A method and system for data extraction based on kernel density analysis

Abstract

The present invention provides a method and system for data extraction based on kernel density analysis, wherein the method comprises: when the number of layers is 1, obtaining input parameters; using a single layer condition management table to retrieve the input parameters, if there are no identical records in the single layer condition management table, using a kernel density calculation model to obtain data extraction results of a target area, and storing the data extraction results in the single layer condition management table; if there are identical records in the single layer condition management table, finding the data extraction results of the target area according to the corresponding input parameters; when the number of layers is multiple layers, using a multi-layer target area extraction method to obtain the data extraction results of the target area. The present invention improves the operating efficiency and display efficiency of the kernel density analysis system by utilizing the condition management table to cache and use intermediate process data and result data.

Description

Technical Field

[0001] The present invention relates to the technical field of kernel density analysis, and in particular to a method and system for data extraction based on kernel density analysis. Background Art

[0002] With the introduction of modern science and technology, especially computer technology, into cartography and geography, geographic information systems (GIS) began to develop. Maps in digital form in computers have shown people a wider range of applications. Using computers to analyze maps, obtain information, and support spatial decision-making has become an important research content of geographic information systems, and the term "spatial analysis" has become a special term in this field. Spatial analysis mainly mines the potential information of spatial targets through the joint analysis of spatial data and spatial models. The basic information of these spatial targets is nothing more than their spatial position, distribution, morphology, distance, orientation, topological relationship, etc. Among them, distance, orientation, and topological relationship constitute the spatial relationship of spatial targets. It is the spatial characteristics between geographic entities and can be used as the basis for data organization, query, analysis, and reasoning. By dividing geographic spatial targets into different types of points, lines, and surfaces, the morphological structure of these different types of targets can be obtained. Combining the spatial data and attribute data of spatial targets can be used to perform spatial calculations and analyses for many specific tasks.

[0003] The existing geographic information system platform contains many relatively independent spatial analysis functions, such as density analysis (including point density, kernel density, etc.), overlay analysis (feature overlay, raster overlay, etc.), network analysis (path analysis, address matching, etc.), etc. It also contains many relatively independent data processing functions, such as reclassification, data conversion, coordinate projection and transformation, etc.

[0004] Kernel density estimation is used in probability theory to estimate unknown density functions. It is one of the non-parametric test methods. Solving the distribution density function of a random variable from a given set of sample points is one of the basic problems in probability statistics.

[0005] The Kernel Density tool calculates the density of features within their surrounding neighborhood. This tool can calculate density for both point and line features. Possible uses include analyzing housing density for community planning or exploring how roads or utility lines affect wildlife habitat. A weight field can be used to give some features more weight than others and also allows a single point to represent multiple observations. For example, a single address can represent a six-unit apartment. For line features, a divided highway may have a greater impact than a narrow dirt road.

[0006] At present, the various spatial analysis functions and data processing functions of the existing platforms are relatively independent, and most of them are manually operated and executed one-time. When faced with some complex business application scenarios, it is often necessary to operate many different tools or call many different services to achieve the desired results, and there is no effective organization and management of the intermediate process data and result data generated in each step. Summary of the invention

[0007] In order to overcome the deficiencies of the prior art, the object of the present invention is to provide a method and system for data extraction based on kernel density analysis.

[0008] A method for data extraction based on kernel density analysis, comprising:

[0009] Step 1: Get the number of layers contained in the target area;

[0010] Step 2: When the number of layers is 1, obtain input parameters; the input parameters include point layer data source connection information and filtering conditions, kernel density analysis parameters and reclassification parameters;

[0011] Step 3: Use the single layer condition management table to retrieve the input parameters. If there is no identical record in the single layer condition management table, use the kernel density calculation model to obtain the data extraction result of the target area, and store the data extraction result in the single layer condition management table;

[0012] Step 4: If there are identical records in the single layer condition management table, find the data extraction result of the target area according to the corresponding input parameters;

[0013] Step 5: When the number of layers is multiple, use the multi-layer target area extraction method to obtain the data extraction result of the target area.

[0014] Preferably, the kernel density calculation model in step 3 is:

[0015]

[0016] in, f ( x , y ) represents the density value of the prediction point, x is the longitude coordinate of the prediction point, y is the latitude coordinate of the prediction point, i =1,2,3.. n Indicates the input point, n is the number of input points, radius Indicates the predicted point ( x , y ) as the center of the search radius, Yes iThe weight field value of Yes i To the predicted point ( x , y ) distance.

[0017] Preferably, the step 5: when the number of layers is multiple layers, using a multi-layer target area extraction method to obtain a data extraction result of the target area includes:

[0018] Step 5.1: Obtain multi-layer input parameters; the multi-layer input parameters include multiple point layer data source connection information and filtering conditions, kernel density analysis parameters and reclassification parameters;

[0019] Step 5.2: Use the multi-layer condition management table to retrieve the multi-layer input parameters. If there is no identical record in the multi-layer condition management table, use the kernel density calculation model to obtain the data extraction result of the target area, and store the data extraction result in the multi-layer condition management table;

[0020] Step 5.3: If there are identical records in the multi-layer condition management table, the data extraction results of the target area are found according to the corresponding multi-layer input parameters.

[0021] Preferably, in step 5.2, the data extraction result of the target area is obtained by using the kernel density calculation model, including:

[0022] Step 5.2.1: Get point data based on the point layer data source connection information and filtering conditions;

[0023] Step 5.2.2: Input the point data and kernel density analysis parameters into the kernel density calculation model to obtain result grid data;

[0024] Step 5.2.3: reclassifying the result raster data to obtain classified result raster data;

[0025] Step 5.2.4: Convert the resulting raster data after the classification to a surface to obtain the data extraction result of the target area.

[0026] The present invention also provides a system for data extraction based on kernel density analysis, comprising:

[0027] The layer number acquisition module is used to obtain the number of layers contained in the target area;

[0028] A parameter acquisition module is used to acquire input parameters when the number of layers is 1; the input parameters include point layer data source connection information and filtering conditions, kernel density analysis parameters and reclassification parameters;

[0029] A retrieval module, used to retrieve the input parameter using the single layer condition management table, and if there is no identical record in the single layer condition management table, obtain the data extraction result of the target area using the kernel density calculation model, and store the data extraction result in the single layer condition management table;

[0030] A data extraction result acquisition module is used to find the data extraction result of the target area according to the corresponding input parameters when there are identical records in the single layer condition management table;

[0031] The multi-layer extraction module is used to obtain the data extraction result of the target area using the multi-layer target area extraction method when the number of layers is multiple.

[0032] Preferably, the kernel density calculation model in the retrieval module is:

[0033]

[0034] in, f ( x , y ) represents the density value of the prediction point, x is the longitude coordinate of the prediction point, y is the latitude coordinate of the prediction point, i =1,2,3.. n Indicates the input point, n is the number of input points, radius Indicates the predicted point ( x , y ) as the center of the search radius, Yes i The weight field value of Yes i To the predicted point ( x , y ) distance.

[0035] Preferably, the multi-layer extraction module includes:

[0036] A multi-layer input parameter acquisition unit is used to acquire multi-layer input parameters; the multi-layer input parameters include multiple point layer data source connection information and filtering conditions, kernel density analysis parameters and reclassification parameters;

[0037] A retrieval unit, used to retrieve the multi-layer input parameters using a multi-layer condition management table, and if there is no identical record in the multi-layer condition management table, obtain a data extraction result of the target area using a kernel density calculation model, and store the data extraction result in the multi-layer condition management table;

[0038] The result extraction unit is used to find the data extraction result of the target area according to the corresponding multi-layer input parameters when there are identical records in the multi-layer condition management table.

[0039] Preferably, the retrieval unit comprises:

[0040] The point data acquisition subunit is used to acquire point data according to the point layer data source connection information and filtering conditions;

[0041] A raster data calculation subunit, used for inputting the point data and kernel density analysis parameters into the kernel density calculation model to obtain result raster data;

[0042] A reclassification subunit, used for reclassifying the result raster data to obtain the classified result raster data;

[0043] The raster-to-surface subunit is used to perform raster-to-surface conversion on the result raster data after the classification to obtain the data extraction result of the target area.

[0044] The present invention also provides an electronic device, comprising a bus, a transceiver, a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the transceiver, the memory and the processor are connected via the bus, and wherein the computer program, when executed by the processor, implements the steps in the above-mentioned method for data extraction based on kernel density analysis.

[0045] The present invention also provides a computer-readable storage medium having a computer program stored thereon, characterized in that when the computer program is executed by a processor, the steps in the above-mentioned method for data extraction based on kernel density analysis are implemented.

[0046] The beneficial effect of the method and system for data extraction based on kernel density analysis provided by the present invention is that, compared with the prior art, the present invention improves the operating efficiency and display efficiency of the kernel density analysis system by utilizing the condition management table to cache and use intermediate process data and result data.

[0047] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, preferred embodiments are given below and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0049] Figure 1 A single-layer target area extraction flow chart provided by an embodiment of the present invention is shown;

[0050] Figure 2 A multi-layer target area extraction flow chart provided by an embodiment of the present invention is shown;

[0051] Figure 3 The functional block diagram of the timing execution strategy management module provided by the embodiment of the present invention is shown. DETAILED DESCRIPTION

[0052] In the description of the present invention, it should be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", "clockwise", "counterclockwise" and the like indicate orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the referred device or element must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as limiting the present invention.

[0053] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the present invention, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.

[0054] In the present invention, unless otherwise clearly specified and limited, the terms "installed", "connected", "connected", "fixed" and the like should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be an indirect connection through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0055] Please also read Figure 1-2 , a method for data extraction based on kernel density analysis, comprising:

[0056] Step 1: Get the number of layers contained in the target area;

[0057] Step 2: When the number of layers is 1, obtain input parameters; the input parameters include point layer data source connection information and filtering conditions, kernel density analysis parameters and reclassification parameters;

[0058] Step 3: Use the single layer condition management table to retrieve the input parameters. If there is no identical record in the single layer condition management table, use the kernel density calculation model to obtain the data extraction result of the target area, and store the data extraction result in the single layer condition management table;

[0059] In practical applications, kernel density calculation is used to calculate the density value of point features around the pixels in each raster result. In theory, each point feature will cover a smooth surface centered on the feature, with the highest value at the location of the point feature. As the distance to the point feature increases, this value will gradually decrease and become zero when it is equal to the search radius. The density value of each pixel in the raster result is the sum of the values ​​of all kernel density surfaces superimposed on this pixel position.

[0060] The kernel density calculation model in the embodiment of the present invention is:

[0061]

[0062] The kernel function uses the quartic kernel function, and the formula is:

[0063]

[0064] The final calculation model is:

[0065]

[0066] f ( x , y ) represents the density value of the prediction point (i.e., the point of the target area in the embodiment of the present invention), x is the longitude coordinate of the prediction point, y The latitude coordinate of the prediction point. Since it is a raster output, this coordinate is the coordinate of the pixel center point. i =1,2,3.. n Represents input points, i.e. point features; n is the number of input points, that is, the total number of point features; radius is the search radius, i.e. the prediction point ( x , y ) as the center of the search radius; Yes i The weight field value is optional. If not, it is 1. Yes i To the predicted point ( x , y ) distance.

[0067] Step 4: If there are identical records in the single layer condition management table, find the data extraction result of the target area according to the corresponding input parameters;

[0068] The following is a further description of the single-layer target area extraction process in conjunction with a specific embodiment:

[0069] See also Figure 1 ,The first step: receive input parameters, including point layer data source connection information and filtering conditions, kernel density analysis parameters (pixel size, analysis radius, weight field, etc.), reclassification parameters (classification level and threshold range corresponding to each level), and whether to regenerate;

[0070] Step 2: Use the received parameters to search the condition management table to see if there are records with the same conditions.

[0071] Step 3: If there is no record of the condition in the condition management table, it means that this is a new analysis, and a unique identifier is generated as the condition number (i.e. Figure 2 The key in the condition management table is stored together with the conditions, and a record is created in the condition result association table to record the condition number.

[0072] If there is a record of the condition in the condition management table, it means that the data of the same condition has been executed, and the condition number is obtained; determine whether it needs to be regenerated; if the input parameter is yes, delete the cached data under the associated number; then enter the fourth step and execute in sequence until the result is regenerated and returned. If the input parameter is no, find the target area result number by the condition number; determine whether the target area result number exists; if it exists, find the corresponding data in the target area result table and return it. If it does not exist, find the kernel density grid result number by the condition number; determine whether the kernel density grid result exists; if it exists, jump to the seventh step and execute in sequence until the result is regenerated and returned; if it does not exist, jump to the fourth step and execute in sequence until the result is regenerated and returned.

[0073] Step 4: Connect information and filter conditions with the data source of the base layer to obtain point data;

[0074] Step 5: Call the kernel density calculation model through point data and kernel density analysis parameters to obtain the resulting raster data;

[0075] Step 6: Generate a unique identifier as the kernel density result number, save it together with the path where the raster result file is stored in the kernel density raster cache table, and update it to the condition result association table through the condition number;

[0076] Step 7: Use the result raster data of kernel density analysis and the part about reclassification in the conditional parameters to call the third-party reclassification tool to obtain the result raster data after classification;

[0077] Step 8: Generate a unique identifier as the reclassification result number, save it together with the path where the raster result file is stored in the reclassification raster cache table, and update it to the condition result association table through the condition number;

[0078] Step 9: The result raster data processed by the reclassification tool calls the third-party raster-to-polygon tool to obtain the target area polygon data that has been reclassified and marked;

[0079] Step 10: Generate unique identifiers according to the levels of the identifiers during reclassification as the target area result numbers, and store them together with the target area surface data of the identifier in the target area extraction result table;

[0080] Step 11: The execution result and surface data are returned together, and the process execution ends.

[0081] Step 5: When the number of layers is multiple, use the multi-layer target area extraction method to obtain the data extraction result of the target area.

[0082] Further, the step 5 comprises:

[0083] Step 5.1: Obtain multi-layer input parameters; the multi-layer input parameters include multiple point layer data source connection information and filtering conditions, kernel density analysis parameters and reclassification parameters;

[0084] Step 5.2: Use the multi-layer condition management table to retrieve the multi-layer input parameters. If there is no identical record in the multi-layer condition management table, use the kernel density calculation model to obtain the data extraction result of the target area, and store the data extraction result in the multi-layer condition management table;

[0085] In the embodiment of the present invention, the data extraction result of the target area is obtained by using the kernel density calculation model in step 5.2, including:

[0086] Step 5.2.1: Get point data based on the point layer data source connection information and filtering conditions;

[0087] Step 5.2.2: Input the point data and kernel density analysis parameters into the kernel density calculation model to obtain result grid data;

[0088] Step 5.2.3: reclassifying the result raster data to obtain classified result raster data;

[0089] Step 5.2.4: Convert the resulting raster data after the classification to a surface to obtain the data extraction result of the target area.

[0090] Step 5.3: If there are identical records in the multi-layer condition management table, the data extraction results of the target area are found according to the corresponding multi-layer input parameters.

[0091] The following is a further description of the single-layer target area extraction process in conjunction with a specific embodiment:

[0092] See also Figure 2 ,The first step: receive input parameters, including multiple point layer data source connection information and filtering conditions, kernel density analysis parameters (pixel size, analysis radius, weight field, etc.), reclassification parameters (classification level and threshold range corresponding to each level), and whether to regenerate;

[0093] Step 2: Use the received parameters to search the condition management table to see if there are records with the same conditions.

[0094] Step 3: If there is no record of the condition in the condition management table, it means that this is a new analysis. Then a unique identifier is generated as the condition number, which is stored in the condition management table together with the conditions. A record is created in the condition result association table to record the condition number.

[0095] Step 4: Traverse the point data source configuration, and search the condition cache table in combination with other parameters to see whether the single layer has been processed under this condition;

[0096] Step 5: If no result is found, it means that the single layer has not been processed and it is the first execution. Then a unique identifier is generated as the condition number, which is stored in the condition management table together with the conditions. A record is created in the condition result association table to record the condition number (note that the record at this time is different from the third step. The record at this time is the single layer related number)

[0097] Step 6: Call the kernel density calculation model through point data and kernel density analysis parameters to obtain the resulting raster data;

[0098] Step 7: Generate a unique identifier as the kernel density result number, save it together with the path where the raster result file is stored in the kernel density raster cache table, and update it to the condition result association table through the condition number (note that it is the condition number of a single layer);

[0099] Step 8: After traversing all point layers and executing the kernel density analysis steps, call the third-party raster fusion tool together with the kernel density analysis raster results of each point layer to obtain the fused raster result data; the raster fusion process is to superimpose multiple grid data, and at the overlapping position, the values ​​of the overlapping pixels are summed up as the result value of the output pixel.

[0100] Step 9: Generate a unique identifier as the fusion result number, save it together with the path where the raster result file is stored in the fused raster cache table, and update it to the condition result association table through the condition number (note that the record where the condition number of multiple layers is located is updated at this time);

[0101] Step 10: The result raster data of raster fusion processing and the part about reclassification in the conditional parameters are used together to call the third-party reclassification tool to obtain the classified result raster data; raster reclassification is to configure several classification intervals to correspond to new values. Here, the application is to convert the kernel density value (decimal) interval into a severity level (integer) to facilitate the subsequent extraction of surface features by level. Generally, there are three types: 1 (severe), 2 (general), and Null (no data).

[0102] Step 11: Generate a unique identifier as the reclassification result number, save it together with the path where the raster result file is stored in the reclassification raster cache table, and update it to the condition result association table through the condition number;

[0103] Step 12: The result raster data processed by the reclassification tool calls the third-party raster to polygon tool to obtain the target area polygon data that has been finally reclassified and identified; raster to polygon feature is to extract the raster into line or polygon data through a field value in the raster. Here, it is applied to extract the polygon feature result composed of the numerical part from the raster data with the severity level identified after reclassification, which is generally classified as 1 (severe) and 2 (general).

[0104] Step 13: Generate unique identifiers according to the levels identified during reclassification as the target area result numbers, and store them together with the target area surface data of the identifier in the target area extraction result table;

[0105] Step 14: The execution result and surface data are returned together, and the process execution ends.

[0106] The advantages of the present invention are:

[0107] 1. The present invention connects various tools in series into a set of processing procedures, saving cumbersome operating steps;

[0108] 2. The present invention manages and reuses conditions, intermediate process data, and result data, saving processing time and making data organization orderly;

[0109] 3. The present invention can achieve the purpose of regularly updating the target area and ensure the timeliness of the data by combining with the automatic execution function.

[0110] The present invention also provides a system for data extraction based on kernel density analysis, comprising:

[0111] The layer number acquisition module is used to obtain the number of layers contained in the target area;

[0112] A parameter acquisition module is used to acquire input parameters when the number of layers is 1; the input parameters include point layer data source connection information and filtering conditions, kernel density analysis parameters and reclassification parameters;

[0113] A retrieval module, used to retrieve the input parameter using the single layer condition management table, and if there is no identical record in the single layer condition management table, obtain the data extraction result of the target area using the kernel density calculation model, and store the data extraction result in the single layer condition management table;

[0114] A data extraction result acquisition module, used to find the data extraction result of the target area according to the corresponding input parameters when there are identical records in the single layer condition management table;

[0115] The multi-layer extraction module is used to obtain the data extraction result of the target area using the multi-layer target area extraction method when the number of layers is multiple.

[0116] Preferably, the kernel density calculation model in the retrieval module is:

[0117]

[0118] in, f ( x , y ) represents the density value of the prediction point, x is the longitude coordinate of the prediction point, y is the latitude coordinate of the prediction point, i =1,2,3.. n Indicates the input point, n is the number of input points, radius Indicates the predicted point ( x , y ) as the center of the search radius, Yes i The weight field value of Yes i To the predicted point ( x , y) distance.

[0119] Preferably, the multi-layer extraction module includes:

[0120] A multi-layer input parameter acquisition unit is used to acquire multi-layer input parameters; the multi-layer input parameters include multiple point layer data source connection information and filtering conditions, kernel density analysis parameters and reclassification parameters;

[0121] A retrieval unit, used to retrieve the multi-layer input parameters using a multi-layer condition management table, and if there is no identical record in the multi-layer condition management table, obtain a data extraction result of the target area using a kernel density calculation model, and store the data extraction result in the multi-layer condition management table;

[0122] The result extraction unit is used to find the data extraction result of the target area according to the corresponding multi-layer input parameters when there are identical records in the multi-layer condition management table.

[0123] Preferably, the retrieval unit comprises:

[0124] The point data acquisition subunit is used to acquire point data according to the point layer data source connection information and filtering conditions;

[0125] A raster data calculation subunit, used for inputting the point data and kernel density analysis parameters into the kernel density calculation model to obtain result raster data;

[0126] A reclassification subunit, used for reclassifying the result raster data to obtain the classified result raster data;

[0127] The raster-to-surface subunit is used to perform raster-to-surface conversion on the result raster data after the classification to obtain the data extraction result of the target area.

[0128] See also Figure 3 It should be noted that the system for extracting data from kernel density analysis of the present invention also has a timed execution strategy management module. The timed execution strategy management module is a separate service independent of the target area analysis and extraction service. By saving the parameters for calling the target area analysis and extraction service and additionally configuring the execution frequency of the timed task, the timed execution task will execute the target area analysis and extraction service regularly according to the configured parameters based on this frequency. For example, by executing it on a daily basis, it can generate new target area analysis result data every day, or update the target area analysis result data for the current month.

[0129] Scheduled execution type and scheduled execution time: These two configurations determine the execution time of the scheduled execution strategy. Generally, there are two types:

[0130] 1. Execute annually, and configure the time to be the month, day, hour, and minute, which means it will be executed once a year at this time;

[0131] 2. Execute on a quarterly basis, and then configure the time to be the hour and minute of the day of the month, that is, execute once every quarter at this time;

[0132] 3. Execute on a monthly basis, and the configured time is the hour and minute of the day, that is, it will be executed once every month at this time;

[0133] 4. Execute on a daily basis, and configure the time to be a few minutes, that is, execute once every day at this time;

[0134] 5. Time interval, the time unit is days and hours. You can set it to 30 days, which means it will be executed for the first time from the second day after the creation time, and then it will be executed every 30 days. You can also set it to one hour, which means it will be executed for the first time from the second hour after the creation time, and then it will be executed every hour.

[0135] 6. Execute once, the configuration is the year, month, day, hour and minute, that is, it will be executed once after this time;

[0136] Calling parameters: the parameters passed in when calling the target area extraction service, mainly including point data source, kernel density analysis parameters, reclassification operation parameters, etc. The specific meaning of the parameters is explained in the function description of condition management in the next section.

[0137] Policy status: Enabled or disabled, only enabled services will be executed;

[0138] Creation time: save the time when the policy is created;

[0139] Last Execution Time: Saves the time when the policy was last executed.

[0140] (2) Condition management

[0141] Condition management is an entity table used to configure and execute a series of subsequent operation parameters. It also provides the data basis for conditional retrieval results. It is the starting point for the extraction, analysis, execution and retrieval of the entire target area.

[0142] During execution, whether the analysis is automatically executed by a scheduled task or performed by manually calling the analysis service, a unique code will be generated as the condition number to facilitate the association of subsequent intermediate process data and result data with the condition. At the same time, the content of the condition is also stored for use when retrieving results.

[0143] Point data source: records the connection parameters of the point data source. When there are multiple data sources, they are stored as an array. This parameter is also used when doing kernel density analysis.

[0144] Kernel density analysis parameters: parameters required to be passed in when calling a third-party kernel density analysis tool, such as pixel size, analysis radius, weight field, etc.

[0145] Reclassification parameters: parameters to be passed in when calling a third-party reclassification tool, which are used to divide the raster results from the kernel density analysis into several levels according to the threshold. These levels are the importance of these analysis area surfaces; areas below the minimum threshold are regarded as non-target areas and will not be extracted into the target area surface result data later;

[0146] Creation time: the time when the condition was first created;

[0147] Update Time: Records the time when the last update was performed.

[0148] (3) Data management of kernel density analysis results

[0149] Kernel density analysis result data management is an entity table used to manage kernel density analysis raster data results. It records the path where the kernel density analysis raster data files are stored, as well as the creation time and update time of the data. Generally, a record is created or updated when the result data is cached after calling the third-party kernel density analysis.

[0150] (4) Raster fusion result data management

[0151] Raster fusion result data management is an entity table used to manage raster fusion result data. It records the fusion result data obtained by raster fusion processing of the kernel density analysis raster results of these point layers when the target area analysis is extracted based on multiple point layers. Generally, a record is created or updated when the result data is cached after calling the third-party raster fusion tool.

[0152] Raster fusion is equivalent to the kernel density analysis result of multiple point layers. Based on this result data, further processing can obtain the target area surface data extracted based on the multi-point layer. Raster fusion is used here to solve the problem of very long analysis time caused by large data volume, as well as to reuse the results based on the previous single layer analysis, thereby further saving analysis time.

[0153] (5) Raster reclassification result data management

[0154] Raster reclassification result data management is an entity table used to manage raster reclassification result data. It records the raster result data obtained by reclassifying and assigning values ​​according to the threshold value for the raster result after kernel density analysis or the raster result after fusion processing. Generally, a record is created or updated when the result data is cached after calling the third-party raster reclassification tool.

[0155] (6) Data management of raster-to-surface conversion results

[0156] Raster-to-polygon result data management is an entity table used to manage raster-to-polygon processing result data. It records the results of polygon feature extraction for the reclassified raster data. The extracted polygon features will be stored separately according to the level of the threshold value during reclassification. Generally, the result data is obtained after calling the third-party raster-to-polygon tool and stored or updated in this table. There may be one or more result data according to the level.

[0157] The beneficial effect of the method and system for data extraction based on kernel density analysis provided by the present invention is that, compared with the prior art, the present invention improves the operating efficiency and display efficiency of the kernel density analysis system by utilizing the condition management table to cache and use intermediate process data and result data.

[0158] The present invention also provides an electronic device, comprising a bus, a transceiver, a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the transceiver, the memory and the processor are connected via the bus, and when the computer program is executed by the processor, the steps in the above-mentioned method for data extraction based on kernel density analysis are implemented. Compared with the prior art, the beneficial effects of the electronic device provided by the present invention are the same as the beneficial effects of the method for data extraction based on kernel density analysis described in the above-mentioned technical solution, and are not elaborated herein.

[0159] The present invention also provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the steps in the above-mentioned method for data extraction based on kernel density analysis are implemented. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided by the present invention are the same as the beneficial effects of the method for data extraction based on kernel density analysis described in the above-mentioned technical solution, which will not be repeated here.

[0160] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any technical solution that can be easily thought of by a person skilled in the art within the technical scope disclosed by the present invention should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be based on the protection scope of the claims.

Claims

1. A method for data extraction based on kernel density analysis, characterized in that: include: Step 1: Get the number of layers contained in the target area; Step 2: When the number of layers is 1, obtain input parameters; the input parameters include point layer data source connection information and filtering conditions, kernel density analysis parameters and reclassification parameters; Step 3: Retrieve the input parameter in the single layer condition management table. If there is no identical record in the single layer condition management table, use the kernel density calculation model to obtain the data extraction result of the target area, and store the data extraction result in the single layer condition management table. Step 4: If there are identical records in the single layer condition management table, find the data extraction result of the target area according to the corresponding input parameters; Step 5: When the number of layers is multiple, the multi-layer target area extraction method is used to obtain the data extraction result of the target area; The step 5: when the number of layers is multiple layers, a multi-layer target region extraction method is used to obtain a data extraction result of the target region, including: Step 5.1: Obtain multi-layer input parameters; the multi-layer input parameters include multiple point layer data source connection information and filtering conditions, kernel density analysis parameters and reclassification parameters; Step 5.2: Retrieve the multi-layer input parameters in the multi-layer condition management table. If there is no identical record in the multi-layer condition management table, use the kernel density calculation model to obtain the data extraction result of the target area, and store the data extraction result in the multi-layer condition management table; Step 5.3: If there are identical records in the multi-layer condition management table, find the data extraction result of the target area according to the corresponding multi-layer input parameters; In step 5.2, the kernel density calculation model is used to obtain the data extraction results of the target area, including: Step 5.2.1: Get point data based on the point layer data source connection information and filtering conditions; Step 5.2.2: Input the point data and kernel density analysis parameters into the kernel density calculation model to obtain result grid data; Step 5.2.3: reclassifying the result raster data to obtain classified result raster data; Step 5.2.4: Convert the resulting raster data after the classification to a surface to obtain the data extraction result of the target area.

2. The method for data extraction based on kernel density analysis according to claim 1, characterized in that: The kernel density calculation model in step 3 is: Where f(x,y) represents the density value of the prediction point, x is the longitude coordinate of the prediction point, y is the latitude coordinate of the prediction point, i=1,2,3..n represents the input point, n is the number of input points, radius represents the search radius centered on the prediction point (x,y), w i is the weight field value of point i, dist i is the distance from point i to the predicted point (x, y).

3. A system for data extraction based on kernel density analysis, characterized in that: include: The layer number acquisition module is used to obtain the number of layers contained in the target area; A parameter acquisition module is used to acquire input parameters when the number of layers is 1; the input parameters include point layer data source connection information and filtering conditions, kernel density analysis parameters and reclassification parameters; A retrieval module, used to retrieve the input parameter in the single layer condition management table, and if there is no identical record in the single layer condition management table, obtain the data extraction result of the target area using the kernel density calculation model, and store the data extraction result in the single layer condition management table; A data extraction result acquisition module is used to find the data extraction result of the target area according to the corresponding input parameters when there are identical records in the single layer condition management table; A multi-layer extraction module is used to obtain the data extraction result of the target area using a multi-layer target area extraction method when the number of layers is multiple; The multi-layer extraction module comprises: A multi-layer input parameter acquisition unit is used to acquire multi-layer input parameters; the multi-layer input parameters include multiple point layer data source connection information and filtering conditions, kernel density analysis parameters and reclassification parameters; A retrieval unit, used for retrieving the multi-layer input parameter in the multi-layer condition management table, and if there is no identical record in the multi-layer condition management table, obtaining a data extraction result of the target area using a kernel density calculation model, and storing the data extraction result in the multi-layer condition management table; A result extraction unit, used to find the data extraction result of the target area according to the corresponding multi-layer input parameters when there are identical records in the multi-layer condition management table; The retrieval unit comprises: The point data acquisition subunit is used to acquire point data according to the point layer data source connection information and filtering conditions; A raster data calculation subunit, used for inputting the point data and kernel density analysis parameters into the kernel density calculation model to obtain result raster data; A reclassification subunit, used for reclassifying the result raster data to obtain the classified result raster data; The raster-to-surface subunit is used to perform raster-to-surface conversion on the result raster data after the classification to obtain the data extraction result of the target area.

4. A system for data extraction based on kernel density analysis according to claim 3, characterized in that: The kernel density calculation model in the retrieval module is: Where f(x,y) represents the density value of the prediction point, x is the longitude coordinate of the prediction point, y is the latitude coordinate of the prediction point, i=1,2,3..n represents the input point, n is the number of input points, radius represents the search radius centered on the prediction point (x,y), w i is the weight field value of point i, dist i is the distance from point i to the predicted point (x, y).

5. An electronic device, comprising a bus, a transceiver, a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the transceiver, the memory, and the processor are connected via the bus, wherein: When the computer program is executed by the processor, the steps in the method for data extraction based on kernel density analysis as described in any one of claims 1 to 2 are implemented.

6. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps in the method for data extraction based on kernel density analysis as described in any one of claims 1 to 2 are implemented.

Images