A method for converting mass coordinate data based on gaode POI data

Abstract

The application discloses a kind of mass coordinate data conversion methods based on high gao POI data, comprising: based on the principle of quadtree index optimization rectangular search algorithm, and obtain POI data of research area;Research area is gridded, constructs research area grid point coordinate database, POI coordinate point is matched with database to determine the WGS-84 coordinate of POI coordinate point, and the WGS-84 coordinate of POI coordinate point that cannot be matched is obtained using bilinear interpolation method;Define chinese address element annotation set, use hidden Markov model to carry out word segmentation processing to source address information in POI data;The first four layers of address information after word segmentation processing are parsed using Trie tree model, and the remaining six layers of address information are parsed using state automaton, to finally obtain complete address information.The application expands the threshold of POI data, so that the acquired area information is more comprehensive and accurate;Effectively solve the deviation problem between the value obtained by gao API interface and the true value, realize the accurate correspondence between longitude and latitude value and real geographical position.

Description

Technical Field

[0001] This invention relates to the field of transformer area address information parsing technology, specifically a method for converting massive coordinate data based on Gaode POI data. Background Technology

[0002] The vigorous promotion of smart grid construction and the continuous growth of information and communication scale have placed higher demands on the refined management of the distribution network. However, low-voltage distribution areas still face problems such as a large user base, complex network structure, and untimely updates of user distribution area affiliations.

[0003] Points of Interest (POIs) are one of the main data sources for geographic information interaction in urban information management systems. With the booming development of internet maps, more and more scientific research fields are beginning to incorporate internet maps into various information management systems. Currently, all internet map service providers in China, including Gaode Maps, use the GCJ-02 coordinate system defined by the State Bureau of Surveying and Mapping for data encryption. Furthermore, the amount of POI data obtained using the rectangular search method through Gaode's Application Programming Interface (API) has a limit. Without coordinate data conversion and overcoming this data limit, the obtained POI data will be incomplete and deviate from reality. In the process of resolving user addresses, the current Chinese word segmentation tool jieba is unsatisfactory in parsing road names, easily leading to the loss of key information during calculations and statistics. To solve these problems, this invention proposes a massive coordinate data conversion method based on Gaode POI data. Summary of the Invention

[0004] The purpose of this section is to outline some aspects of embodiments of the present invention and to briefly describe some preferred embodiments. Simplifications or omissions may be made in this section, as well as in the abstract and title of this application, to avoid obscuring the purpose of these documents; however, such simplifications or omissions should not be construed as limiting the scope of the invention.

[0005] In view of the above-mentioned problems, the present invention is proposed.

[0006] Therefore, the technical problem solved by this invention is that the latitude and longitude of existing open-source user data deviate from the actual values, resulting in inaccurate user address identification.

[0007] To solve the above-mentioned technical problems, the present invention provides the following technical solution: a method for converting massive coordinate data based on Gaode POI data, comprising:

[0008] The rectangular search algorithm of the Gaode API interface is optimized based on the quadtree index principle, and the optimized Gaode API interface is used to obtain POI data of the research area.

[0009] The study area is gridded to construct a database of grid point coordinates. The POI coordinates are matched with the database to determine the WGS-84 coordinates of the POI coordinates. The bilinear interpolation method is used to obtain the WGS-84 coordinates of the POI coordinates that cannot be matched.

[0010] Define a set of Chinese address elements and use a hidden Markov model to segment the source address information in the POI data.

[0011] The first four levels of address information after word segmentation are parsed using a Trie tree model, and the remaining six levels of address information are parsed using a state automaton to finally obtain the complete address information.

[0012] As a preferred embodiment of the massive coordinate data conversion method based on Gaode POI data described in this invention, the quadtree index principle includes: when the number of POIs satisfying the conditions in the divided region is less than the threshold, the region will not be further divided; when the number of POIs satisfying the conditions in the region is higher than the threshold, the region will continue to be divided into four parts until the number of POIs satisfying the conditions in all sub-regions is less than the threshold, at which point the division stops.

[0013] As a preferred embodiment of the massive coordinate data conversion method based on Gaode POI data described in this invention, the quadtree index principle optimizes the Gaode API interface, including: continuously dividing the research area into four parts, recursively until the number of POIs in all square areas is lower than the upper limit of acquisition, and using the number of polygonal POIs obtained at this time as the threshold for acquiring the number of POIs.

[0014] As a preferred embodiment of the massive coordinate data conversion method based on Gaode POI data described in this invention, the POI data includes coordinate data and geographic location information data.

[0015] As a preferred embodiment of the massive coordinate data transformation method based on Gaode POI data described in this invention, the step of gridding the study area includes: establishing a latitude and longitude interval of [missing information] in the GCJ-02 latitude and longitude coordinate system. The specification network grids the rectangular study area and ensures that the latitude and longitude satisfy the following conditions:

[0016]

[0017] in, These are the coordinates of the bottom right corner of the rectangular area; These are the coordinates of the top-left corner of the rectangular area.

[0018] As a preferred embodiment of the massive coordinate data conversion method based on Gaode POI data described in this invention, the construction of the research area grid point coordinate database includes:

[0019] First, the grid points are represented in the GCJ-02 coordinate system as follows: ;in, and These represent the row and column numbers, respectively.

[0020] Secondly, the Gaode API interface is used to convert the divided latitude and longitude grid points into coordinate points in WGS-84 coordinates. At this point, the grid points are represented as: ;

[0021] Secondly, add a three-digit hexadecimal code consisting of row and column numbers to the grid point coordinates. The grid points are numbered as follows: ;in, The number is represented as: ;

[0022] Finally, the coordinate data of each grid point is recorded and stored in the database to construct the grid point coordinate database.

[0023] As a preferred embodiment of the massive coordinate data conversion method based on Gaode POI data described in this invention, the step of obtaining the WGS-84 coordinates of unmatched POI coordinate points using bilinear interpolation includes: if the POI coordinate data fails to match in the grid point database, obtaining the row and column numbers of the nearest grid point to that point. The acquisition method can be represented as:

[0024]

[0025] Then obtain the row and column numbers of the grid points on the diagonal of the nearest grid point, represented as follows: ;

[0026] Using bilinear interpolation to determine the coordinates of the POI Convert to WGS-84 coordinates The calculation formula is expressed as:

[0027]

[0028] in, The GCJ-02 coordinates represent the coordinates of the nearest grid point. The GCJ-02 coordinates represent the coordinates of grid points on the diagonal. The WGS-84 coordinates representing the coordinates of the nearest grid point. The WGS-84 coordinates of the grid points on the diagonal.

[0029] As a preferred embodiment of the massive coordinate data conversion method based on Gaode POI data described in this invention, the word segmentation process includes: using the four states B, E, M, and S in the Hidden Markov Model to segment the source address, storing the segmentation results in a list, and extracting address elements at each level for address information matching and parsing.

[0030] As a preferred embodiment of the massive coordinate data conversion method based on Gaode POI data described in this invention, the method of parsing using a Trie tree model includes: establishing a standard address Trie tree and storing the administrative region code of the address at the last node of each branch; writing string data of the address to be matched, and determining the matching entry of the Trie tree based on the first character of the string; performing matching from the index according to the state transition function, and obtaining the administrative region code when the state ends, indicating successful address matching; when state transition is not possible, address matching fails; completing the address matching operation; and deleting the first four layers of address information matching fields from the word segmentation result list.

[0031] As a preferred embodiment of the massive coordinate data conversion method based on Gaode POI data described in this invention, the step of using a state automaton for parsing includes: merging the remaining word segments in the word segmentation result list after parsing by the Trie tree model, and inputting them into a finite state automaton. Perform address information matching for the last six levels and obtain address resolution results; store the matching results of the Trie tree model into the first four address levels in order, store the matching results of the finite state automaton into the last six address levels in order, and finally obtain the resolution results of the ten address levels and output the address information.

[0032] in, A finite set; The state corresponding to the address level; express and direct product The mapping represents the rules for transitioning between different states, also known as the state transition function; This is the initial state of the state machine; This represents the set of termination states of M.

[0033] The beneficial effects of this invention are as follows: This invention proposes a massive coordinate data conversion method based on Gaode POI data. By introducing the quadtree index principle to optimize the rectangle search method of the polygon search part in the Gaode POI interface, the threshold of POI data is expanded, making the obtained station area information more comprehensive and accurate. The bilinear interpolation method is used to realize the WGS-84 coordinate conversion of POI data, effectively solving the deviation problem between the value obtained from the Gaode API interface and the actual value, and realizing the accurate correspondence between latitude and longitude values ​​and the actual geographical location. The hidden Markov model is used to perform word segmentation processing on the source address information in the POI data, and the parsing of the POI source address information is realized based on Tire trees and automata. Finally, address information matching the actual geographical location is obtained, and a database for address matching is formed, which improves the accuracy and efficiency of address matching. Attached Figure Description

[0034] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. Wherein:

[0035] Figure 1 This is an overall flowchart of a method for converting massive coordinate data based on Gaode POI data, provided in one embodiment of the present invention.

[0036] Figure 2 A schematic diagram of a quadtree index for a massive coordinate data conversion method based on Gaode POI data provided in one embodiment of the present invention;

[0037] Figure 3 An example diagram of administrative region addresses using a Trie tree structure, provided as an embodiment of the present invention, illustrates a method for converting massive coordinate data based on Gaode POI data.

[0038] Figure 4 An example diagram of the state transition structure of a finite state automaton for a massive coordinate data transformation method based on Gaode POI data, provided in one embodiment of the present invention;

[0039] Figure 5 A sampling example diagram of a massive coordinate data conversion method based on Gaode POI data provided in an embodiment of the present invention;

[0040] Figure 6 This is an example diagram of coordinate information for a massive coordinate data conversion method based on Gaode POI data, provided as an embodiment of the present invention. Detailed Implementation

[0041] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below 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. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of the present invention.

[0042] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0043] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0044] This invention is described in detail with reference to the schematic diagrams. When detailing the embodiments of this invention, for ease of explanation, the cross-sectional views illustrating the device structure may be partially enlarged, not adhering to the usual scale. Furthermore, the schematic diagrams are merely examples and should not be construed as limiting the scope of protection of this invention. In actual fabrication, the three-dimensional spatial dimensions of length, width, and depth should be included.

[0045] Furthermore, in the description of this invention, it should be noted that the terms "upper," "lower," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. These terms are used solely for the convenience of describing the invention and for simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the invention. In addition, the terms "first," "second," or "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0046] Unless otherwise explicitly specified and limited, the terms "installation," "connection," and "joining" in this invention should be interpreted broadly. For example, they can refer to fixed connections, detachable connections, or integral connections; similarly, they can refer to mechanical connections, electrical connections, or direct connections, or indirect connections through an intermediate medium, or internal connections between two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0047] Example 1

[0048] Reference Figures 1-4As an embodiment of the present invention, a method for converting massive coordinate data based on Gaode POI data is provided, including:

[0049] S1: Optimize the rectangular search method of Gaode API interface based on the quadtree index principle, and use the optimized Gaode API interface to obtain POI data of the research area;

[0050] Specifically, such as Figure 2 As shown, the quadtree index principle is introduced, and the research area is continuously divided into four parts, recursively until the number of POIs in all square areas is lower than the upper limit of acquisition. The number of polygon POIs obtained at this time is used as the threshold for acquiring the number of POIs.

[0051] It should be noted that when the number of POIs meeting the criteria in a divided region is less than the threshold, the region will not be further divided; when the number of POIs meeting the criteria in a region is greater than the threshold, the region will continue to be divided into four parts until the number of POIs meeting the criteria in all sub-regions is less than the threshold, at which point the division will stop.

[0052] It should also be noted that the existing Gaode Map API interface provides four methods: keyword search, surrounding search, polygon search, and ID query. However, the rectangle search method in polygon search has a limited number of POIs, which cannot comprehensively and accurately obtain POI data of the study area. Therefore, this embodiment proposes to use a quadtree index to optimize the rectangle search method in order to obtain more comprehensive and accurate geographic information data of the study area.

[0053] Furthermore, the optimized Amap API interface was used to obtain POI data for the study area, including coordinate data and address information.

[0054] S2: Grid the study area, construct a grid point coordinate database for the study area, match the POI coordinates with the coordinate database to determine the WGS-84 coordinates of the POI coordinates, and use bilinear interpolation to obtain the WGS-84 coordinates of the POI coordinates that cannot be matched.

[0055] Furthermore, in the GCJ-02 latitude and longitude coordinate system, a latitude and longitude interval of [missing information] is established. The specification network grids the rectangular study area so that the latitude and longitude satisfy the following conditions:

[0056]

[0057] in, These are the coordinates of the bottom right corner of the rectangular area; These are the coordinates of the top-left corner of the rectangular area.

[0058] Furthermore, the grid points in the GCJ-02 coordinate system are represented as follows: ;in, and These represent the row and column numbers, respectively.

[0059] Furthermore, the Gaode Map API interface is used to convert the divided latitude and longitude grid points into coordinate points in WGS-84 coordinates. The grid point coordinates are represented as: .

[0060] Furthermore, add to the grid point coordinates Number, represented as: ,in, The number is composed of a three-digit hexadecimal code consisting of a row number and a column number, represented as follows: .

[0061] Furthermore, the coordinate data of each grid point is recorded and stored in the database to construct a grid point coordinate database.

[0062] Furthermore, the collected POI coordinate data is input into the grid point database for matching. If the coordinate data point is in the grid point database, the database directly outputs its converted WGS-84 coordinates. If the coordinate data fails to match in the grid point database, the row and column numbers of the nearest grid point are obtained. The acquisition method can be represented as:

[0063] .

[0064] Furthermore, the row and column numbers of the grid points on the diagonal of the nearest grid point are represented as follows: Use bilinear interpolation to determine the POI coordinates. Convert to WGS-84 coordinates The calculation formula is expressed as:

[0065] .

[0066] It should be noted that the source address information in the POI data obtained using the Gaode API is represented in GCJ-02 coordinate form. However, the GCJ-02 coordinate system is a geographic coordinate system that has been encrypted, and its values ​​have a significant deviation from the actual geographic location. Therefore, it is necessary to convert the GCJ-02 coordinates to WGS-84 coordinates to resolve the discrepancy between the values ​​obtained from the Gaode API and the actual values, and to achieve an accurate correspondence between the latitude and longitude values ​​and the actual geographic location.

[0067] S3: Define a set of Chinese address elements and use a hidden Markov model to process the source address information in the POI data into words;

[0068] Furthermore, based on the principles of address representation and the common ways of describing addresses, the address hierarchy element annotation set is defined as shown in Table 1;

[0069] Table 1. Address Hierarchical Element Labeling

[0070] Address Element Hierarchy Remark First layer First-level administrative divisions; provinces, autonomous regions, municipalities directly under the central government, and special administrative regions Second floor Second-level administrative divisions; districts and counties under the jurisdiction of cities, prefectures, autonomous prefectures, leagues, and municipalities directly under the central government. Third layer Third-level administrative divisions; counties, municipal districts, county-level cities, banners Fourth floor Fourth-level administrative division; subdistrict, town, township Fifth floor The address is located on the road, street, alley, or village. Sixth floor The street number, address number, etc. Seventh floor Address, building number, etc. Eighth floor Address unit number, building area, etc. Ninth floor Address, apartment number, room number, etc. tenth floor Remarks

[0071] Furthermore, a hidden Markov model is used to segment the POI source address information, and the segmentation results are stored in a list for matching ten layers of address elements.

[0072] It is important to know that Hidden Markov Models have four states: B, E, M, and S, which represent the start position, end position, middle position, and position of the source address as a standalone word, respectively. The address elements of the last six levels are extracted according to the labels, and the extracted address elements are input into the automaton matching model together with the results of the Trie tree model.

[0073] It should be noted that, since the jieba word segmentation module used for conventional word segmentation processing does not achieve satisfactory accuracy in parsing road names during address parsing, this embodiment uses a hidden Markov model to segment the POI source address information, thereby effectively improving the accuracy of address parsing.

[0074] S4: Use the Trie tree model to parse the first four levels of address information after word segmentation, and use the state automaton to parse the remaining six levels of address information to finally obtain the complete address information.

[0075] Furthermore, refer to Figure 3 Establish a standard address Trie tree and store the administrative region code of the address at the last node of each branch; write the string data of the address to be matched, and determine the matching entry of the Trie tree based on the first character of the string; perform matching according to the state transition function starting from the index; when the state ends, obtain the administrative region code, and the address match is successful; when the state transition cannot be performed, the address match fails; complete the address matching operation; delete the first four levels of address information matching fields from the word segmentation result list.

[0076] It should be noted that the Trie tree structure mainly includes a root node, child nodes, and node values. The state of the automaton is expressed by the corresponding node values. In natural language processing, the Trie tree can establish a hierarchical data retrieval structure for parsing and matching the first four levels of administrative region address information of POI source address information.

[0077] Furthermore, the remaining word segments in the word segmentation result list are merged, input into a finite state automaton for the last six layers of address information matching, and the address resolution result is obtained.

[0078] It's important to know that the basic structure of a deterministic finite state automaton (DFA) is as follows:

[0079]

[0080] in, A finite set; The state corresponding to the address level; express and direct product The mapping represents the rules for transitioning between different states, also known as the state transition function; This is the initial state of the state machine; This represents the set of termination states of M.

[0081] Furthermore, the matching results of the Trie tree model are stored sequentially in the first four address levels, and the matching results of the finite state automaton are stored sequentially in the last six address levels. Finally, the parsing results of the ten address levels are obtained and the address information is output.

[0082] It should be noted that, as Figure 4 As shown, after the address information is segmented into words according to the Hidden Markov Model, the segmented words are input into the automaton. The automaton matches according to the predefined address hierarchy and performs state transitions according to the state transition rules. When the rule is met, the state transition is performed until all states have been traversed. The automaton completes the layer-by-layer matching of the address information and then combines them according to the hierarchical order, thus completing the parsing of the address information with high accuracy.

[0083] Furthermore, the latitude and longitude conversion results and the output address information are used as the final POI data conversion results as the output.

[0084] It should be noted that using a combination of Trie tree model and automata (FA) to parse the source address information in POI data can more accurately parse the road name information, thereby improving the positioning accuracy and ultimately enabling the establishment of a database of user address information for the transformer substation area, which is used to match user address information in the transformer substation area.

[0085] Example 2

[0086] Reference Figures 5-6 This invention provides a method for converting massive coordinate data based on Gaode POI data, as one embodiment of the present invention. To verify the beneficial effects of the present invention, a simulation experiment is conducted for scientific demonstration.

[0087] In this embodiment, the study area is located within the Third Ring Road of Wuhan City. The vehicle service tags within a rectangular area defined by longitudes 114.148085°E—114.456820°E and latitudes 30.457560°N—30.699075°N are used as the sampling objects. The sampling area and sample points are as follows: Figure 5 As shown, a control point database is established with the coordinates of the lower left corner of the rectangle at 114.148 085° E and 30.457 560° N.

[0088] Reference Figure 6 Python was used as the implementation tool to call the Gaode API interface to obtain the coordinate information of the points that meet the conditions. The coordinates of the sample points were then transformed to obtain the data table, as shown in Table 1.

[0089] Table 1 Address Data

[0090] Province prefecture-level cities County Streets, towns and townships road Road number Building number Unit number Account Number Remark Hubei Province Wuhan Wuchang District / Donghu Road No. 84 / / / / Hubei Province Wuhan Wuchang District Luojia Mountain Subdistrict Office Bayi Road No. 106 / / / / Hubei Province Wuhan Wuchang District / Tuanjie Road No. 18 / / / / Hubei Province Wuhan Wuchang District / Zhongbei Road No. 235 / / / Next to the PetroChina Hotel Hubei Province Wuhan Wuchang District East Lake Ecological Tourism Scenic Area Donghu Road No. 150 / / / / Hubei Province Wuhan Wuchang District / Zhongshan Road No. 264 / / / Appendix 4 Hubei Province Wuhan Wuchang District / Bapu Street Exit C of subway station / / / 200 meters Hubei Province Wuhan Wuchang District / Sanjiaolu Village Hubei University Qin Garden Community shops No. 12 / / …… …… …… …… …… …… …… …… …… ……

[0091] After completing the offline conversion of latitude and longitude coordinates, the geographical location descriptions represented by the coordinates need to be further processed. Using the T-FA model, the above-mentioned address source data is parsed and classified into the corresponding 10-level address label set. As can be seen from Table 1, the POI data conversion method of this invention can achieve accurate correspondence between latitude and longitude values ​​and real geographical locations. Further parsing of the POI source address information finally obtains address information that matches the real geographical location, forming a database for address matching, which improves the accuracy and efficiency of address matching and has high processing capacity and precision.

[0092] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for converting massive coordinate data based on Gaode POI data, characterized in that, include: The rectangular search algorithm of the Gaode application API interface is optimized based on the quadtree index principle, and the optimized Gaode API interface is used to obtain the POI data of the research area. The study area is gridded to construct a grid point coordinate database. Points of interest (POIs) are matched against the database to determine their WGS-84 coordinates. Bilinear interpolation is then used to obtain the WGS-84 coordinates of POIs that cannot be matched. Specifically, if a POI coordinate match fails in the grid point database, the row and column numbers of the nearest grid point are obtained. The acquisition method can be represented as: ； Then obtain the row and column numbers of the grid points on the diagonal of the nearest grid point, represented as follows: ; Use bilinear interpolation to determine the coordinates of points of interest (POIs). Convert to WGS-84 coordinates The calculation formula is expressed as: ； in, The GCJ-02 coordinates represent the coordinates of the nearest grid point. The GCJ-02 coordinates represent the coordinates of grid points on the diagonal. The WGS-84 coordinates representing the coordinates of the nearest grid point. WGS-84 coordinates representing the coordinates of grid points on the diagonal; Define a set of Chinese address elements and use a hidden Markov model to perform word segmentation on the source address information in the Point of Interest (POI) data. The first four levels of address information after word segmentation are parsed using a Trie tree model, and the remaining six levels of address information are parsed using a state automaton to finally obtain the complete address information.

2. The massive coordinate data conversion method based on Gaode POI data as described in claim 1, characterized in that: The principle of the quadtree index includes: when the number of points of interest (POIs) that meet the conditions in the partitioned region is less than the threshold, the region will not be further partitioned; when the number of points of interest that meet the conditions in the region is higher than the threshold, the region will continue to be divided into four parts until the number of points of interest that meet the conditions in all sub-regions is less than the threshold, at which point the division will stop.

3. The massive coordinate data conversion method based on Gaode POI data as described in claim 1 or 2, characterized in that: The quadtree indexing principle optimizes the Gaode application API interface, including: continuously dividing the research area into four parts, recursively until the number of points of interest (POIs) in all square areas is lower than the upper limit that can be obtained, and using the number of polygonal points of interest (POIs) obtained at this time as the threshold for obtaining the number of points of interest (POIs).

4. The massive coordinate data conversion method based on Gaode POI data as described in claim 3, characterized in that: The Point of Interest (POI) data includes coordinate data and geographic location information data.

5. The massive coordinate data conversion method based on Gaode POI data as described in claim 4, characterized in that: The process of gridding the study area includes: establishing a latitude and longitude interval of [missing information] in the GCJ-02 latitude and longitude coordinate system. The specification network grids the rectangular study area and ensures that the latitude and longitude satisfy the following conditions: ； in, These are the coordinates of the bottom right corner of the rectangular area; These are the coordinates of the top-left corner of the rectangular area.

6. The massive coordinate data conversion method based on Gaode POI data as described in claim 5, characterized in that: The construction of the grid point coordinate database for the study area includes: First, the grid points are represented in the GCJ-02 coordinate system as follows: ;in, and These represent the row and column numbers, respectively. Secondly, the Gaode API interface is used to convert the divided latitude and longitude grid points into coordinate points in WGS-84 coordinates. At this point, the grid points are represented as: ; Secondly, add a three-digit hexadecimal code consisting of row and column numbers to the grid point coordinates. The grid points are numbered as follows: ;in, The number is represented as: ; Finally, the coordinate data of each grid point is recorded and stored in the database to construct the grid point coordinate database.

7. The massive coordinate data conversion method based on Gaode POI data as described in claim 6, characterized in that: The word segmentation process includes: using the four states B, E, M, and S in the Hidden Markov Model to segment the source address, storing the segmentation results in a list, and extracting address elements at each level for address information matching and parsing.

8. The method for converting massive coordinate data based on Gaode POI data as described in claim 7, characterized in that: The process of parsing using a Trie tree model includes: establishing a standard address Trie tree and storing the administrative region code of the address at the last node of each branch; writing string data of the address to be matched and determining the matching entry point of the Trie tree based on the first character of the string; performing matching from the index according to the state transition function, and obtaining the administrative region code when the state ends, indicating a successful address match; when a state transition cannot be performed, the address match fails; completing the address matching operation; and deleting the first four layers of address information matching fields from the word segmentation result list.

9. The massive coordinate data conversion method based on Gaode POI data as described in claim 8, characterized in that: The process of using a state automaton for parsing includes: merging the remaining word segments in the word segmentation result list after parsing by the Trie tree model, and inputting the result into a finite state automaton. Perform address information matching for the last six levels and obtain address resolution results; store the matching results of the Trie tree model into the first four address levels in order, store the matching results of the finite state automaton into the last six address levels in order, and finally obtain the resolution results of the ten address levels and output the address information. in, A finite set; The state corresponding to the address level; express and direct product to The mapping represents the rules for transitioning between different states, also known as the state transition function; This is the initial state of the state machine; This represents the set of termination states of M.

Images