target set, and obtaining the corresponding first candidate combination and second candidate combination respectively; obtaining the first quantity of the position data included in the first candidate combination and the second quantity of the position data included in the second candidate combination; the step of obtaining the target mean of each position data in the target combination according to the three means corresponding to the three types of data in the target set includes: determining the mean corresponding to the target ratio as the target mean in the target set according to the ratio of the first quantity and the second quantity corresponding to each of the three means; wherein, the difference between the target ratio and 1 is the smallest.

[0155] In one embodiment, when the computer program is executed by a processor, the following steps are further implemented: clustering the position data in each combination by using a preset density clustering algorithm with noise to obtain a plurality of the sets.

[0156] In one embodiment, a computer program product is provided, including a computer program, and when the computer program is executed by a processor, the following steps are implemented:

[0157] Obtaining the position data of multiple vehicles at the parking points;

[0158] Grouping each of the position data according to the position characteristics of each of the position data to obtain a plurality of combinations; wherein, the number of the position data included in each combination does not exceed a preset threshold;

[0159] Clustering the position data in each combination to obtain a plurality of sets;

[0160] Obtaining the position information of the center point of each of the sets; wherein, the center point is the aggregation point where the vehicles park.

[0161] In one embodiment, when the computer program is executed by a processor, the following steps are further implemented: grouping the location data according to the geographical location features of the location data to obtain a plurality of initial combinations; grouping the location data in the target combination according to the longitude and latitude parameters of the location data in the target combination to obtain a plurality of candidate combinations; wherein the target combination is a combination in which the number of location data included in the plurality of initial combinations exceeds the preset threshold; when the number of location data included in the candidate combination exceeds the preset threshold, re-grouping the candidate combination as the target combination until the number of location data in each group after grouping is less than or equal to the preset threshold.

[0162] In one embodiment, when the computer program is executed by a processor, the following steps are further implemented: obtaining the target mean of each location data in the target combination; wherein the target mean includes one of the target longitude mean and the target latitude mean; grouping the location data in the target combination according to the target direction according to the target mean to obtain two candidate combinations; wherein the target direction includes one of the longitude direction and the latitude direction, and the longitude and latitude identifiers in the target mean and the target direction are the same.

[0163] In one embodiment, when the computer program is executed by a processor, the following steps are further implemented: respectively performing truncation processing on the longitude parameter and the latitude parameter of each location data in the target combination to obtain the truncated first longitude data and the first latitude data; wherein the number of decimal places of the first longitude data and the first latitude data does not exceed the preset length; respectively performing deduplication processing on the first longitude data and the first latitude data to obtain the deduplicated second longitude data and the second latitude data; respectively obtaining the number of longitudes of the second longitude data and the number of latitudes of the second latitude data; determining the target direction according to the direction corresponding to the target number; wherein the target number is the smaller value of the number of longitudes and the number of latitudes.

[0164] In one embodiment, when the computer program is executed by a processor, the following steps are further implemented: respectively obtaining three means corresponding to three types of data in the target set; wherein the target set is a longitude set or a latitude set, the target set is determined according to the target direction, the longitude set includes the first longitude data, the second longitude data, and the longitude parameters of each location data in the target combination, and the latitude set includes the first latitude data, the second latitude data, and the latitude parameters of each location data in the target combination; obtaining the target mean of each location data in the target combination according to the three means corresponding to the three types of data in the target set.

[0165] In one embodiment, when the computer program is executed by a processor, the following steps are further implemented: according to the three means of the three types of data in the target set, the position data in the target combination are grouped in the target direction respectively, and the corresponding first candidate combination and second candidate combination are obtained; obtaining a first quantity of the position data included in the first candidate combination and a second quantity of the position data included in the second candidate combination; the obtaining of the target mean of each position data in the target combination according to the three means corresponding to the three types of data in the target set includes: determining, in the target set, the mean corresponding to the target ratio as the target mean according to the ratio of the first quantity and the second quantity corresponding to each of the three means; wherein, the difference between the target ratio and 1 is the smallest.

[0166] In one embodiment, when the computer program is executed by a processor, the following steps are further implemented: clustering the position data in each combination by using a preset density clustering algorithm with noise to obtain a plurality of the sets.

[0167] It should be noted that the data involved in this application (including but not limited to the data for analysis, the stored data, the displayed data, etc.) are all data authorized by the user or fully authorized by all parties.

[0168] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, database, or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memories. Non-volatile memories can include read-only memory (ROM), magnetic tapes, floppy disks, flash memories, optical memories, high-density embedded non-volatile memories, resistive random-access memories (ReRAM), magnetoresistive random-access memories (MRAM), ferroelectric random-access memories (FRAM), phase change memories (PCM), graphene memories, etc. Volatile memories can include random access memory (RAM) or external cache memories, etc. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc. The databases involved in the embodiments provided in the present application can include at least one of relational databases and non-relational databases. Non-relational databases can include distributed databases based on blockchain, etc., without limitation. The processors involved in the embodiments provided in the present application can be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logics, data processing logics based on quantum computing, etc., without limitation.

[0169] The technical features of the above embodiments can be combined arbitrarily. For the sake of concise description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope described in this specification.

[0170] The above-described embodiments only represent several implementation manners of the present application. The description is relatively specific and detailed, but it should not be construed as a limitation on the patent scope of the present application. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present application, several modifications and improvements can still be made, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the appended claims.

Claims

1. A method for identifying a parking point, characterized in that, Including: Obtain the position data of multiple vehicles at the parking point; Group the position data according to the position characteristics of each position data to obtain multiple combinations; wherein, the number of position data included in each combination does not exceed a preset threshold; the position characteristics include geographical location characteristics; group the position data according to the geographical location characteristics of each position data to obtain multiple initial combinations; group the position data in the target combination according to the longitude and latitude parameters of each position data in the target combination to obtain multiple candidate combinations; wherein, the target combination is a combination in which the number of position data included in multiple initial combinations exceeds the preset threshold; obtain the target mean of each position data in the target combination; wherein, the target mean includes one of the target longitude mean and the target latitude mean; group the position data in the target combination according to the target direction according to the target mean to obtain two candidate combinations; wherein, the target direction includes one of the longitude direction and the latitude direction, and the longitude and latitude identifiers in the target mean and the target direction are the same; in the case where the number of position data included in the candidate combination exceeds the preset threshold, use the candidate combination as the target combination for re-grouping until the number of position data in each combination after grouping is less than or equal to the preset threshold; Cluster the position data in each combination to obtain multiple sets; Obtain the position information of the center point of each set; wherein, the center point is the aggregation point where the vehicle parks.

2. The parking point identification method according to claim 1, characterized in that Before obtaining the target mean of each position data in the target combination, the parking point recognition method further includes: Perform truncation processing on the longitude parameter and the latitude parameter of each position data in the target combination respectively to obtain the truncated first longitude data and first latitude data; wherein, the number of decimal places of the first longitude data and the first latitude data does not exceed a preset length; Perform de-duplication processing on the first longitude data and the first latitude data respectively to obtain the de-duplicated second longitude data and second latitude data; Obtain the number of longitudes of the second longitude data and the number of latitudes of the second latitude data respectively; Determine the target direction based on the direction corresponding to the target number; wherein, the target number is the decimal of the number of longitudes and the number of latitudes.

3. The parking spot identification method according to claim 2, wherein, The obtaining of the target mean of each position data in the target combination includes: Obtain three means corresponding to three types of data in the target set respectively; wherein, the target set is a longitude set or a latitude set, the target set is determined according to the target direction, the longitude set includes the first longitude data, the second longitude data and the longitude parameters of each position data in the target combination, and the latitude set includes the first latitude data, the second latitude data and the latitude parameters of each position data in the target combination; Obtain the target mean of each position data in the target combination according to the three means corresponding to the three types of data in the target set.

4. The parking point recognition method according to claim 3, characterized in that, The parking point recognition method further includes: According to the three means of the three types of data in the target set, group the position data in the target combination in the target direction respectively to obtain the corresponding first candidate combination and second candidate combination; Obtain the first quantity of the position data included in the first candidate combination and the second quantity of the position data included in the second candidate combination; The obtaining of the target mean of each position data in the target combination according to the three means corresponding to the three types of data in the target set includes: According to the ratio of the first quantity and the second quantity corresponding to each of the three means, determine the mean corresponding to the target ratio in the target set as the target mean; wherein, the difference between the target ratio and 1 is the smallest.

5. The parking point recognition method according to claim 1, characterized in that, The clustering of the position data in each combination to obtain multiple sets includes: Use the preset density-based spatial clustering of applications with noise (DBSCAN) algorithm to cluster the position data in each combination to obtain multiple sets.

6. A parking spot recognition device, characterized in that, Includes: A position module, configured to obtain the position data of multiple vehicles at the parking point; A grouping module, configured to group the position data according to the position characteristics of each position data to obtain multiple combinations; wherein, the number of position data included in each combination does not exceed a preset threshold; the position characteristics include geographical location characteristics; A clustering module, configured to cluster the position data in each combination to obtain multiple sets; A center module, configured to obtain the position information of the center point of each set; wherein, the center point is the aggregation point where the vehicles park; The grouping module is further configured to group the position data according to the geographical location characteristics of each position data to obtain multiple initial combinations; group the position data in the target combination according to the longitude and latitude parameters of each position data in the target combination to obtain multiple candidate combinations; wherein, the target combination is a combination in which the number of position data included in multiple initial combinations exceeds the preset threshold; when the number of position data included in the candidate combination exceeds the preset threshold, use the candidate combination as the target combination for re-grouping until the number of position data in each grouped combination is less than or equal to the preset threshold; The grouping module is further configured to obtain the target mean of each position data in the target combination; wherein, the target mean includes one of a target longitude mean and a target latitude mean; group the position data in the target combination in the target direction according to the target mean to obtain two candidate combinations; wherein, the target direction includes one of the longitude direction and the latitude direction, and the longitude and latitude identifiers in the target mean and the target direction are the same.

7. A computer device, comprising a memory and a processor, the memory storing a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 5.

8. A computer storage medium, on which a computer program is stored, characterized in that, When the computer program is executed by the processor, it implements the steps of the method according to any one of claims 1 to 5.

Images