A method and system for monitoring and checking rural farmland

Abstract

The application provides a rural cultivated land monitoring and checking method and system, applied to the technical field of data processing, which comprises the following steps: identifying a cultivated land area to be checked, acquiring the coordinate information of the cultivated land area, planning a UAV path, generating UAV cruise route information, and collecting image information of the cultivated land area. The cultivated land area image information is processed by a cultivated land area state recognition model to generate cultivated land area state information, wherein the cultivated land area state information comprises a cultivated land building state area. According to the cultivated land building state area, the identity information of the owner of the cultivated land is matched. According to the identity information of the owner of the cultivated land, the application form of the owner of the cultivated land in the cultivated land building state area is collected. According to the application form of the owner of the cultivated land, the cultivated land building state area is marked as illegal, and a rural cultivated land building state checking result is generated. The technical problems of low monitoring and checking efficiency and slow checking progress in the prior art are solved.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, specifically to a method and system for monitoring and verifying rural arable land. Background Technology

[0002] With urbanization, illegal construction and occupation of farmland have emerged in rural areas on the outskirts of cities. However, current technology for monitoring and verifying rural farmland mainly relies on on-site inspections to check for illegal occupation. Due to the wide distribution of farmland, inspectors need to conduct extensive on-site investigations, resulting in low efficiency and slow progress in monitoring and verifying rural farmland.

[0003] Therefore, existing methods for monitoring and verifying rural farmland suffer from technical problems such as low monitoring and verification efficiency and slow verification progress. Summary of the Invention

[0004] This application provides a method and system for monitoring and verifying rural arable land, which addresses the technical problems of low monitoring and verification efficiency and slow verification progress in existing methods for monitoring and verifying rural arable land.

[0005] In view of the above problems, this application provides a method and system for monitoring and verifying rural arable land.

[0006] The first aspect of this application provides a method for monitoring and verifying rural cultivated land. The method includes: identifying cultivated land areas to be verified and obtaining coordinate information of the cultivated land areas; performing path planning based on the cultivated land area coordinate information to generate aerial drone patrol route information; dispatching aerial drones based on the aerial drone patrol route information to collect image information of the cultivated land areas; processing the cultivated land area image information using a cultivated land area status recognition model to generate cultivated land area status information, wherein the cultivated land area status information includes cultivated land areas with housing construction status; matching the cultivated land owner's identity information based on the cultivated land owner's identity information; collecting the cultivated land owner's application form for the cultivated land area with housing construction status based on the cultivated land owner's application form; marking the cultivated land area with a violation status based on the cultivated land owner's application form, and generating a rural cultivated land construction status verification result.

[0007] A second aspect of this application provides a monitoring and verification system for rural cultivated land. The system includes: a coordinate information acquisition module for identifying cultivated land areas to be verified and acquiring coordinate information of the cultivated land areas; a cruise route information acquisition module for performing path planning based on the cultivated land area coordinate information and generating cruise route information for aerial drones; a cultivated land area image acquisition module for dispatching aerial drones based on the aerial drone cruise route information and collecting image information of the cultivated land areas; a cultivated land area status information recognition module for processing the cultivated land area image information using a cultivated land area status recognition model to generate cultivated land area status information, wherein the cultivated land area status information includes cultivated land areas with housing construction status; a cultivated land owner identity acquisition module for matching cultivated land owner identity information based on the cultivated land areas with housing construction status; a cultivated land owner application form collection module for collecting cultivated land owner applications based on the cultivated land owner identity information for the cultivated land areas with housing construction status; and a status verification module for marking the cultivated land areas with housing construction status as having violations based on the cultivated land owner applications and generating rural cultivated land construction status verification results.

[0008] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0009] The method provided in this application embodiment obtains the coordinate information of cultivated land areas by identifying the cultivated land areas to be verified. Path planning is performed based on the cultivated land area coordinate information to generate aerial drone patrol route information. An aerial drone is dispatched based on the aerial drone patrol route information to collect image information of the cultivated land areas. The cultivated land area image information is processed through a cultivated land area status recognition model to generate cultivated land area status information, wherein the cultivated land area status information includes cultivated land areas with housing construction status. The identity information of the cultivated land owner is matched based on the cultivated land owner's identity information. Application forms from the cultivated land owners for the cultivated land areas with housing construction status are collected based on the cultivated land owner's application forms. The cultivated land areas with housing construction status are marked as having a violation status based on the cultivated land owner's application forms, generating a rural cultivated land construction status verification result. By constructing a cultivated land area status recognition model, verification personnel can quickly and accurately identify the status of cultivated land areas and automatically obtain the identity of the cultivated land owner and the corresponding application form status. This achieves automated monitoring and verification of rural cultivated land, further reducing the workload of verification personnel and improving the efficiency of rural cultivated land monitoring and verification. This solves the technical problems of low monitoring and verification efficiency and slow verification progress in existing methods for monitoring and verifying rural farmland.

[0010] The above description is only an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, specific embodiments of this application are given below. Attached Figure Description

[0011] Figure 1 A flowchart illustrating a method for monitoring and verifying rural cultivated land provided in this application;

[0012] Figure 2 A schematic diagram illustrating the process of generating aerial drone cruise route information in a method for monitoring and verifying rural cultivated land provided in this application;

[0013] Figure 3 A flowchart illustrating the process of obtaining farmland regional status information in a method for monitoring and verifying rural farmland provided in this application;

[0014] Figure 4 This application provides a schematic diagram of the structure of a monitoring and verification system for rural cultivated land.

[0015] Explanation of reference numerals in the attached diagram: Coordinate information acquisition module 11, Cruise route information acquisition module 12, Farmland area image acquisition module 13, Farmland area status information identification module 14, Farmland owner identity acquisition module 15, Farmland owner application form collection module 16, Status verification module 17. Detailed Implementation

[0016] Example 1

[0017] To make the objectives, technical solutions, and advantages of this application clearer, the application will be further described in detail below with reference to the accompanying drawings. The described embodiments should not be regarded as limitations on this application. All other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0018] In the following description, references are made to “some embodiments”, which describe a subset of all possible embodiments. However, it is understood that “some embodiments” may be the same subset or different subsets of all possible embodiments and may be combined with each other without conflict.

[0019] In the following description, the terms "first, second, third" are used merely to distinguish similar objects and do not represent a specific ordering of objects. It is understood that "first, second, third" may be interchanged in a specific order or sequence where permitted, so that the embodiments of this application described herein can be implemented in an order other than that illustrated or described herein.

[0020] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only.

[0021] While this application makes various references to certain modules of the system according to embodiments of this application, any number of different modules may be used and run on user terminals and / or servers. These modules are merely illustrative, and different aspects of the system and method may use different modules.

[0022] This application uses flowcharts to illustrate the operations performed by the system according to embodiments of this application. It should be understood that the preceding or following operations are not necessarily performed in exact order. Instead, various steps can be processed in reverse order or simultaneously, as needed. Furthermore, other operations can be added to these processes, or one or more steps can be removed from them.

[0023] like Figure 1 As shown, this application provides a method for monitoring and verifying rural cultivated land, the method comprising:

[0024] S10: Identify the cultivated land area to be verified and obtain the coordinate information of the cultivated land area;

[0025] S20: Based on the coordinate information of the cultivated land area, perform path planning to generate the aerial drone's cruise route information;

[0026] S30: Dispatch the aerial photography drone according to the drone's cruise route information to collect image information of the cultivated land area;

[0027] Specifically, rural arable land verification is the foundation for rural arable land statistical analysis. It involves identifying the arable land areas to be verified, thus defining the boundaries of these areas. This identification utilizes arable land delineation records to mark the boundaries and obtain coordinate information, which can be specific latitude and longitude coordinates or fixed reference points. Subsequently, path planning is performed based on these coordinates to generate aerial drone patrol routes. The drones are then dispatched according to these routes, and their image acquisition equipment collects images of the arable land areas, facilitating further image processing.

[0028] like Figure 2 As shown, the method S20 provided in this application embodiment further includes:

[0029] S21: Based on the coordinate information of the cultivated land area, construct path nodes and generate a path node distribution network;

[0030] S22: Input the cruise start point coordinate information into the path node distribution network to generate multiple cruise distance information;

[0031] S23: Sort the path nodes according to the multiple cruise distance information to generate multiple cruise order information;

[0032] S24: Based on the multiple cruise sequence information, construct multiple cruise sub-routes;

[0033] S25: Add the multiple cruise sub-routes to the cruise route information of the aerial photography drone.

[0034] Specifically, when acquiring the drone's cruise route, path nodes are constructed based on the coordinates of cultivated land areas, generating a path node distribution network, where each path node corresponds to the coordinates of a cultivated land area. Then, the cruise starting point coordinates are input into the path node distribution network to generate multiple cruise distance information, which represents the straight-line distance from the cruise starting point coordinates to the cultivated land area coordinates. Since the cruise distance information at this stage only reflects the path distance, the path nodes need to be sorted based on the cruise distance information to generate multiple cruise sequences. These cruise sequences represent the order in which the drone reaches the nodes during its cruise. Based on these multiple cruise sequence information, multiple cruise sub-routes are constructed. Finally, the planned multiple cruise sub-routes are added to the drone's cruise route information, and the drone cruises according to these sub-routes.

[0035] The method S23 provided in this application embodiment further includes:

[0036] S231: Based on the multiple cruise distance information, obtain the k-th cruise distance and the (k+1)-th cruise distance;

[0037] S232: Connect the kth path node of the kth cruise distance to the cruise starting point coordinate information, connect the k+1th path node of the (k+1)th cruise distance to the cruise starting point coordinate information, and calculate the angle deviation.

[0038] S233: When the angle deviation is less than or equal to the angle deviation threshold, add the kth cruise distance and the (k+1)th cruise distance into the same cruise distance cluster group;

[0039] S234: Based on the clusters with the same cruising distance, the path nodes are sorted from closest to furthest to generate the multiple cruising order information.

[0040] Specifically, based on multiple cruise distance information, the k-th cruise distance and the (k+1)-th cruise distance are obtained. Then, the k-th path node of the k-th cruise distance is connected to the cruise starting point coordinates, and the (k+1)-th path node of the (k+1)-th cruise distance is connected to the cruise starting point coordinates, and the angular deviation is calculated. Since the UAV has a certain flight altitude, image acquisition can be performed on path nodes within a certain deviation range when flying in a straight line. When the angular deviation is less than or equal to an angular deviation threshold, the k-th and (k+1)-th cruise distances are added to a cluster group with the same cruise distance, where the cluster group with the same cruise distance consists of clusters with the same cruise route. Finally, the path nodes are sorted from nearest to farthest from the cluster group with the same cruise distance to generate the multiple cruise order information.

[0041] S40: The cultivated land area image information is processed by the cultivated land area status recognition model to generate cultivated land area status information, wherein the cultivated land area status information includes cultivated land areas with housing construction status.

[0042] S50: Match the identity information of the person who owns the farmland based on the area where the farmland is being used for housing construction;

[0043] S60: Collect the application form of the farmland owner in the area where the farmland is under construction, based on the identity information of the farmland owner;

[0044] S70: Based on the application from the owner of the cultivated land, mark the area of ​​the cultivated land with illegal construction status, and generate the rural cultivated land construction status verification result.

[0045] Specifically, the image information of the cultivated land area is processed using a cultivated land area status recognition model to generate cultivated land area status information. This model identifies the cultivated land area image and obtains its status, including areas with housing construction status. Subsequently, based on these housing construction status areas, the identity information of the cultivated land owner is matched, specifically the owner information for the corresponding housing construction status area. Based on the owner's identity information, application forms from the owners of the cultivated land in these areas are collected. These applications include applications for converting cultivated land to residential land, and their approval status is obtained. Since building on cultivated land requires a specific application process, otherwise it is considered illegal construction, obtaining these applications allows for the determination of whether the housing construction status area is illegal. Finally, based on the owner's application, the housing construction status area is marked as illegal, generating a rural cultivated land construction status verification result. By constructing a farmland area status identification model, inspectors can quickly and accurately identify the status of farmland areas and automatically obtain the identity of the farmland owner and the status of the corresponding application, thereby realizing automated monitoring and verification of rural farmland, further reducing the workload of inspectors and improving the efficiency of rural farmland monitoring and verification.

[0046] like Figure 3 As shown, the method S40 provided in this application embodiment further includes:

[0047] S41: Collect a set of images of houses built on farmland, wherein any one of the images of houses built on farmland represents a type of pre-defined farmland housing construction.

[0048] S42: Traverse the set of farmland building images and copy the images to generate a basic image set for state recognition, wherein the image copying is performed at least once;

[0049] S43: Traverse the set of farmland housing construction images to extract features and generate multiple sets of farmland housing construction feature information, wherein any set of farmland housing construction feature information corresponds to one of the set of farmland housing construction images;

[0050] S44: Based on the state recognition base image set, the multiple sets of farmland housing construction feature information are copied to construct multiple isolated tree nodes, wherein any isolated tree node stores a type of preset farmland housing construction image feature, and the data volume is at least 2 sets.

[0051] S45: Construct a farmland housing identification tree based on the multiple isolated tree nodes;

[0052] S46: Input the image information of the cultivated land area into the cultivated land area status recognition model to generate the cultivated land area status information.

[0053] Specifically, when constructing the farmland area status recognition model, images of farmland housing construction are collected through big data to obtain a farmland housing construction image set. Any image in this set represents a predefined type of farmland housing construction. The farmland housing construction image set is traversed and images are copied to generate a basic image set for status recognition. Each image copy is performed at least once to expand the sample for each type of farmland housing construction. Further, feature extraction is performed on the farmland housing construction image set to generate multiple sets of farmland housing construction feature information. Each set of farmland housing construction feature information corresponds to one image in the farmland housing construction image set. Further, based on the basic image set for status recognition (i.e., the image set from which the farmland housing construction image set was copied), the multiple sets of farmland housing construction feature information are copied to construct multiple isolated tree nodes. The isolated tree nodes are nodes in the anomaly detection tree that identify a pre-defined type of farmland housing construction image features. Multiple sets of farmland housing construction feature information are used to construct the isolated tree nodes. Each isolated tree node stores one type of pre-defined farmland housing construction image feature; that is, each isolated tree node stores one type of pre-defined farmland housing construction image feature, and the data volume is at least two sets. By combining multiple isolated tree nodes, a farmland housing construction recognition tree is constructed. Finally, the farmland area image information is input into the farmland area state recognition model to generate the farmland area state information.

[0054] The method S46 provided in this application embodiment further includes:

[0055] S461: Extract features from the cultivated land area image information to generate multiple sets of cultivated land area image features;

[0056] S462: The features of the multiple sets of cultivated land area images are sequentially input into the cultivated land area status recognition model for feature comparison and branching, generating cultivated land housing recognition tree branch results;

[0057] S463: When the branch result of the farmland construction identification tree does not include a branch with a data size of 1, the input farmland area image information is identified as the farmland construction status area.

[0058] Specifically, when inputting farmland area image information into the farmland area status recognition model, feature extraction is performed on the farmland area image information. Common feature extraction methods can be used, which will not be elaborated here, generating multiple sets of farmland area image features. Subsequently, these multiple sets of farmland area image features are sequentially input into the farmland area status recognition model for feature comparison and branching, generating farmland housing construction identification tree branch results. When the farmland housing construction identification tree branch result does not include branches with a data value of 1, meaning all features of the farmland area image feature have been identified and classified, then the farmland area image feature belongs to the farmland housing construction state, and the input farmland area image information is identified as the farmland housing construction state region. By constructing the farmland area status recognition model, accurate identification of farmland area status is achieved.

[0059] The method S46 provided in this application embodiment further includes:

[0060] S464: Based on the feature comparison analysis results, obtain the feature overlap degree of the nth model node;

[0061] S465: When the feature overlap of the nth model node is greater than or equal to the feature overlap threshold, the input farmland area image features are added to the nth model node, and the input farmland area image information is identified as the farmland housing status area.

[0062] S466: When the feature overlap of the nth model node is less than the feature overlap threshold, traverse the (n+1)th model node;

[0063] S467: When all model nodes have been traversed and the farmland housing status area has not yet been generated, a new model branch is generated based on the input image features of the farmland area, wherein the image feature data of the new model branch is 1 set.

[0064] S468: After adding the farmland housing status area and the newly added model branch to the farmland housing identification tree branch result, delete the newly added model branch.

[0065] Specifically, based on the feature comparison analysis results, the feature overlap degree of the nth model node is obtained, which is the proportion of the nth model node features in the cultivated land area image features. The nth model node is the node with the nth type of preset cultivated land housing image features. When the feature overlap degree of the nth model node is greater than or equal to the feature overlap degree threshold (which can be set according to actual conditions), the input cultivated land area image features are added to the nth model node, expanding the features in the nth model node to facilitate better feature recognition in subsequent processing, and the input cultivated land area image information is identified as the cultivated land housing state region. When the feature overlap degree of the nth model node is less than the feature overlap degree threshold, the (n+1)th model node is traversed to determine if other model nodes meet the feature overlap degree threshold. If, after traversing all model nodes, the cultivated land housing state region is still not generated, it means that the cultivated land area image features do not belong to any cultivated land housing state. Based on the input farmland area image features, a new model branch is generated, where the image feature data of the new model branch consists of one set. Since some features in the new model branch have already been identified by some model node features, this new model branch has a high probability of representing farmland construction. In this case, the new model branch is used to mark the features of this category of farmland construction, and the corresponding farmland construction status area is added to the farmland construction identification tree branch result. The new model branch is then deleted. This facilitates subsequent processing by staff, avoids processing in a fixed way that could lead to model misjudgment, and improves the accuracy of model judgment.

[0066] The method S70 provided in this application embodiment further includes:

[0067] S71: When the application from the owner of the cultivated land includes an application for conversion of the cultivated land to residential land in the area where the house is being built, obtain the approval status of the application for conversion of the residential land.

[0068] S72: When the application of the owner of the cultivated land does not include the application for conversion of the cultivated land to residential land in the area where the house is built, or the approval status is not approved or the approval status is under review, the area where the house is built is marked as having a violation status, and the verification result of the rural cultivated land house status is generated.

[0069] Specifically, when the application from the farmland owner includes an application for conversion of farmland to residential land in the area where houses are currently being built, and the application is approved, the farmland owner has already submitted an application for conversion of farmland to residential land in the area where houses are currently being built, and the application has been approved. Therefore, the farmland in this area is in a compliant state. When the application from the farmland owner does not include an application for conversion of farmland to residential land in the area where houses are currently being built, or when the approval status is not approved, or when the approval status is under review, the farmland in this area is classified as a non-compliant state. In this case, a violation status is marked for the farmland in this area, and a verification result for the rural farmland construction status is generated.

[0070] In summary, the method provided in this application identifies the cultivated land area to be verified, obtains the coordinate information of the cultivated land area, performs UAV path planning, generates UAV cruise route information, and collects cultivated land area image information. The cultivated land area image information is processed through a cultivated land area status recognition model to generate cultivated land area status information, which includes cultivated land areas with housing construction status. Based on the cultivated land construction status areas, the identity information of the cultivated land owner is matched. Based on the cultivated land owner's identity information, the application form of the cultivated land owner for the cultivated land construction status area is collected. Based on the cultivated land owner's application form, the cultivated land construction status area is marked as having a violation status, generating a rural cultivated land construction status verification result. By constructing a cultivated land area status recognition model, verification personnel can quickly and accurately identify the cultivated land area status and automatically obtain the identity of the cultivated land owner and the corresponding application form status, realizing automated monitoring and verification of rural cultivated land, further reducing the workload of verification personnel, and improving the efficiency of rural cultivated land monitoring and verification. This solves the technical problems of low monitoring and verification efficiency and slow verification progress in existing rural cultivated land monitoring and verification methods.

[0071] Example 2

[0072] Based on the same inventive concept as the monitoring and verification method for rural arable land in the foregoing embodiments, such as Figure 4 As shown, this application provides a monitoring and verification system for rural cultivated land, the system comprising:

[0073] The coordinate information acquisition module 11 is used to identify the cultivated land area of ​​the area to be verified and acquire the coordinate information of the cultivated land area.

[0074] The cruise route information acquisition module 12 is used to perform path planning based on the coordinate information of the cultivated land area and generate cruise route information for the aerial photography drone.

[0075] The cultivated land area image acquisition module 13 is used to dispatch the aerial photography drone according to the drone's cruise route information to collect cultivated land area image information.

[0076] The cultivated land area status information identification module 14 is used to process the cultivated land area image information through the cultivated land area status identification model to generate cultivated land area status information, wherein the cultivated land area status information includes cultivated land housing status areas.

[0077] The cultivated land owner identity acquisition module 15 is used to match the cultivated land owner identity information based on the area of ​​the cultivated land building status.

[0078] The cultivated land owner application form collection module 16 is used to collect the cultivated land owner application forms for the cultivated land construction status area based on the cultivated land owner's identity information;

[0079] The status verification module 17 is used to mark the illegal status of the area where houses are built on the cultivated land according to the application of the owner of the cultivated land, and generate the status verification result of the rural cultivated land construction.

[0080] Furthermore, the cruise route information acquisition module 12 is also used for:

[0081] Based on the coordinate information of the cultivated land area, path nodes are constructed, and a path node distribution network is generated.

[0082] Input the cruise start point coordinates into the path node distribution network to generate multiple cruise distance information;

[0083] The path nodes are sorted according to the multiple cruise distance information to generate multiple cruise order information;

[0084] Based on the multiple cruise sequence information, multiple cruise sub-routes are constructed;

[0085] The multiple cruise sub-routes are added to the cruise route information of the aerial photography drone.

[0086] Furthermore, the cruise route information acquisition module 12 is also used for:

[0087] Based on the multiple cruise distance information, obtain the k-th cruise distance and the (k+1)-th cruise distance;

[0088] Connect the kth path node of the kth cruise distance to the coordinate information of the cruise starting point, and connect the k+1th path node of the (k+1)th cruise distance to the coordinate information of the cruise starting point, and calculate the angle deviation.

[0089] When the angle deviation is less than or equal to the angle deviation threshold, the k-th cruise distance and the (k+1)-th cruise distance are added to the same cruise distance cluster group.

[0090] Based on the clusters with the same cruising distance, the path nodes are sorted from closest to furthest to generate the multiple cruising order information.

[0091] Furthermore, the cultivated land area status information identification module 14 is also used for:

[0092] Collect a set of images of houses built on arable land, wherein any one of the images in the set represents a type of pre-defined arable land housing construction.

[0093] The image set of farmland housing construction is traversed and copied to generate a basic image set for state recognition, wherein the image copying is performed at least once;

[0094] The image set of farmland housing construction is traversed to extract features and generate multiple sets of farmland housing construction feature information, wherein any set of farmland housing construction feature information corresponds to one of the farmland housing construction image sets;

[0095] Based on the state recognition base image set, the multiple sets of farmland housing construction feature information are copied to construct multiple isolated tree nodes. Each isolated tree node stores a type of preset farmland housing construction image feature, and the data volume is at least 2 sets.

[0096] Based on the multiple isolated tree nodes, construct a farmland housing identification tree;

[0097] The image information of the cultivated land area is input into the cultivated land area status recognition model to generate the cultivated land area status information.

[0098] Furthermore, the cultivated land area status information identification module 14 is also used for:

[0099] Feature extraction is performed on the image information of the cultivated land area to generate multiple sets of cultivated land area image features;

[0100] The features of the multiple sets of cultivated land area images are sequentially input into the cultivated land area status recognition model for feature comparison and branching, generating cultivated land housing construction recognition tree branch results;

[0101] When the branch results of the farmland housing construction identification tree do not include branches with a data size of 1, the input farmland area image information is identified as the farmland housing construction status area.

[0102] Furthermore, the cultivated land area status information identification module 14 is also used for:

[0103] Based on the feature comparison analysis results, the feature overlap degree of the nth model node is obtained;

[0104] When the feature overlap of the nth model node is greater than or equal to the feature overlap threshold, the input farmland area image features are added to the nth model node, and the input farmland area image information is identified as the farmland housing status area.

[0105] If the feature overlap of the nth model node is less than the feature overlap threshold, traverse the (n+1)th model node.

[0106] If the farmland housing status area is not generated after traversing all model nodes, a new model branch is generated based on the input image features of the farmland area. The image feature data of the new model branch is 1 set.

[0107] After adding the farmland housing status area and the newly added model branch to the farmland housing identification tree branch result, delete the newly added model branch.

[0108] Furthermore, the status verification module 17 is also used for:

[0109] When the application from the owner of the cultivated land includes an application for conversion to residential land in the area where the cultivated land is currently used for housing construction, obtain the approval status of the application for conversion to residential land.

[0110] If the application from the owner of the cultivated land does not include the application for conversion of the cultivated land to residential land in the area where the house is being built, or if the approval status is not approved or the approval status is under review, the area where the house is being built will be marked as having a violation status, and the verification result of the rural cultivated land house status will be generated.

[0111] The above-described Embodiment 2 is used to execute the method as described in Embodiment 1. Its execution principle and basis can be obtained from the content described in Embodiment 1, and will not be elaborated further here. Although this application has been described in conjunction with specific features and embodiments, this application is not limited to the exemplary embodiments described herein. Based on the embodiments of this application, those skilled in the art can make various modifications and variations to this application without departing from the scope of this application, and the content obtained in this way also falls within the protection scope of this application.

Claims

1. A method for monitoring and verifying rural arable land, characterized in that, include: Identify the cultivated land areas to be verified and obtain the coordinate information of the cultivated land areas; Based on the coordinate information of the cultivated land area, path planning is performed to generate the cruise route information of the aerial photography drone. Based on the aerial drone's cruise route information, an aerial drone is dispatched to collect image information of the cultivated land area. The cultivated land area image information is processed by the cultivated land area status recognition model to generate cultivated land area status information, wherein the cultivated land area status information includes cultivated land areas with housing construction status. Based on the area of ​​the farmland where houses are built, match the identity information of the farmland owner; Based on the identity information of the farmland owner, collect the application form of the farmland owner in the area where the farmland is under construction. Based on the application from the owner of the cultivated land, the area where houses are being built on the cultivated land is marked as having an illegal status, and a rural cultivated land housing status verification result is generated. The step of generating aerial drone cruise route information based on the coordinate information of the cultivated land area includes: Based on the coordinate information of the cultivated land area, path nodes are constructed, and a path node distribution network is generated. Input the cruise start point coordinates into the path node distribution network to generate multiple cruise distance information; The path nodes are sorted according to the multiple cruise distance information to generate multiple cruise order information; Based on the multiple cruise sequence information, multiple cruise sub-routes are constructed; Add the multiple cruise sub-routes to the cruise route information of the aerial photography drone; The step of sorting the path nodes according to the multiple cruise distance information to generate multiple cruise order information includes: Based on the multiple cruise distance information, obtain the k-th cruise distance and the (k+1)-th cruise distance; Connect the kth path node of the kth cruise distance to the coordinate information of the cruise starting point, and connect the k+1th path node of the (k+1)th cruise distance to the coordinate information of the cruise starting point, and calculate the angle deviation. When the angle deviation is less than or equal to the angle deviation threshold, the k-th cruise distance and the (k+1)-th cruise distance are added to the same cruise distance cluster group. Based on the clusters with the same cruising distance, the path nodes are sorted from closest to furthest distance to generate the multiple cruising order information; The process of processing the cultivated land area image information using the cultivated land area status recognition model generates cultivated land area status information, wherein the cultivated land area status information includes cultivated land areas with housing construction status, including: Collect a set of images of houses built on arable land, wherein any one of the images in the set represents a type of pre-defined arable land housing construction. The image set of farmland housing construction is traversed and copied to generate a basic image set for state recognition, wherein the image copying is performed at least once; The image set of farmland housing construction is traversed to extract features and generate multiple sets of farmland housing construction feature information, wherein any set of farmland housing construction feature information corresponds to one of the farmland housing construction image sets; Based on the state recognition base image set, the multiple sets of farmland housing construction feature information are copied to construct multiple isolated tree nodes. Each isolated tree node stores a type of preset farmland housing construction image feature, and the data volume is at least 2 sets. Based on the multiple isolated tree nodes, construct a farmland housing identification tree; The image information of the cultivated land area is input into the cultivated land area status recognition model to generate the cultivated land area status information.

2. The method as described in claim 1, characterized in that, The step of inputting the cultivated land area image information into the cultivated land area state recognition model to generate the cultivated land area state information includes: Feature extraction is performed on the image information of the cultivated land area to generate multiple sets of cultivated land area image features; The features of the multiple sets of cultivated land area images are sequentially input into the cultivated land area status recognition model for feature comparison and branching, generating cultivated land housing construction recognition tree branch results; When the branch results of the farmland housing construction identification tree do not include branches with a data size of 1, the input farmland area image information is identified as the farmland housing construction status area.

3. The method as described in claim 2, characterized in that, The process of sequentially inputting the multiple sets of farmland area image features into the farmland area status recognition model for feature comparison and branching, and generating farmland housing construction recognition tree branch results, includes: Based on the feature comparison analysis results, the feature overlap degree of the nth model node is obtained; When the feature overlap of the nth model node is greater than or equal to the feature overlap threshold, the input farmland area image features are added to the nth model node, and the input farmland area image information is identified as the farmland housing status area. If the feature overlap of the nth model node is less than the feature overlap threshold, traverse the (n+1)th model node. If the farmland housing status area is not generated after traversing all model nodes, a new model branch is generated based on the input image features of the farmland area. The image feature data of the new model branch is 1 set. After adding the farmland housing status area and the newly added model branch to the farmland housing identification tree branch result, delete the newly added model branch.

4. The method as described in claim 1, characterized in that, The step of marking the area of ​​farmland with illegal construction status according to the application of the farmland owner, and generating a rural farmland construction status verification result, includes: When the application from the owner of the cultivated land includes an application for conversion to residential land in the area where the cultivated land is currently used for housing construction, obtain the approval status of the application for conversion to residential land. If the application from the owner of the cultivated land does not include the application for conversion of the cultivated land to residential land in the area where the house is being built, or if the approval status is not approved or the approval status is under review, the area where the house is being built will be marked as having a violation status, and the verification result of the rural cultivated land house status will be generated.

5. A monitoring and verification system for rural arable land, characterized in that, The system is used to perform the method according to any one of claims 1-4, comprising: The coordinate information acquisition module is used to identify the cultivated land area of ​​the area to be verified and acquire the coordinate information of the cultivated land area; The cruise route information acquisition module is used to perform path planning based on the coordinate information of the cultivated land area and generate cruise route information for the aerial photography drone. The cultivated land area image acquisition module is used to dispatch the aerial photography drone according to the drone's cruise route information to collect cultivated land area image information. The cultivated land area status information identification module is used to process the cultivated land area image information through the cultivated land area status identification model to generate cultivated land area status information, wherein the cultivated land area status information includes cultivated land areas with housing construction status. The farmland owner identity acquisition module is used to match farmland owner identity information based on the farmland construction status area; The farmland ownership application collection module is used to collect farmland ownership applications in the area where farmland is under construction, based on the farmland owner's identity information. The status verification module is used to mark the illegal status of the farmland construction area based on the application form of the farmland owner, and generate the rural farmland construction status verification result.

Images