A method and system for identifying abnormal information of a graph patch based on remote sensing monitoring data

Abstract

The application discloses a kind of based on remote sensing monitoring data's graph spot abnormal information identification method and system, the method includes: based on the remote sensing monitoring data of to be checked graph spot, obtain the geographical position and graph spot attribute of to be checked graph spot;The land use management information corresponding to the geographical position is obtained, and the land use management information is based on the graph spot attribute and predicts the land type information corresponding to the to be checked graph spot;The land type information of field investigation of the to be checked graph spot is obtained, and based on the land type information of field investigation and the land type information obtains the abnormal check result of the to be checked graph spot, to improve the accuracy of graph spot check.

Description

Technical Field

[0001] This invention relates to the field of land and resources management technology, specifically to a method and system for identifying abnormal information of map patches based on remote sensing monitoring data. Background Technology

[0002] Land use change survey refers to a comprehensive, accurate, and systematic survey of land resources and spatial utilization within a region. It is usually based on the results of the previous year's land use change survey, and utilizes the latest satellite remote sensing imagery, through field investigations and internal verification, to promptly grasp the annual changes in land use, thereby ensuring the timeliness and accuracy of the land use survey results.

[0003] When conducting land use change surveys, if remote sensing images reveal a potential land use change in a particular area, a field survey task is generated for that area to obtain on-site information about the current land use and upload the survey results. When updating the land use change data for that area based on the survey results, the uploaded results must be verified to ensure their authenticity and accuracy.

[0004] In existing technologies, the review of survey results relies solely on verification personnel to conduct checks on the authenticity, standardization, and correctness of the reported findings. This method of verifying survey results uses a relatively singular verification dimension, lacking an understanding of land use prediction changes based on remote sensing images. Consequently, it lacks a basis for accuracy verification, failing to guarantee the accuracy of the review. Furthermore, existing technologies are heavily dependent on verification personnel, lacking objectivity and impacting the accuracy rate, thus affecting the quality of the survey results. Summary of the Invention

[0005] To address the aforementioned technical problems, this invention discloses a method and system for identifying anomaly information in image patches based on remote sensing monitoring data, thereby improving the accuracy of image patch verification. Specifically, the method for identifying anomaly information in image patches based on remote sensing monitoring data includes:

[0006] Based on the remote sensing monitoring data of the map features to be verified, the geographical location and map feature attributes of the map features to be verified are obtained;

[0007] Obtain the land use management information corresponding to the geographical location, and predict the land category information corresponding to the map patch to be verified based on the land use management information and the map patch attributes;

[0008] Obtain the field survey land category information of the map patch to be verified, and obtain the anomaly verification result of the map patch to be verified based on the field survey land category information and the land category information.

[0009] This invention discloses a method for identifying anomalies in map features based on remote sensing monitoring data. The method verifies the field survey results of the map features to be verified using remote sensing monitoring data, thereby providing a reference basis for anomaly verification and improving the accuracy of map feature verification. Specifically, during map feature verification, the geographical location and attributes of the map features to be verified are obtained from the remote sensing monitoring data. Based on the geographical location and attributes, the possible land use type of the map features is predicted. Then, the predicted land use type is used to verify the reported land use type information of the map features from the field survey, further increasing the accuracy of map feature verification.

[0010] As a preferred example, the step of obtaining the geographical location and attributes of the map patch to be verified based on the remote sensing monitoring data of the map patch to be verified includes:

[0011] Obtain the attribute structure corresponding to the remote sensing monitoring data;

[0012] The remote sensing monitoring data is parsed according to the attribute structure to obtain the geographical location, change monitoring type, and type of the patch to be verified; wherein, the patch attributes include the change monitoring type and the type of patch.

[0013] In the above scheme, the remote sensing monitoring data is parsed based on the attribute structure of the remote sensing monitoring data of the land parcels to be verified, so as to ensure the accuracy of the parsing. Secondly, the parsing is used to obtain the geographical location, land parcel change monitoring type, and land parcel type corresponding to the land parcels to be verified, providing a data foundation for subsequent land use prediction, thereby ensuring the accuracy of land use prediction.

[0014] As a preferred example, before obtaining the land use management information corresponding to the geographical location, the process includes:

[0015] Based on the reported change monitoring type, a pre-built map patch type attribute table is queried to obtain multiple original map patch types corresponding to the reported change monitoring type;

[0016] Obtain the comparison result between each of the original patch types and the patch type;

[0017] Based on the comparison results, it was determined that the attributes of the image patch were not abnormal.

[0018] In the above scheme, before predicting the land use category of the map patch, anomaly detection is first performed on the parsed data. Only after the data is found to be free of anomalies can subsequent land use category prediction be performed. This avoids the execution of useless and erroneous land use category prediction steps, which would waste time in map patch verification and thus ensure the efficiency of map patch verification.

[0019] As a preferred example, obtaining the land use management information corresponding to the geographical location includes:

[0020] Construct a geographic layer for the map patch to be verified based on the geographic location;

[0021] The geographic layer of the map features is overlaid with the geographic layer of the pre-acquired land use management information;

[0022] Extract the overlapping layer when the land use management information geographic layer and the map patch geographic layer overlap, and obtain the land use management information corresponding to the overlapping layer.

[0023] In the above scheme, a geographic layer is constructed based on the geographical location of the map patch to be verified, and the geographic layer is superimposed with the pre-acquired land use management information geographic layer to obtain the land use management information corresponding to the map patch to be verified. This enables subsequent identification of the land use type of the map patch based on the dynamically changing land use management information, thereby improving the accuracy of land use type prediction.

[0024] As a preferred example, the step of predicting the land category information corresponding to the map patch to be verified based on the land use management information and the map patch attributes includes:

[0025] Based on the land use management information, determine whether the plot to be verified is within the land use area, and record the determination result;

[0026] Based on the judgment result and the change monitoring type and type of the map patch to be verified, the pre-constructed land use attribute table is queried to obtain the predicted land use type corresponding to the map patch to be verified.

[0027] In the above scheme, the land use management information and the change monitoring type and type of the map patch to be verified are used to query the pre-constructed land category attribute table to predict the land category that the map patch may correspond to, which improves both the efficiency and accuracy of the prediction.

[0028] As a preferred example, obtaining the land use information from the field survey of the map patch to be verified includes:

[0029] From the obtained single-plot survey results, select the first single-plot survey results that have passed the verification;

[0030] The survey results of each of the first single map patches are analyzed to obtain the corresponding survey data; wherein, the survey data includes the verification area and survey land category information of the map patch;

[0031] Match the geographical locations of the verification area and the map patch to be verified, and obtain the second map patch survey results corresponding to the map patch to be verified from a plurality of first map patch survey results;

[0032] The land use information of the second single-plot survey results shall be used as the field survey land use information of the plot to be verified.

[0033] In the above scheme, when a potential land use change is identified in a certain area, a single map patch corresponding to that area is generated, and a field verification task corresponding to that single map patch is issued to obtain the field survey results. Therefore, when reviewing the map patch for anomalies, the corresponding single map patch survey results can be found based on geographic location matching to improve data query efficiency. Furthermore, in order to review the map patch results for anomalies based on predicted land categories, surveyed land category information can be retrieved from the map patch results to identify anomalies in the map patch results.

[0034] As a preferred example, obtaining the anomaly verification result of the map patch to be verified based on the field survey land category information and the land category information includes:

[0035] Based on the land category information and land category information obtained from the field survey, the rationality of the second single-plot survey results is verified.

[0036] When the survey results of the second single patch are determined to be unreasonable, the patch to be verified is identified as an abnormal patch.

[0037] In the above scheme, the reasonableness of the reported single-plot survey results is judged by comparing the predicted land types and the survey land type information in the plot results. Then, based on the judgment result, it is identified whether the plot to be verified has undergone a real field survey. If the judgment is unreasonable, it is determined that the plot survey results of the plot to be verified are incorrect, and the current plot is judged as an abnormal plot so that the abnormal plot can be re-investigated.

[0038] On the other hand, the map patch anomaly information identification system based on remote sensing monitoring data disclosed in this invention includes a remote sensing monitoring data parsing module, a land type prediction module, and a map patch verification module;

[0039] The remote sensing monitoring data parsing module is used to obtain the geographical location and attributes of the map features to be verified based on the remote sensing monitoring data of the map features to be verified.

[0040] The land use prediction module is used to obtain land use management information corresponding to the geographical location, and predict the land use information corresponding to the land use parcel to be checked based on the land use management information and the parcel attributes;

[0041] The map patch verification module is used to obtain the field survey land category information of the map patch to be verified, and to obtain the anomaly verification result of the map patch to be verified based on the field survey land category information and the land category information.

[0042] This invention discloses a system for identifying anomalies in map features based on remote sensing monitoring data. The system verifies the field survey results of the map features to be verified based on the remote sensing monitoring data of the features to be verified, thereby providing a reference basis for anomaly verification and improving the accuracy of map feature verification. Specifically, during map feature verification, the system obtains the geographical location and attributes of the map features to be verified from the remote sensing monitoring data. Based on the geographical location and attributes, it predicts the possible land use type of the map features. Then, it uses the predicted land use type information to verify the reported land use type information of the map features to be verified, thereby increasing the accuracy of map feature verification.

[0043] As a preferred example, the remote sensing monitoring data parsing module includes an attribute analysis unit and a data extraction unit;

[0044] The attribute analysis unit is used to obtain the attribute structure corresponding to the remote sensing monitoring data;

[0045] The data extraction unit is used to parse the remote sensing monitoring data according to the attribute structure to obtain the geographical location, change monitoring type and type of the patch to be verified; wherein, the patch attributes include the change monitoring type and the type of patch.

[0046] In the above scheme, the remote sensing monitoring data is parsed based on the attribute structure of the remote sensing monitoring data of the land parcels to be verified, so as to ensure the accuracy of the parsing. Secondly, the parsing is used to obtain the geographical location, land parcel change monitoring type, and land parcel type corresponding to the land parcels to be verified, providing a data foundation for subsequent land use prediction, thereby ensuring the accuracy of land use prediction.

[0047] As a preferred example, the image patch verification system further includes a data verification module; wherein, the data verification module includes a type query unit and an anomaly detection unit;

[0048] The type query unit is used to query a pre-built map patch type attribute table according to the map patch change monitoring type to obtain multiple original map patch types corresponding to the map patch change monitoring type;

[0049] The anomaly detection unit is used to obtain the comparison result between each of the original patch types and the patch type; based on the comparison result, it is determined that the patch attributes do not have anomalies.

[0050] In the above scheme, before predicting the land use category of a map patch, anomaly detection is first performed on the parsed data. Only after detecting that there are no anomalies in the data will subsequent land use category prediction be performed, thereby avoiding the execution of useless and erroneous land use category prediction steps and wasting map patch verification time, thus ensuring the efficiency of map patch verification. Attached Figure Description

[0051] Figure 1 This is a flowchart illustrating a method for identifying abnormal information of map patches based on remote sensing monitoring data, as disclosed in an embodiment of the present invention.

[0052] Figure 2 This is a schematic diagram of the structure of a patch anomaly information identification system based on remote sensing monitoring data, as disclosed in an embodiment of the present invention.

[0053] Figure 3 This is a flowchart illustrating a method for identifying abnormal information of map patches based on remote sensing monitoring data, as disclosed in another embodiment of the present invention. Detailed Implementation

[0054] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0055] Example 1

[0056] This embodiment discloses a method for identifying anomaly information in image patches based on remote sensing monitoring data, which improves the accuracy of image patch verification. Specifically, the implementation process of the image patch verification method can be found in [reference needed]. Figure 1 It mainly includes steps 101 to 103, and the steps are mainly as follows:

[0057] Step 101: Based on the remote sensing monitoring data of the patch to be verified, obtain the geographical location and patch attributes of the patch to be verified.

[0058] Step 102: Obtain the land use management information corresponding to the geographical location, and predict the land type information corresponding to the map patch to be verified based on the land use management information and the map patch attributes.

[0059] Step 103: Obtain the field survey land category information of the map patch to be verified, and obtain the anomaly verification result of the map patch to be verified based on the field survey land category information and the land category information.

[0060] The above steps involve parsing the remote sensing monitoring data of the acquired map features to be verified to obtain the geographical location and map feature attributes included in the remote sensing monitoring data. The geographical location is used to determine whether the area where the map feature is located corresponds to land use management information, thus providing a reference for subsequent land use type prediction. Land use type prediction is performed based on the map feature attributes and the land use management information to ensure that the predicted land use type matches the actual type of the map feature and the actual land use situation, thereby improving the accuracy of land use type prediction. Next, based on the predicted land use type information, anomaly detection is performed on the field survey land use information of the map features to be verified to identify whether the field survey of the map features to be verified was conducted correctly. Based on the identification results, the map features are marked as anomalies to facilitate subsequent re-surveys, thereby improving the accuracy and effectiveness of the map feature survey. It should be noted that the field survey results refer to the process where, when a potential land use change is identified in a certain area based on remote sensing images, a single map patch in that area is collected, and a survey task corresponding to that single map patch is generated. After receiving the survey task, the user conducts a field survey of the single map patch to generate the field survey results for the area where the single map patch is located. The field survey results include land use information identified by the user, the geographical location of the area, etc.

[0061] In one embodiment of this example, when extracting the geographical location and attributes of the patch to be verified from the remote sensing monitoring data, to ensure the accuracy of data extraction and thus the accuracy of subsequent land use prediction, step 101 can preferably generate the geographical location and patch attributes using the following steps. Specifically, the steps are as follows:

[0062] Step 1011: Obtain the attribute structure corresponding to the remote sensing monitoring data;

[0063] Step 1012: Parse the remote sensing monitoring data according to the attribute structure to obtain the geographical location, change monitoring type and type of the patch to be checked; wherein, the patch attributes include the change monitoring type and the type of patch.

[0064] The above steps parse the remote sensing monitoring data based on the attribute structure of the distributed map features to be verified, ensuring the accuracy of the parsing. Secondly, the parsing is used to obtain the geographical location, change monitoring type, and type of the map features to be verified, providing a data foundation for subsequent land use prediction and thus ensuring the accuracy of the prediction.

[0065] In one embodiment of this example, before extracting the corresponding geographical location and patch attributes for land use prediction, to ensure the accuracy of the data extraction or the correctness of the acquired remote sensing monitoring data, and to avoid repeated errors in subsequent land use prediction steps, the extracted data can be anomaly detected through the following steps before performing step 102. Specifically, the steps are as follows:

[0066] Step S1: Query the pre-built map patch type attribute table according to the map patch change monitoring type to obtain multiple original map patch types corresponding to the map patch change monitoring type;

[0067] Step S2: Obtain the comparison result between each of the original patch types and the patch type;

[0068] Step S3: Based on the comparison results, determine that the attributes of the patch are not abnormal.

[0069] In the above steps, before predicting the land use category of the map patch, anomaly detection is first performed on the parsed data. Only after the data is found to be free of anomalies will subsequent land use category prediction be performed. This avoids executing useless or erroneous land use category prediction steps, which would waste time in map patch verification and thus ensure the efficiency of map patch verification.

[0070] In one embodiment of this example, when extracting the geographical location for acquiring geographical information, in order to improve the accuracy or efficiency of land use management information acquisition, accurate land use management information is used to predict accurate land category information. Step 102 can predict land categories through the following steps. Specifically, the steps are:

[0071] Step 1021: Construct a geographic layer for the patch to be verified based on the geographic location;

[0072] Step 1022: Overlay the geographic map layer with the pre-acquired land use management information range layer;

[0073] Step 1023: Extract the overlapping layer when the land use management information scope layer and the geographic patch layer overlap from the land use management information scope layer, and obtain the land use management information corresponding to the overlapping layer.

[0074] Step 1024: Determine whether the plot to be verified is within the land use area based on the land use management information, and record the determination result;

[0075] Step 1025: Based on the judgment result and the change monitoring type and type of the map patch to be verified, query the pre-constructed land use attribute table to obtain the predicted land use type corresponding to the map patch to be verified.

[0076] In the above steps, a geographic layer is constructed based on the geographical location of the map patch to be verified, and the geographic layer is superimposed on the pre-acquired land use management information range layer to obtain the land use management information corresponding to the map patch to be verified. This enables subsequent identification of the land use type of the map patch based on the dynamically changing land use management information, thereby improving the accuracy of land use type prediction. At the same time, the land use management information and the map patch change monitoring type and map patch type of the map patch to be verified are used to query the pre-constructed land use attribute table to predict the possible land use type corresponding to the map patch, which improves both the efficiency and accuracy of prediction.

[0077] In one embodiment of this example, after predicting the land type of the map patch to be verified, the field survey land type information corresponding to the map patch to be verified can be obtained. This allows for comparison between the predicted land type and the field survey land type information to identify whether any anomalies have occurred during the verification of the map patch. Specifically, step 103 can identify abnormal map patches through the following steps:

[0078] Step 1031: Select several first single-plot survey results with a verification conclusion of "passed" from the obtained single-plot survey results;

[0079] Step 1032: Analyze the survey results of each of the first single map patches to obtain the corresponding survey data; wherein, the survey data includes the verification area and survey land type information of the map patch;

[0080] Step 1033: Match the geographical location of the verification area and the map patch to be verified, and obtain the second map patch corresponding to the map patch to be verified from several first map patch survey results;

[0081] Step 1034: Use the survey land category information of the second single map patch survey results as the field survey land category information of the map patch to be verified.

[0082] Step 1035: Based on the field survey land type information and the land type information, verify the rationality of the second single-plot survey results;

[0083] Step 1036: When it is determined that the survey results of the second single patch are unreasonable, the patch to be verified is identified as an abnormal patch.

[0084] In the above steps, when a potential land use change is identified in a certain area, a single map patch corresponding to that area is generated, and a field verification task corresponding to that single map patch is issued to obtain the field survey results. Therefore, when reviewing the map patch for anomalies, the corresponding single map patch survey results can be found based on geographical location matching to improve data query efficiency. Furthermore, to review the map patch results for anomalies based on predicted land categories, surveyed land category information can be retrieved from the map patch results to identify anomalies in the map patch results. Then, by comparing the predicted land categories with the surveyed land category information in the map patch results, the reasonableness of the reported single map patch survey results is judged. Based on the judgment result, it is determined whether the current map patch to be verified has undergone a genuine field survey. If the judgment is unreasonable, the map patch survey results of the map patch to be verified are determined to be incorrect, and the current map patch is identified as an anomaly, so that the anomaly map patch can be re-investigated.

[0085] On the other hand, this embodiment also discloses a system for identifying anomaly information in image patches based on remote sensing monitoring data. For the specific structural composition of the verification system, please refer to... Figure 2 It mainly includes a remote sensing monitoring data analysis module 201, a land category prediction module 202, and a map patch verification module 203.

[0086] The remote sensing monitoring data parsing module 201 is used to obtain the geographical location and attributes of the map patch to be verified based on the remote sensing monitoring data of the map patch to be verified.

[0087] The land use prediction module 202 is used to obtain the land use management information corresponding to the geographical location, and predict the land use information corresponding to the land use parcel to be checked based on the land use management information and the parcel attributes.

[0088] The map patch verification module 203 is used to obtain the field survey land category information of the map patch to be verified, and to obtain the abnormal verification result of the map patch to be verified based on the field survey land category information and the land category information.

[0089] In this embodiment, the remote sensing monitoring data parsing module 201 includes an attribute analysis unit and a data extraction unit.

[0090] The attribute analysis unit is used to obtain the attribute structure corresponding to the remote sensing monitoring data.

[0091] The data extraction unit is used to parse the remote sensing monitoring data according to the attribute structure to obtain the geographical location, change monitoring type and type of the patch to be verified; wherein, the patch attributes include the change monitoring type and the type of patch.

[0092] In this embodiment, as Figure 2The map patch verification system shown also includes a data verification module; wherein, the data verification module includes a type query unit and an anomaly detection unit.

[0093] The type query unit is used to query a pre-built map patch type attribute table based on the map patch change monitoring type to obtain multiple original map patch types corresponding to the map patch change monitoring type.

[0094] The anomaly detection unit is used to obtain the comparison result between each of the original patch types and the patch type; based on the comparison result, it is determined that the patch attributes do not have anomalies.

[0095] This embodiment provides a method and system for identifying anomaly information in map features based on remote sensing monitoring data. It verifies the uploaded field survey results of the map features to be verified based on the remote sensing monitoring data of the map features to be verified, thereby providing a reference basis for anomaly verification and improving the accuracy of map feature verification. Specifically, during map feature verification, the geographical location and attributes of the map features to be verified are obtained from the remote sensing monitoring data. Based on the geographical location and attributes, the possible land use type information currently corresponding to the map features is predicted. Then, the predicted land use type information is used to perform anomaly verification on the reported field survey land use type information of the map features to be verified, thereby increasing the accuracy of map feature verification.

[0096] Example 2

[0097] Accurate and timely acquisition of land use change information for each region is a crucial part of land surveying. This acquisition typically involves higher-level units using remote sensing images to identify potential land use changes in a specific area, generating map features and remote sensing monitoring data for those features. This data is then sent to intermediate-level units, which compile and distribute it to lower-level units for on-site investigation and feedback. The intermediate-level units then verify the on-site investigation results. Current verification methods only check the authenticity, standardization, and accuracy of land use classification and attribute identification for individual map features. This relatively narrow verification scope can lead to omissions in verification, such as local authorities deliberately omitting buildings reported as agricultural land in their field evidence photos, thus compromising the accuracy of the verification.

[0098] Meanwhile, existing technologies rely heavily on verification personnel, resulting in a lack of objectivity in the review process. Under the multiple pressures of reviewing a large number of map features, tight deadlines, and high quality requirements, the review status of personnel involved in verification, review, and spot checks fluctuates, their professional skills vary, and human factors may even introduce errors. The lack of objective verification in the process affects the accuracy of the verification to some extent, further impacting the quality of the investigation results.

[0099] To address the aforementioned technical problems, this embodiment discloses a method for verifying map features by combining remote sensing monitoring data issued by higher-level units with data reported by lower-level units from multiple dimensions, thereby improving the accuracy of map feature verification. Specifically, the detailed implementation process of the map feature verification method is described in [reference needed]. Figure 3 It mainly includes steps 301 to 304, and the steps are mainly as follows:

[0100] Step 301: Analyze the acquired remote sensing monitoring data to obtain the geographical location, monitoring type, and type of the patch to be verified.

[0101] In this embodiment, the attribute structure of the distributed remote sensing monitoring patch data is obtained. Based on the attribute structure, fields such as "monitoring type," "geographical location," and "patch type" in the remote sensing monitoring patch data are obtained. Then, the monitoring type, patch type, and geographical location of the patch to be verified, i.e., the remote sensing monitoring patch, are sorted out from these fields. Among them, the monitoring types are mainly identified and sorted out into six categories: suspected newly added construction patches, cultivated land process change patches, construction land and facility agricultural land change patches, non-cultivated agricultural land change patches, unused land change patches, and newly added reclamation patches.

[0102] Furthermore, after obtaining the patch monitoring type and patch type, to ensure the accuracy of data parsing, it can be determined whether the parsed patch type corresponds to the patch monitoring type based on the patch monitoring type and the pre-acquired patch attribute table, thereby identifying the accuracy of data parsing. In this embodiment, the patch attribute table is shown in Table 1:

[0103] Table 1. Attributes of Image Patch

[0104]

[0105]

[0106]

[0107]

[0108]

[0109]

[0110] The map patch attribute table shows that different map patch monitoring types correspond to different map patch attribute information. This map patch attribute information includes the map patch type name, map patch type code, and a description of each map patch type. The map patch type name column shows that the map patch types involve a total of 42 categories, including buildings / structures with clearly defined construction uses, buildings / structures with unknown uses, hardening, bulldozing / piling, completed roads, and roads under construction.

[0111] When performing anomaly detection on the data, the various types of patches that the patch to be tested may correspond to and the patch type code of each patch type can be obtained according to the identified monitoring type. Then, when the patch type or patch type code obtained after parsing the remote sensing monitoring data is consistent with a certain patch type or patch type code corresponding to the monitoring type, the result of data parsing is determined to be correct.

[0112] Step 302: Construct a geographic layer of the map patch to be verified based on the geographic location, and overlay the geographic layer with the pre-acquired land use management information range layer to obtain the land use management information of the map patch to be verified.

[0113] In this embodiment, given that multiple batches of remote sensing monitoring data may be received in real time, to ensure data standardization, the remote sensing monitoring data can be deduplicated according to the order of its latest and oldest iterations. Then, a geographic layer for the patches to be verified is constructed based on the deduplicated remote sensing monitoring data. Specifically, the geographic layer can be constructed using techniques in geographic information software such as ArcGIS.

[0114] Furthermore, the geographic layer is jointly overlaid with the pre-acquired land use management information range layer to determine whether the patch to be verified is within the land use range corresponding to the land use management information, thereby obtaining the remote sensing monitoring data, i.e., the patch to be verified, which is divided into the range outside the land use management information range. Furthermore, the newly generated layer after overlay can also be named the monitoring change range layer.

[0115] It should be noted that the land use management information range layer is a geographic layer constructed based on the geographic locations contained in the acquired land use management information.

[0116] The land use management information includes data on approved agricultural land conversion and land acquisition projects, data from the approved construction land filing system, data on the linkage between urban and rural construction land increases and decreases, map data from the temporary land use management information system, data on construction land implementation plan filings, and data on land use authorization approvals for transfer, etc. The overlay of these layers can also be performed using techniques within geographic information software such as ArcGIS.

[0117] Step 303: Obtain the predicted land use category of the plot to be verified based on the land use management information, the plot monitoring type, and the plot type.

[0118] In this embodiment, based on the land use management information, the monitoring type of the land parcel, and the land parcel type, the predicted land type of the land parcel to be verified is obtained by querying the pre-acquired predicted land type attribute table. Specifically, the predicted land type attribute table is shown in Table 2:

[0119] Table 2 Predicted Land Category Attributes

[0120]

[0121]

[0122]

[0123]

[0124] Referring to the predicted land use attribute table, it can be seen that the table includes various monitoring patch types and the corresponding monitoring patch attribute information for each type. The monitoring patch attribute information includes the patch type and the predicted land use category. Specifically, when a suspected newly constructed patch is identified as being of the monitoring type GF (i.e., photovoltaic land), and the patch to be verified was identified in the previous step as belonging to the land use area, then the predicted land use category for the patch to be verified is 0601.

[0125] Step 304: Obtain the land classification of the local reported results of the map patch to be verified, and obtain the anomaly verification results of the map patch to be verified based on the land classification of the local reported results and the predicted land classification.

[0126] In this embodiment, the map patches with a verification conclusion of "pass" are obtained from the reported single map patch results. Then, the map patches with the conclusion of "pass" are compared in detail to compare the consistency between the land use attributes of the reported map patches and the predicted land use attributes in the single map patch results.

[0127] During the comparison, the acquired remote sensing monitoring patch change range layer and the single patch results are overlaid to extract the single patch results corresponding to the patch to be verified. Fields such as monitoring type, patch type, predicted land use type, and reported land use type code are retained from the vector data of the matched single patch results. Specifically, when matching the patch to be verified with the single patch results, the geographical location corresponding to each single patch result can be obtained, and a geographic layer of the single patch result can be constructed. Then, the geographic layer of the patch to be verified and the single patch results can be overlaid using techniques in geographic information software such as ArcGIS.

[0128] Next, based on relevant land use classification standards, a reasonable relationship database is established between the predicted land use types of various monitored map patches and the land use information of reported single map patches. This database is used to determine the reasonable correspondence between the land use information in the single map patch results and the predicted land use types. Furthermore, based on the reasonableness of this correspondence, it is determined whether the map patch to be verified is an abnormal map patch, i.e., whether the single map patch result corresponding to the map patch to be verified is abnormal. Specifically, the relationship database is shown in Table 3:

[0129] Table 3 Relationship Database

[0130]

[0131]

[0132]

[0133]

[0134]

[0135]

[0136]

[0137]

[0138]

[0139]

[0140] For example, if the monitoring type obtained after parsing remote sensing monitoring data is "suspected newly added construction patch", and the patch type is "20 (buildings / structures with clear construction purpose)", then the corresponding predicted land category is 20 construction land. Under this predicted land category, if the land category in the single patch result is linear features (such as roads, pipelines, etc.) or construction land (such as commercial and industrial land), it is considered a reasonable correspondence; if the land category in the single patch result is arable land, garden land, or other agricultural land, it is considered an abnormal correspondence and is extracted as an abnormal patch for output. It should be noted that for monitoring patches with monitoring types of "changes in construction land and facility agricultural land", "changes in non-arable agricultural land", and "changes in unused land", and the patch type is "01", they belong to newly added arable land, and all newly added arable land must be submitted for manual key verification.

[0141] This embodiment provides a method for identifying anomaly information in map features based on remote sensing monitoring data. This method automatically and easily automates and batches the map feature verification process. Specifically, by predicting the land use type of the issued remote sensing monitoring map features in advance, it provides a basis for subsequent verification of survey results, allowing for timely and accurate correction of verification results. Secondly, it enriches the verification dimensions, reduces reliance on verifiers, and makes the verification results more objective and accurate. Specifically, by comparing the consistency between the land use type of the issued monitoring map features and the land use type of the reported results, the verification dimensions are enriched, reducing the possibility of errors introduced by human factors, improving the verification accuracy to a certain extent, and further ensuring the quality of the survey results.

[0142] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above descriptions are merely specific embodiments of the present invention and are not intended to limit the scope of protection of the present invention. In particular, it should be noted that any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention for those skilled in the art.

Claims

1. A method for identifying anomaly information in image patches based on remote sensing monitoring data, characterized in that, include: Based on the remote sensing monitoring data of the map features to be verified, the geographical location and map feature attributes of the map features to be verified are obtained; including: obtaining the attribute structure corresponding to the remote sensing monitoring data; parsing the remote sensing monitoring data according to the attribute structure to obtain the geographical location, map feature change monitoring type, and map feature type corresponding to the map features to be verified; wherein, the map feature attributes include the map feature change monitoring type and the map feature type; Obtain land use management information corresponding to the geographical location, and predict the land category information corresponding to the map patch to be verified based on the land use management information and the map patch attributes; including: determining whether the map patch to be verified is within the land use area based on the land use management information, and recording the determination result; querying a pre-constructed land category attribute table based on the determination result and the map patch change monitoring type and map patch type of the map patch to be verified, and obtaining the predicted land category corresponding to the map patch to be verified; Obtain the field survey land category information of the map patch to be verified, and obtain the anomaly verification result of the map patch to be verified based on the field survey land category information and the land category information.

2. The method for identifying anomaly information in image patches based on remote sensing monitoring data according to claim 1, characterized in that, Before obtaining the land use management information corresponding to the geographical location, the process includes: Based on the reported change monitoring type, a pre-built map patch type attribute table is queried to obtain multiple original map patch types corresponding to the reported change monitoring type; Obtain the comparison result between each of the original patch types and the patch type; Based on the comparison results, it was determined that the attributes of the image patch were not abnormal.

3. The method for identifying anomaly information in image patches based on remote sensing monitoring data according to claim 1, characterized in that, The step of obtaining the land use management information corresponding to the geographical location includes: Construct a geographic layer for the map patch to be verified based on the geographic location; The geographic layer of the map features is overlaid with the geographic layer of the pre-acquired land use management information; Extract the overlapping layer when the land use management information geographic layer and the map patch geographic layer overlap, and obtain the land use management information corresponding to the overlapping layer.

4. The method for identifying anomaly information in image patches based on remote sensing monitoring data according to claim 1, characterized in that, The process of obtaining the land use information from the field survey of the map patch to be verified includes: From the obtained single-plot survey results, select the first single-plot survey results that have passed the verification; The survey results of each of the first single map patches are analyzed to obtain the corresponding survey data; wherein, the survey data includes the verification area and survey land category information of the map patch; Match the geographical locations of the verification area and the map patch to be verified, and obtain the second map patch survey results corresponding to the map patch to be verified from a plurality of first map patch survey results; The land use information of the second single-plot survey results shall be used as the field survey land use information of the plot to be verified.

5. The method for identifying anomaly information in image patches based on remote sensing monitoring data according to claim 4, characterized in that, The process of obtaining the anomaly verification results for the plots to be verified based on the field survey land category information and the land category information includes: Based on the land category information and land category information obtained from the field survey, the rationality of the second single-plot survey results is verified. When the survey results of the second single patch are determined to be unreasonable, the patch to be verified is identified as an abnormal patch.

6. A system for identifying anomaly information in image patches based on remote sensing monitoring data, characterized in that, It includes a remote sensing monitoring data analysis module, a land category prediction module, and a map patch verification module; The remote sensing monitoring data parsing module is used to obtain the geographical location and attributes of the map features to be verified based on the remote sensing monitoring data of the map features to be verified. The remote sensing monitoring data parsing module includes an attribute analysis unit and a data extraction unit. The attribute analysis unit is used to obtain the attribute structure corresponding to the remote sensing monitoring data. The data extraction unit is used to parse the remote sensing monitoring data according to the attribute structure to obtain the geographical location, map feature change monitoring type, and map feature type corresponding to the map feature to be verified. The map feature attributes include the map feature change monitoring type and the map feature type. The land use prediction module is used to obtain land use management information corresponding to the geographical location, and predict the land use information corresponding to the land use parcel to be verified based on the land use management information and the parcel attributes; including: determining whether the land use parcel to be verified is within the land use area based on the land use management information, and recording the determination result; querying a pre-constructed land use attribute table based on the determination result and the parcel change monitoring type and parcel type of the land use parcel to be verified, and obtaining the predicted land use corresponding to the land use parcel to be verified; The map patch verification module is used to obtain the field survey land category information of the map patch to be verified, and to obtain the anomaly verification result of the map patch to be verified based on the field survey land category information and the land category information.

7. A system for identifying anomaly information in image patches based on remote sensing monitoring data according to claim 6, characterized in that, It also includes a data verification module; wherein the data verification module includes a type query unit and an anomaly detection unit; The type query unit is used to query a pre-built map patch type attribute table according to the map patch change monitoring type to obtain multiple original map patch types corresponding to the map patch change monitoring type; The anomaly detection unit is used to obtain the comparison result between each of the original patch types and the patch type; based on the comparison result, it is determined that the patch attributes do not have anomalies.

Images