Method, device and equipment for monitoring and supervising supplied and unused land in normalized mode

Abstract

The invention relates to a method, a device and equipment for monitoring and supervising supplied and unused land in a normalized mode, and relates to the technical field of natural resource management, the method comprises the following steps: determining supplied land to be monitored, obtaining supplied land red line vector data, supplied land project business materials and multi-source image data of the supplied land, and according to the multi-source image data, determining the supplied land to be monitored according to the supplied land red line vector data, the supplied land project business materials and the multi-source image data; and performing classification monitoring processing by combining the supply land red line vector data and supply land project business materials to obtain a working condition classification state and associated monitoring data of the supplied land, and performing change monitoring on the working condition classification state and the associated monitoring data by using a preset supply and unused land change monitoring model according to a normalized supervision mechanism. According to the method, the land monitoring result of the supplied land is obtained, and the long-time-sequence tracking and active monitoring of the land development and utilization state are realized through the intelligent identification of the land working condition and the timely monitoring of the land development and utilization change, so that the timeliness and authenticity of the supervision information are improved.

Description

Technical Field

[0001] The present application relates to the field of natural resource management technology, and in particular to a method, device and equipment for monitoring and supervising unused land in a normalized mode. Background Art

[0002] "Supply but unused" land refers to the supplied construction land that has not been developed and utilized according to the established time node, which has caused a serious waste of land resources. Among them, idle land is a typical type of "supply but unused" land.

[0003] At present, the status of construction start and completion after the supply of construction land and the disposal of idle land mainly rely on self-reporting, supplemented by remote sensing image verification and supervision. However, this mechanism faces many challenges. Some potential problems in the supervision process are difficult to detect. Some areas fail to report in time. The speed of new land that is "supplied but not used" easily exceeds the speed of disposal. There are many difficulties. It can be seen that the existing land monitoring or supervision methods cannot obtain the actual construction status of the land in a timely and accurate manner. Summary of the invention

[0004] The present application provides a method, device and equipment for monitoring and supervising unused land in a normalized mode. Through intelligent identification of land construction status and timely monitoring of changes in land development and utilization, long-term tracking and active monitoring of land development and utilization status are achieved, the land supervision mechanism is improved, the timeliness and authenticity of supervision information are improved, and the intensive and efficient use of land is promoted, thereby solving the problem that existing land monitoring or supervision methods cannot obtain the actual construction status of the land in a timely and accurate manner.

[0005] First, the present application provides a method for monitoring and supervising unused land in a normalized mode, including:

[0006] Identify the supplied land to be monitored;

[0007] Obtaining land supply red line vector data, land supply project business materials and multi-source image data of the supplied land;

[0008] Based on the multi-source image data, combined with the land supply red line vector data and the land supply project business materials, classified monitoring processing is performed to obtain the working condition classification status and related monitoring data of the supplied land;

[0009] According to the normalized supervision mechanism, the preset supplied but unused land change monitoring model is used to monitor the changes of the working condition classification status and the related monitoring data to obtain the land monitoring results of the supplied land.

[0010] Optionally, obtaining the land supply red line vector data, land supply project business materials and multi-source image data of the supplied land, including:

[0011] Obtaining the land supply red line vector data and land supply project business materials of the supplied land;

[0012] Based on the land supply red line vector data, the monitoring scope of the supplied land is locked in the supervision process;

[0013] With respect to the monitoring range, continuous optical image acquisition is performed on the supplied land to obtain the multi-source image data, and the multi-source image data is used to characterize the actual working condition of the supplied land.

[0014] Optionally, classified monitoring and processing is performed based on the multi-source image data in combination with the land supply red line vector data and the land supply project business materials to obtain the working condition classification status and related monitoring data of the supplied land, including:

[0015] Performing data comparison based on the business materials of the land supply project and the vector data of the land supply red line to obtain a data comparison result;

[0016] Based on the data comparison result and the actual start and completion information in the land supply project business materials, determine the working condition classification status, and, based on the working condition classification status, determine the time difference between the agreed start time in the land supply project business materials and the current time to obtain a time mark;

[0017] Performing overlay analysis on the land supply red line vector data and the multi-source image data to obtain long-time series image data within the normalized monitoring range;

[0018] For the working condition classification state, classification monitoring is performed based on the time mark and the long time series image data to obtain related monitoring data;

[0019] The operating condition classification status includes a started state and a not started state.

[0020] Optionally, a superposition analysis is performed based on the land supply red line vector data and the multi-source image data to obtain long-term series image data within the normalized monitoring range, including:

[0021] Performing overlay analysis based on the land supply red line vector data and the multi-source image data to obtain image data of the supplied land at different periods;

[0022] Long time series image data matching the monitoring range is extracted from the image data.

[0023] Optionally, for the working condition classification state, classification monitoring is performed based on the time mark and the long time series image data to obtain associated monitoring data, including:

[0024] If the working condition classification state is the started state, then according to the time mark, the image data corresponding to the current time period is extracted from the long time series image data as the associated monitoring data;

[0025] If the working condition classification status is a non-operating state, then the associated monitoring data is extracted from the long time series image data according to a preset monthly monitoring cycle combined with a time stamp.

[0026] Optionally, the change monitoring of the working condition classification state and the associated monitoring data is performed according to the normalized supervision mechanism using a preset supplied but unused land change monitoring model to obtain the land monitoring result of the supplied land, including:

[0027] If the working condition classification state is the started state, then according to the preset supervision mechanism corresponding to the started state, the associated monitoring data is used to identify buildings through the supplied but unused land change monitoring model to obtain the land monitoring result of the supplied land;

[0028] If the working condition classification state is a non-operating state, extracting the previous phase data and the subsequent phase data from the associated monitoring data;

[0029] According to the normalized supervision mechanism, based on the previous phase data and the later phase data, the change of construction status is monitored through the provided but unused land change monitoring model to obtain the land monitoring result;

[0030] The image data of the previous month of the month corresponding to the later phase data is the earlier phase data.

[0031] Optionally, before monitoring the change of the working condition classification state and the associated monitoring data by using a preset land change monitoring model, the method further includes:

[0032] Acquire an engineering activity target recognition sample of the sample land, wherein the engineering activity target recognition sample includes a sample image and a sample vector;

[0033] respectively determining the spectral characteristics and morphological characteristics of the pile foundation image and the foundation pit image in the construction change sample;

[0034] Performing cropping processing based on the sample image and the sample vector to obtain a tile sample and a grid label tile corresponding to the tile sample;

[0035] Based on the engineering activity target recognition samples, a database of unused samples is constructed in combination with the tile samples and the grid label tiles, and based on the samples showing the features of the foundation pit protection frame and the pile foundation in the engineering activity target recognition sample images, a mixed sample library is constructed;

[0036] Using the sample data in the unused sample database and the sample data in the mixed sample database to perform data preprocessing to obtain a training set, a validation set, and a test set;

[0037] Performing model training based on the training set to obtain the unused land change monitoring model;

[0038] The test set is used to perform model testing on the unused land change monitoring model, and the verification set is used to verify and evaluate the model accuracy of the unused land change monitoring model and the accuracy of the test set.

[0039] Optionally, after obtaining the land monitoring result of the supplied land, the method further includes:

[0040] Determine a monthly monitoring report based on the land monitoring results;

[0041] Building an early warning list based on the monthly monitoring report, and issuing an early warning for the supplied land through the early warning list;

[0042] For the early warning list, based on the multi-source image data combined with field inspections, field verification results are obtained, and based on the field verification results, the supplied land is subject to enhanced supervision;

[0043] Through preset routine monitoring applications, combined with collaborative supervision methods and supervision cycles, regular monitoring after land supply will be carried out.

[0044] In a second aspect, the present application provides a device for monitoring and supervising unused land, comprising:

[0045] A supplied land determination module, used to determine the supplied land to be monitored;

[0046] A multi-source data acquisition module is used to acquire the land supply red line vector data, land supply project business materials and multi-source image data of the supplied land;

[0047] A classification monitoring and processing module is used to perform classification monitoring and processing based on the multi-source image data, combined with the land supply red line vector data and the land supply project business materials, to obtain the working condition classification status and related monitoring data of the supplied land;

[0048] The normalized monitoring module is used to monitor the changes of the working condition classification status and the related monitoring data according to the normalized supervision mechanism using the preset supplied but unused land change monitoring model to obtain the land monitoring results of the supplied land.

[0049] In a third aspect, the present application provides an electronic device, including a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other via the communication bus;

[0050] Memory, used to store computer programs;

[0051] The processor is used to implement the steps of the method for monitoring and supervising unused land in the normalized mode as described in any one of the embodiments of the first aspect when executing the program stored in the memory.

[0052] In summary, the embodiment of the present application determines the supplied land to be monitored, obtains the land supply red line vector data, land supply project business materials and multi-source image data of the supplied land, and then performs classified monitoring and processing based on the multi-source image data, combined with the land supply red line vector data and the land supply project business materials, to obtain the working condition classification status and related monitoring data of the supplied land, and then according to the normalized supervision mechanism, uses the preset supplied but unused land change monitoring model to monitor the changes in the working condition classification status and related monitoring data to obtain the land monitoring results of the supplied land. The present application realizes long-term tracking and active monitoring of the land development and utilization status through intelligent identification of land commencement status and timely monitoring of changes in land development and utilization. While improving the timeliness and authenticity of supervision information, it effectively reduces the consumption of manpower and material costs, and solves the problem that the existing unused land monitoring and supervision methods under the normalized mode cannot obtain the actual land commencement status in a timely and accurate manner. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0054] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0055] Figure 1 A flow chart of a method for monitoring and supervising supplied but unused land in a normalized mode provided in an embodiment of the present application;

[0056] Figure 2This is a schematic diagram of the steps of a method for monitoring and supervising supplied but unused land in a normalized mode provided by an optional embodiment of the present application;

[0057] Figure 3 This is a diagram of a normalized supervision mechanism for "supplied but unused" land provided by an optional embodiment of the present application;

[0058] Figure 4 It is a sample diagram of pile foundation provided by an optional example of this application;

[0059] Figure 5 It is a sample diagram of a foundation pit provided by an optional example of this application;

[0060] Figure 6 It is a pile foundation change monitoring diagram provided by an optional example of the present application;

[0061] Figure 7 It is a foundation pit change monitoring diagram provided by an optional example of the present application;

[0062] Figure 8 A structural block diagram of a device for monitoring and supervising supplied but unused land provided in an embodiment of the present application;

[0063] Fig. 9 It is a structural schematic diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0064] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0065] In the relevant technologies, the supervision of "supplied but unused" land mainly relies on the cities to report idle land, and then use image recognition supervision. This supervision method is essentially a post-event supervision, that is, when the land has reached the idle time point, its image will be analyzed, which is difficult to prevent the problem of idle land and inefficiency. In addition, the existing land monitoring technology mainly relies on remote sensing images, and automatically identifies the current status of land use and its changes through target detection technology, but this method cannot grasp the process of land development and utilization, and has certain limitations. Specifically, in the process of land construction, there is often a construction process of driving pile foundations and excavating foundation pits. Since the existing technology cannot identify pile foundations and foundation pits, it can only identify them as bare land and judge them as unstarted. The classification of detection types is still rough, and it is difficult to accurately identify the land construction status including complex construction processes such as pile foundations and foundation pits. At the same time, due to factors such as image resolution and cloud and rain occlusion, it is difficult to accurately judge by relying solely on automatic recognition of remote sensing images in the field, and a large number of manual comprehensive evaluations in the field and field surveys in the field are required to verify, which not only consumes a lot of costs, but also reduces monitoring efficiency.

[0066] In summary, the existing land monitoring and supervision technologies have the following problems: ① Lack of long-term monitoring. The accuracy of judging the construction status of a plot of land based solely on the latest single-period image data still needs to be improved. In the actual implementation of the project, some plots have complex situations such as demolition before construction, piling before covering, etc., which significantly increases the workload of manual verification in and outside the field; ② Unable to identify pile foundations and foundation pits. Land monitoring and identification technology has certain limitations in its application, especially in the automatic identification of pile foundations and foundation pits. At present, this technology mainly focuses on the identification and monitoring of land use types. For subtle and critical construction activities such as pile foundations or foundation pits, existing technologies are often difficult to accurately identify, and often misjudge them as bare land, which lacks pertinence and accuracy in the monitoring process of land development and utilization; ③ Insufficient supervision initiative. At present, the post-supply supervision of land is mainly based on local reporting. There is a lack of long-term tracking and supervision of the development and utilization status of land after supply, and potential problems are difficult to be discovered and solved in a timely and proactive manner. In the face of the problem of "land supply but not use", especially the problem of idle land, the existing mechanism has not paid enough attention to preventing the problem from the source. Therefore, it is necessary to advance the supervision time and appropriately extend it to comprehensively improve the efficiency and level of post-supply supervision of land; ④ Inadequate use of data resources. In the existing natural resource management work, the natural resources department has accumulated a wealth of satellite remote sensing images, and with the normalization of projects such as geographic national conditions monitoring, a large amount of land change monitoring data resources have not been fully utilized.

[0067] In order to solve the above-mentioned problems existing in the prior art, one of the technical concepts of the present application is to propose a monitoring and supervision method for supplied but unused land under a normalized mode, which uses multiple data such as land supply red line vector data of the supplied land, land supply project business materials and multi-source image data to carry out classified monitoring and processing, and obtain the working condition classification status and related monitoring data of the supplied land, and then according to the normalized supervision mechanism, use the supplied but unused land change monitoring model to carry out change monitoring to obtain the land monitoring results of the supplied land. The present application can combine image data and business data to jointly carry out continuous monitoring of the supplied land, and through intelligent identification of land commencement status and timely monitoring of changes in land development and utilization, it realizes long-term tracking and active monitoring of land development and utilization status, which can improve the accuracy of land commencement process judgment, automatically identify the land commencement construction status including complex construction processes such as pile foundations and foundation pits, reduce the consumption of manpower and material costs, improve the timeliness and authenticity of supervision information, and solve various problems existing in the existing monitoring and supervision method for supplied but unused land under a normalized mode.

[0068] To facilitate the understanding of the embodiments of the present application, further explanation will be given below in conjunction with the drawings and specific embodiments. The embodiments do not constitute a limitation on the embodiments of the present application.

[0069] Figure 1 A flow chart of a method for monitoring and supervising unused land in a normalized mode provided in an embodiment of the present application. Figure 1 As shown, the method for monitoring and supervising unused land in a normalized mode provided in the embodiment of the present application may specifically include the following steps:

[0070] Step 110, determine the supplied land to be monitored.

[0071] In this embodiment, the supplied land can be understood as construction land that has been supplied in accordance with legal procedures. The supplied land may include but is not limited to land that has started construction and land that has not started construction. This embodiment does not limit the quantity and size of the supplied land.

[0072] It should be noted that the method for monitoring and supervising supplied but unused land under the normalized mode provided in this application can carry out normalized monitoring and supervision on multiple supplied lands at the same time, and one or more supplied lands can be selected according to actual needs as targets to be monitored for normalized monitoring and supervision.

[0073] Step 120, obtaining the land supply red line vector data, land supply project business materials and multi-source image data of the supplied land.

[0074] Specifically, the multi-source image data is essentially image data that reflects the actual construction status of the supplied land. It can be image data obtained by photographing the supplied land. The sources of the multi-source image data may include but are not limited to multi-source satellite remote sensing data, aerial survey images, drone images, manual inspection site photos, etc.

[0075] In this embodiment, the land supply red line vector data can be used to characterize the actual scope of land that needs to be monitored to avoid monitoring irrelevant areas in multi-source image data; the land supply project business materials can be used to characterize the key information required in the land supervision and monitoring process, such as key information that can at least include the project's electronic supervision number, the start time agreed in the contract, the completion time agreed in the contract, the construction permit, the planning and construction drawings, the piling report, the completion acceptance report, and the actual start and completion information. Among them, the actual start and completion information includes the actual start / completion time, such as the actual start time and / or the actual completion time, etc. The contract-agreed start time and the contract-agreed completion time represent the estimated completion time, and the actual start / completion time represents the actual completion time.

[0076] In actual implementation, since the multi-source image data of the supplied land may contain image data other than the supplied land, that is, the actual land range represented by the multi-source image data actually captured may exceed the land area that needs to be monitored in the supplied land. Therefore, this embodiment can use the land supply red line vector data in the vector business data to lock the monitoring range of the supplied land, that is, use the vector business data to lock the image corresponding to the land range for monitoring and supervision of the supplied land in the remote sensing image data, so as to achieve the purpose of reducing the spatial scope of monitoring for the "supply but not used" business, greatly reduce the data processing volume of computer recognition monitoring, avoid the problem of waste of monitoring resources due to monitoring of irrelevant land, and improve the accuracy of land monitoring.

[0077] Step 130, based on the multi-source image data, combined with the land supply red line vector data and the land supply project business materials, classified monitoring processing is performed to obtain the working condition classification status and related monitoring data of the supplied land.

[0078] In this embodiment, the working condition classification status may include but is not limited to: the started state and the not started state; the associated monitoring data may correspond to the working condition classification status of the supplied land. According to different working condition classification states, different supervision methods can be selected to monitor and supervise the supplied land. According to the corresponding supervision method, the corresponding image data is selected from the multi-source image data as the associated monitoring data, so as to realize reasonable monitoring of the supplied land.

[0079] In the specific implementation, based on the information represented by the business materials of the land supply project and the land supply red line vector data, this embodiment divides the supplied land into a started state or a non-started state as the working condition classification state. Specifically, in this embodiment, there are two ways to determine the working classification state of the supplied land: ①, according to the agreed start time in the business materials of the land supply project and the current time, when the agreed start time is greater than or equal to the current time, the supplied land is determined to be in a started state; when the agreed start time is less than the current time, the supplied land is determined to be in a non-started state. ②, according to the actual start and completion time in the business materials of the land supply project, the working condition classification state of the supplied land is obtained. In addition, the actual start and completion time may not exist in the business materials of the land supply project. The currently supplied land can be set to a non-started state, or, it can be determined by comparing the agreed start time in the business materials of the land supply project with the current time.

[0080] In actual implementation, for supplied land that has been started, key image data from the start time of the supplied land to the present can be extracted from multi-source image data in a key monitoring manner as associated monitoring data; for supplied land that has not been started, image data for a certain period of time can be extracted from multi-source image data in a normalized monitoring and supervision mode as associated monitoring data. Therefore, this embodiment introduces land supply red line vector data, land supply project business materials and multi-source image data on the basis of the existing land monitoring and identification that only relies on remote sensing images, and combines the three types of data together for land supervision, solving the problems of low monitoring efficiency and high cost consumption caused by relying solely on remote sensing images for identification and monitoring.

[0081] Step 140, according to the normalized supervision mechanism, using the preset supplied but unused land change monitoring model to monitor the changes of the working condition classification status and the related monitoring data, to obtain the land monitoring results of the supplied land.

[0082] In this embodiment, the supplied but unused land change monitoring model can also be called: "supplied but unused" land change monitoring model or "supplied but unused" change monitoring model, etc. Of course, it can also be directly referred to as a change monitoring model or a land monitoring model. The specific name of the model is not limited in this embodiment. Among them, the supplied but unused land change monitoring model can be used to conduct continuous change monitoring of supplied land in different working condition classification states, and the embodiments of the present application do not limit this. By monitoring the changes of the supplied land with the supplied but unused land change monitoring model, it can be determined whether the supplied land has "supplied but unused", "idle land" and "false commencement" and other situations.

[0083] In the specific implementation, for the supplied land whose working condition classification status is the started state, the associated monitoring data can be continuously monitored for changes according to the normalized supervision mechanism using the supplied but unused land change monitoring model. By monitoring and supervising the supplied land, the actual construction situation of the supplied land can be determined. When the actual construction situation is inconsistent with the construction expectation represented by the land supply project business materials (which can be judged in combination with the site project business materials and the current time), it indicates that the supplied land may have problems such as "false commencement", "supply but not use", "idleness", etc. As a land monitoring result, it indicates that the supplied land is a problem plot. At this time, the problem plot can be strengthened for supervision and timely warning. When the actual construction situation is inconsistent with the construction expectation represented by the land supply project business materials, it can be determined that the supplied land does not currently have problems such as "supply but not use". As a normal supplied land, continuous supervision and monitoring is carried out through the supplied but unused land change monitoring model to avoid being called "supplied but not used" land.

[0084] For supplied land whose working condition classification status is not started, a normalized mode monitoring can be set up. According to a certain period, combined with the normalized supervision mechanism, the change monitoring model of supplied but unused land can be used to continuously and periodically monitor the changes of related monitoring data. Once a problem is found, timely intervention can be made to achieve the purpose of proactively preventing the waste of land resources. Based on the normalized mode monitoring of the supplied land, the time nodes are grasped, and the monitoring and supervision of the "supplied but unused" and "idle" land problems are intervened in advance, leaving enough time to prevent the waste of resources caused by the undeveloped land, and realize the transformation of land use supervision from end-point governance to front-end control.

[0085] In actual implementation, the unused land change monitoring model of this embodiment can also integrate image features such as pile foundations and foundation pits during the model training stage to realize the identification of pile foundations and foundation pits. Therefore, in actual applications, the model can accurately and effectively identify the construction conditions of pile foundations and foundation pits based on image features such as pile foundations and foundation pits represented by multi-source image data, which solves the technical bottleneck of the existing unused land change monitoring model that roughly classifies the engineering activities of pile foundations and foundation pits as bare land, and can more accurately reflect the actual situation of engineering activities on the supplied land. This application will not elaborate on this.

[0086] It can be seen that the embodiment of the present application determines the supplied land to be monitored, obtains the land supply red line vector data, land supply project business materials and multi-source image data of the supplied land, and then performs classified monitoring and processing based on the multi-source image data, combined with the land supply red line vector data and the land supply project business materials, to obtain the working condition classification status and related monitoring data of the supplied land, and then according to the normalized supervision mechanism, uses the preset supplied but unused land change monitoring model to monitor the changes in the working condition classification status and related monitoring data to obtain the land monitoring results of the supplied land. The present application realizes long-term tracking and active monitoring of the land development and utilization status through intelligent identification of land commencement status and timely monitoring of changes in land development and utilization. While improving the timeliness and authenticity of supervision information, it effectively reduces the consumption of manpower and material costs, and solves the problem that the existing unused land monitoring and supervision methods under the normalized mode cannot obtain the actual land commencement status in a timely and accurate manner.

[0087] Reference Figure 2 , shows a schematic flow chart of the steps of a method for monitoring and supervising unused land in a normalized mode provided by an optional embodiment of the present application. The method may specifically include the following steps:

[0088] Step 210, determine the supplied land to be monitored.

[0089] For the description of step 210, reference may be made to the above step 110, which will not be described in detail in this embodiment.

[0090] Step 220, obtaining the land supply red line vector data of the supplied land and the land supply project business materials.

[0091] Step 230, based on the land supply red line vector data, lock the monitoring scope of the supplied land in the supervision process.

[0092] A unified description of steps 220 to 230 is given as follows:

[0093] In this embodiment, the land supply red line vector data can lock the monitoring scope of the supplied land in the supervision process, making the monitoring work more targeted. The land supply red line vector data can be formed by collecting the land supply red line data of each region, and the land supply red line vector data can be used to lock the monitoring scope of the supplied land in the supplied land, and then multi-source image data can be obtained for the monitoring scope; the land supply project business materials (or business materials of the supplied land) contain a lot of key information required for land monitoring or land supervision. This embodiment prepares relevant business materials as land supply project business materials, so as to ensure the integrity and accuracy of data information and provide a judgment basis for subsequent continuous supervision.

[0094] It should be noted that the monitoring and supervision method for unused land under the normalized mode provided by this application can be applied to the field of natural resource management. Use basic image data such as high-resolution satellite orthophotos and aerial images as multi-source image data. Collect business data such as annual land change survey data results, land use status data, provincial land supply red line data, and normalized monitoring map data from various regions. In this way, these data can be combined to comprehensively consider the ex ante prevention and supervision of "unused" land, and use multidisciplinary technical means such as land management, remote sensing, and geographic information to deeply analyze the structural characteristics and attribute information of each data, and extract effective information that serves the monitoring and supervision of "unused" land, and construct a composite construction progress monitoring model of "remote sensing intelligent interpretation + manual field verification", and establish a set of methods and processes for normalized monitoring and supervision of "unused" land.

[0095] Step 240 , for the monitoring range, continuously collect optical images of the supplied land to obtain the multi-source image data.

[0096] The multi-source image data is used to characterize the actual working conditions of the supplied land.

[0097] In this embodiment, the image acquisition rules can be used to continuously acquire optical images of the supplied land within the monitoring range to obtain multi-source image data. The image acquisition rules can include but are not limited to: acquisition cycle, acquisition source, etc. Among them, the acquisition cycle is used to periodically obtain / acquire multi-source image data from the acquisition source. The acquisition cycle can be real-time acquisition or intermittent acquisition, such as once a month.

[0098] In actual implementation, there may be two situations for continuous image acquisition of the supplied land within the monitoring scope: ① First, directly lock the supplied land through the monitoring scope and perform continuous image acquisition. The multi-source image data collected can directly correspond to the monitoring scope, avoiding acquisition of the range outside the supplied land, and reducing the amount of calculation for subsequent processing of image data; ② First, continuously acquire images of the supplied land according to the preset image acquisition rules to obtain image data. At this time, the collected image data may contain images outside the monitoring scope. The monitoring scope can be used to lock the image data corresponding to the monitoring scope, and the image cropping and other methods can be used to crop the area corresponding to the monitoring scope from the image data to obtain multi-source image data, or the image identifier (such as an identifier box) can be used to calibrate the area corresponding to the monitoring scope in the image data. Subsequently, only the image processing needs to be performed on the image corresponding to the calibrated area, which can also effectively reduce the amount of calculation for processing image data.

[0099] In actual processing, when this embodiment collects multi-source image data, it can try to ensure that the corresponding image data can be obtained every month for the monitored plots, so as to meet the timely, high-frequency, and long-time monitoring needs of plot changes. Therefore, this implementation targets the timeliness and stage characteristics of engineering activity elements in land development and utilization, as well as the uncertainty of remote sensing image shooting. By deeply analyzing the landmark engineering activity elements and construction process steps in each stage of construction activities, it explores and establishes a mapping relationship between remote sensing image engineering activity target elements and construction activities. By combining multi-source image data, land supply red line vector data, and land supply project business materials, it is used for regular monitoring and supervision of supplied land, thereby ensuring that the established mapping relationship can correspond to the relevant links in the land use life cycle, providing a solid foundation for the subsequent "supply but not use" identification.

[0100] In the specific implementation, since the construction process is time-sensitive, it is very important to implement high-frequency monitoring of the development and utilization of land. By comparing the images of the land at different times, the changes can be extracted and a set of normalized monitoring mechanisms can be established. This normalized monitoring mechanism can be used to conduct continuous monitoring of "supplied but unused" land. Among them, the normalized monitoring mechanism can also be called the normalized supervision mechanism of "supplied but unused" land. Figure 3 As shown in the figure, the implementation process of the normalized monitoring mechanism can be divided into four stages / steps: data preparation stage, data preprocessing, normalized change monitoring of "supplied but unused" land, and verification feedback. In the data preparation stage, the monitoring scope can be determined using the land supply red line vector data of the supplied land, the land supply project business materials can be used to determine the land working condition classification status and related monitoring data, and the actual engineering status can be determined based on multi-source image data.

[0101] Furthermore, in the process of routine monitoring and supervision of supplied land, a diversified target identification sub-model can be constructed, and through the coordinated application of diversified target identification sub-models, an efficient monitoring model for changes in supplied but unused land can be constructed. This embodiment of the present application will not be described in detail.

[0102] Step 250, performing data comparison based on the land supply project business materials and the land supply red line vector data to obtain a data comparison result.

[0103] Specifically, the data comparison result is used to characterize the process verification status corresponding to each key information in the land supply project business materials. Specifically, this embodiment uses the land supply project business materials and the land supply red line vector data for data comparison (or data comparison), associates the business data with the vector data, and performs a detailed data comparison to check the key information contained in the project business materials to obtain the data comparison results, where the checked information mainly includes but is not limited to: the project's electronic supervision number, the start time agreed in the contract, the completion time agreed in the contract, the construction permit, the planning and construction drawings, the piling report, the completion acceptance report, the actual start and completion time, etc.

[0104] Step 260, based on the data comparison result and the actual start and completion information in the land supply project business materials, determine the working condition classification status, and, based on the working condition classification status, determine the time difference between the agreed start time in the land supply project business materials and the current time to obtain a time mark.

[0105] The operating condition classification status includes a started state and a not started state.

[0106] In this embodiment, the actual data comparison situation can be determined based on the data comparison result. When the data comparison situation is consistent, the supplied land can be divided into a started state or a non-started state based on the actual start and completion information in the land supply project business materials, and the current time can be combined with the agreed start time in the land supply project business materials (the start time agreed in the contract) to calculate the time difference between the agreed start time and the current time as a time identifier.

[0107] In the specific implementation, through data comparison, the relevant business materials of "unfinished" projects, the land supply red line vector data and the "unfinished" supplied land can be screened out. The "unfinished" supplied land refers to the land area that has "started" but has not completed all the engineering goals, as well as the land area that has "not started".

[0108] In the specific implementation, the land supply red line vector data of the unfinished project is superimposed and analyzed with the image data of different times, and the different image times are attached to the attribute data of the land supply red line vector data, so that the land supply red line vector data has a time mark corresponding to the image data. In addition, the time difference ΔT between the project in the "unstarted" state and the agreed start time is calculated according to the current time. The long-term vector data of the "unfinished" project and the corresponding image data are the basic data required to monitor the "supplied but not used" land.

[0109] For example, the time difference ΔT between the agreed start time and the current time can be calculated and identified by the following formula (1):

[0110]

[0111] Furthermore, for the plots in the "started" state, the image data is combined to verify that there are actually engineering activities for development and utilization of the plots, that is, step 270-step 290 is executed; for the plots in the "not started" state, the normalized monitoring mode is used to continuously monitor the progress of engineering activities in the plots. Among them, the plots that have not reached the agreed start time (that is, ΔT≥0) should be prevented from becoming "supplied but not used" land that "should be started but is not started", and the plots that have exceeded the agreed start time will be the focus of supervision (that is, ΔT<0), and supervision will be strengthened to prevent them from further becoming "idle land".

[0112] Step 270, performing overlay analysis based on the land supply red line vector data and the multi-source image data to obtain long time series image data within the normalized monitoring range.

[0113] In the specific implementation, according to the monitoring range represented by the red line vector data, the multi-source image data of the corresponding monitoring range is sorted in a certain time sequence to obtain long-time series image data. Subsequently, the corresponding associated monitoring data can be determined for the working condition classification status of the supplied land, and then the associated monitoring data can be used for continuous monitoring.

[0114] In an optional embodiment, the embodiment of the present application performs overlay analysis based on the land supply red line vector data and the multi-source image data to obtain long-term series image data within the normalized monitoring range, which may specifically include: performing overlay analysis based on the land supply red line vector data and the multi-source image data to obtain image data of the supplied land at different periods; extracting long-term series image data matching the monitoring range from the image data.

[0115] In actual implementation, for the supplied land, the multi-source image data is sorted in a certain time sequence using the shooting time corresponding to the multi-source image data, and the image data of the supplied land in different periods is obtained as multi-period image data. Subsequently, the red line vector data and the multi-source image data are combined for overlay analysis. For the monitoring range represented by the red line vector data, the long time series image data matching the monitoring range is extracted from the multi-period image data, and then the long time series image data matching the spatial range of each land supply project is obtained, thereby narrowing the monitoring range of normalized changes, reducing the amount of data processing, and effectively improving the monitoring efficiency and pertinence.

[0116] Step 280 , for the working condition classification status, classification monitoring is performed based on the time mark and the long time series image data to obtain associated monitoring data.

[0117] In actual implementation, for different working condition classification states, a normalized monitoring method can be used in combination with different supervision methods to meet the normalized mode monitoring and supervision needs of different working conditions. Specifically, for supplied land whose working condition classification state is the started state, the image data of the current time period or the most recent time period can be obtained from the long-term series image data as the associated monitoring data, which is used to verify whether the actual working condition of the supplied land is consistent with the construction situation corresponding to the started state; for supplied land whose working condition classification state is not started, the normalized supervision mechanism can be used to carry out continuous change monitoring of the progress of engineering activities in the land parcel, and the image data of a certain time period or a periodic time period can be obtained from the long-term series image data as the associated monitoring data. It can be seen that this embodiment uses the normalized mode to monitor the land after supply, which can effectively grasp the time node, intervene in the monitoring and supervision of the problems of "supplied but not used" and "idle" land in advance, leave enough time to prevent the waste of resources caused by the undeveloped land, and realize the transformation of land use supervision from end-point governance to front-end control.

[0118] In an optional embodiment, the embodiment of the present application performs classified monitoring on the operating condition classification status based on the time stamp and the long time series image data to obtain associated monitoring data, which may specifically include: if the operating condition classification status is the started state, then according to the time stamp, the image data corresponding to the current time period is extracted from the long time series image data as the associated monitoring data; if the operating condition classification status is the non-started state, then according to the preset monthly monitoring cycle combined with the time stamp, the associated monitoring data is extracted from the long time series image data.

[0119] For example, for the supplied land whose working condition classification status is not started, the monthly monitoring cycle can be used as the preset monthly monitoring cycle. According to the time stamp, two consecutive months of image data are extracted from the long time series image data as associated monitoring data. The associated monitoring data can characterize the changes in engineering activities of the supplied land within two consecutive months, so that the associated monitoring data can be used to achieve continuous monitoring in the future.

[0120] Step 290, in accordance with the normalized supervision mechanism, using the preset supplied but unused land change monitoring model to monitor the changes of the working condition classification status and the associated monitoring data, to obtain the land monitoring results of the supplied land.

[0121] In a specific implementation, for supplied land whose working condition classification status is that it has been started, this embodiment can use the supplied but unused land change monitoring model to perform verification monitoring on the associated monitoring data, that is, use the image data contained in the associated monitoring data to verify that there are actually engineering activities for development and utilization of the land. The supplied but unused land change monitoring model can identify the image data, such as identifying whether there are buildings on the land, etc., to verify and ensure that the supplied land has indeed been started for construction, and obtain the land monitoring results of the supplied land. For supplied land whose working condition classification status is not started, this embodiment can set the supplied land to a normalized monitoring and supervision mode, and according to the normalized supervision mechanism, combined with a certain period, use the supplied but unused land change monitoring model to monitor the changes in the associated monitoring data to obtain the land monitoring results of the supplied land.

[0122] Therefore, the embodiments of the present application shift from the existing end-point governance model of simply monitoring the current situation to a management method based on monitoring changes in the spatiotemporal dimension, thereby realizing front-end prevention. The front-end governance of the "supplied but unused" land normalization supervision process is utilized to monitor land use from the spatiotemporal dimension. By optimizing the monitoring model for changes in supplied but unused land, the integration of "supplied but unused" land supervision and normalization mode is realized. Satellite image data, normalized change map automatic extraction data, and natural resource business data are fully utilized to reduce the data resource investment cost of natural resource management departments in idle land supervision. By scientifically and rationally integrating and utilizing data, the operating cost in the natural resource field is reduced, making a positive contribution to promoting the sustainable development of natural resource management.

[0123] In an optional embodiment, the above-mentioned normalized supervision mechanism is used to monitor the changes of the working condition classification status and the related monitoring data using the preset supplied but unused land change monitoring model to obtain the land monitoring result of the supplied land, which may specifically include: if the working condition classification status is the started state, then according to the preset supervision mechanism corresponding to the started state, the associated monitoring data is used to identify the buildings to obtain the land monitoring result of the supplied land; if the working condition classification status is the unstarted state, the previous phase data and the later phase data are extracted from the associated monitoring data; according to the normalized supervision mechanism, based on the previous phase data and the later phase data, the construction status change is monitored by the supplied but unused land change monitoring model to obtain the land monitoring result; wherein, the image data of the previous month corresponding to the later phase data is the previous phase data.

[0124] In the specific implementation, this embodiment further subdivides the screened "unfinished" land into "started" and "not started" according to the data start status. For the "started" plots, it is necessary to verify through monitoring that there are indeed engineering activities that have been developed and utilized on the plots. The most intuitive verification is that there are existing buildings on the plots, and there is a construction progress of the building from the time and space dimensions; for the "not started" plots, according to the marked time difference ΔT, it is further divided into plots that are agreed to be started soon (ΔT≥0) and plots that are overdue for construction (ΔT<0) to carry out change monitoring. In addition, for the "not started" plots, the monitoring cycle is monthly, that is, the monthly monitoring cycle. Change monitoring can be carried out on the plot every month. After the change monitoring is completed, the plot time mark is updated. At the same time, the construction status of the plot is updated and recorded until the plot is confirmed to be a "started" plot after verification.

[0125] For example, for a plot of land that has not been started, any month can be randomly selected and the image data corresponding to that month can be extracted from the associated monitoring data as the post-phase data. Based on that month, the image data of the month before that month can be obtained as the pre-phase data set.

[0126] In actual implementation, the existing technology often generally identifies the construction process of driving pile foundations and excavating foundation pits as bare land, and judges it as an unstarted state. The classification of detection types is still rough, and it is difficult to accurately identify the land construction status that includes complex construction processes such as pile foundations and foundation pits. In order to solve the problems existing in the existing technology, the embodiment of the present application introduces a change monitoring model for unused land to monitor changes in related monitoring data. In order to ensure that the change monitoring model for unused land can effectively identify the construction process, the present application can construct and collect a large amount of sample data of pile foundations and foundation pits for deep learning training of the model. Since the change monitoring model for unused land is based on the change monitoring pre-training model and the target recognition model, this embodiment produces a sample library that meets the "unused" supervision requirements, uses deep learning methods to train and optimize the model, and constructs a targeted change monitoring model for unused land.

[0127] In an optional embodiment, before the embodiment of the present application performs change monitoring on the working condition classification state and the associated monitoring data through a preset land change monitoring model, the embodiment of the present application may also specifically include: obtaining an engineering activity target identification sample of the sample land, the engineering activity target identification sample comprising a sample image and a sample vector; respectively determining the spectral characteristics and morphological characteristics of the pile foundation image and the foundation pit image in the construction change sample; performing cropping based on the sample image and the sample vector to obtain a tile sample and a grid label tile corresponding to the tile sample; based on the engineering activity target identification sample, combining the tile sample and the grid label tile to construct A database of unused samples, and a mixed sample library based on samples showing the features of foundation pit protection frames and pile foundations in the sample images of the engineering activity target identification; data preprocessing is performed using the sample data in the unused sample database and the sample data in the mixed sample library to obtain a training set, a validation set and a test set; model training is performed based on the training set to obtain the unused land change monitoring model; wherein the test set is used to perform model testing on the unused land change monitoring model, and the validation set is used to verify and evaluate the model accuracy of the unused land change monitoring model and the accuracy of the test set.

[0128] In the specific implementation, refer to Figure 3 , for the sample data preparation stage, the embodiment of the present application acquires a variety of sample data by collecting and preprocessing the sample data, including: ① finding the image data before the actual start time from the "started" data, and the image before the pile foundation or foundation pit is used as the pre-phase data, and the image in the period of pile foundation or foundation pit is used as the post-phase data; ② manually outlining the vector range of engineering changes in the pre-phase and post-phase; ③ identifying the specific change type and its corresponding image time and image name in the vector data; ④ using the outlined vector data to crop the corresponding image data to obtain image tiles, and the vector data is converted into raster label tiles containing attribute information. Then, samples are made for the acquired sample data. Specifically, according to relevant regulations and requirements, the start of construction and development is closely related to the pile foundation driven or the foundation pit excavated in the land engineering activities. For construction activities such as digging foundation pits or driving foundation piles, samples for engineering activity target recognition are collected, including collecting construction change samples based on the pre- and post-high-resolution remote sensing images, and collecting construction element samples based on single-period high-resolution remote sensing images. The engineering activity target recognition samples are composed of sample images (the images used for sample annotation) and sample vectors (the vector data generated in the sample annotation process).

[0129] In actual implementation, when making samples, it is preferred to select clear, cloudless images with as accurate geometric positions as possible. When marking samples, outline vectors according to the boundaries of the changed areas or features and record specific categories. In view of the different land use types and construction requirements, the morphology of pile foundations or foundation pits will vary. Therefore, when making samples, it is necessary to ensure that as many morphological types as possible are covered, and give priority to those samples with obvious features and representativeness. Tile sample production is based on sample image and sample vector cropping. The cropped tile sample consists of image tiles (small tiles generated after sample image cropping) and label tiles (small tiles cropped after sample vector rasterization). The two correspond one to one, and tile samples are input data for deep learning model training.

[0130] It should be noted that for image data containing pile foundations, Figure 4 As shown in the figure, the morphological features of the pile foundation in the image are mainly regular and closely arranged dot-matrix structures. The shape of the pile foundation dot matrix is ​​mostly similar to the shape of the supported building, but its spectral characteristics are more complex and changeable. Some pile foundations still have exposed soil, while others have been poured with cement. These pile foundations in different states show the characteristics of "same object, different spectrum" in the image. For image data containing foundation pits, such as Figure 5 As shown in the figure, the morphological characteristics of the foundation pit are mainly reflected in the protective frame structure built and the shadow characteristics formed at the edge of the foundation pit. The middle part of the foundation pit still retains a large area of ​​soil cover, while the shadow at the edge is narrow and not obvious. This leads to the edge area of ​​the foundation pit being easily misjudged as bare land in practical applications, which increases the difficulty of identification. To ensure the accuracy of identification, areas with obvious protective frame structures are preferred as foundation pit training samples in sample preparation.

[0131] In this embodiment, after the sample production is completed, a sample database can be further constructed (such as the sample database can be a gdb database), and the sample information of the pile foundation and the foundation pit can be stored in the corresponding database, that is, the vector sample data with the pile foundation and the foundation pit and the corresponding image data are stored in the sample database, and the vector and image corresponding to the data that has been verified and started in the field are also updated to the sample database by outlining the changed area and re-marking. Among them, the sample database can include a first sample library and a second sample library, and the first sample library can be used to simultaneously store vector sample data with pile foundations, vector sample data with foundation pits, and vectors and images after outlining the changed area and re-marking. Because in actual work, there are special situations in which some high-rise buildings need to dig foundation pits and pile driving operations at the same time during the construction process. In this special situation, the features of the foundation pit protection frame and the pile foundation may be shown in the image at the same time, which is also a start-up state. For this, a mixed sample library is established, and the second sample library is used as a mixed sample library to store such samples with composite features.

[0132] In this embodiment, after completing the construction of the sample database, model training can be carried out. The selected sample data are randomly divided into a training set, a validation set and a test set. According to the training results, the optimized samples are re-made and the relevant parameters of the model training are adjusted until the recognition results of the test set basically meet the monitoring requirements, and a trained land change monitoring model is obtained.

[0133] In order to improve the speed and accuracy of model training, the embodiment of the present application can also perform data preprocessing on the sample data, including normalization, enhancement, etc. Data enhancement can handle sample distribution problems, such as adding random disturbances, rotating pictures, etc. When the data distribution is reasonable, the training set, validation set, and test set are uniformly and randomly divided. The division ratio can be determined according to the business type and the number of samples. After setting the model hyperparameters, start training. After the model training is completed, the model accuracy can be evaluated, including the training set, test set accuracy, training speed, and inference speed. Among them, the importance of the test set accuracy can be higher, so that when the model accuracy does not meet the requirements, the problem can be diagnosed and re-operated. The re-operation mainly includes but is not limited to: re-enhancing samples, dividing data, etc.; when there is a problem with the model structure, return to the construction stage and reset; when there is a problem with the hyperparameters, the hyperparameters can be reset. Finally, a trained land change monitoring model that is not used is formed.

[0134] Furthermore, after completing the model training, the model can be applied. The pre-processed "construction started" and "not started" monitoring data can be input into the unused land change monitoring model. The model can be used to automatically identify and extract the changes in land use type and construction progress of the plots, so as to realize monthly monitoring of the progress of engineering activities on the supplied plots.

[0135] For example, the pile foundation change monitoring diagram is as follows: Figure 6 As shown in the figure, the foundation pit change monitoring diagram is as follows Figure 7 As shown in the figure, according to the pile foundation change monitoring diagram and the foundation pit change monitoring diagram, the image data corresponding to the pile foundation and the foundation pit have obvious changes in the previous and next phases. The change monitoring model of supplied but unused land can quickly identify the changes in the construction progress, thereby realizing the monthly monitoring of the progress of the engineering activities of the supplied plots.

[0136] Therefore, the embodiment of the present application sets pile foundations and foundation pits as target elements of remote sensing intelligent monitoring of land development and utilization engineering activities through systematic combing and analysis, integrates the long-term dimension, and establishes a mapping relationship between pile foundations and foundation pits in engineering activities and the actual land development and utilization process. By using the spectral characteristics and morphological characteristics of pile foundations and foundation pits in optical images, the land change monitoring model that has not been used is introduced into the land construction process monitoring, and the existing change monitoring technology is optimized by combining deep learning to achieve intelligent interpretation and recognition of pile foundations and foundation pits in remote sensing images, improve the accuracy and efficiency of automatic recognition of land development and utilization engineering activities in remote sensing images, effectively obtain the dynamic changes of engineering construction, and provide strong support for the monitoring and management of land development and utilization.

[0137] In an optional embodiment, after the embodiment of the present application obtains the land monitoring results of the supplied land, it may further specifically include: determining a monitoring monthly report based on the land monitoring results; constructing an early warning list based on the monitoring monthly report, and issuing an early warning for the supplied land through the early warning list; for the early warning list, obtaining a field verification result based on the multi-source image data combined with field inspections, and strengthening the supervision of the supplied land based on the field verification results; through a preset normalized monitoring application, combined with a collaborative supervision method and supervision cycle, carrying out normalized monitoring after land supply.

[0138] In the specific implementation, refer to Figure 3According to the land monitoring results, the changes in the supplied land can be determined, and the plots that have not changed in the construction status for a long time can be sorted out, statistically summarized, and formed into a monitoring monthly report to timely grasp the construction and development situation. Then, based on the monitoring monthly report, an early warning list can be constructed directly through field verification and other methods, such as using field verification to monitor the "supplied but not used" land, sorting out the monitoring results, and obtaining the actual inspection results. The early warning list can list the plots that are reflected in the image as not started. If the land parcel is in the "started" state, but the inspection image shows that it is a land parcel that has not started construction, there may be a "false start" problem. The suspected "false start" land parcel can be marked and continuously monitored for changes; for the land parcel that is in the "not started" state within three months before the agreed start time, there is no engineering activity after normalized monitoring, or it is indeed not started three months before the agreed start time, a "supply but not used" warning can be issued; if the land parcel is in the "not started" state, there is no engineering activity after normalized monitoring, and it has exceeded the agreed start time and still has not started, then a "idle land" warning can be issued. In addition, this embodiment can also further confirm whether these land parcels have "false start", "supply but not used", "idle" and other problems in a combination of multiple images and field inspections. According to the results of the field inspection, the problem land parcels are supervised more strictly, and the land parcels are used as new samples to update the samples of the first sample library, so as to further optimize and improve the accuracy of model monitoring.

[0139] Furthermore, this embodiment can be used for normalized monitoring, and normalized monitoring after land supply can be carried out on a monthly basis to achieve multi-departmental collaborative supervision of "monthly monitoring and quarterly verification". Therefore, this embodiment of the application effectively improves the level of supervision and control of the land system after supply, and through monthly monitoring and statistics, it can identify the status of land development and utilization, actively discover potential problems, and issue early warnings, thereby improving the accuracy of internal verification in idle land supervision work, helping responsible persons at all levels to grasp the status and disposal process of the land, coordinate and promote disposal work, prevent false reports such as "false commencement", and after reasonable screening of land areas, use key areas as field verification objects to reduce unnecessary field work and reduce the cost of field verification.

[0140] In summary, the embodiment of the present application determines the supplied land to be monitored, obtains the land supply red line vector data and the land supply project business materials of the supplied land, and based on the land supply red line vector data, locks the monitoring scope of the supplied land in the supervision process, so as to continuously collect optical images of the supplied land for the monitoring scope to obtain multi-source image data, and then compares the data according to the land supply project business materials and the land supply red line vector data to obtain the data comparison result, and determines the time difference between the agreed start time and the current time for the land supply project business materials based on the data comparison result to obtain the time mark, and determines the working condition classification status based on the time mark, and performs overlay analysis based on the land supply red line vector data and the multi-source image data to obtain long-term series image data within the normalized monitoring range, and then performs classified monitoring on the working condition classification status based on the time mark and the long-term series image data to obtain related monitoring data, and finally, according to the normalized supervision mechanism, uses the preset supplied but unused land change monitoring model to monitor the changes of the working condition classification status and the related monitoring data to obtain the land monitoring results of the supplied land. Therefore, the present application achieves: 1. Establishing the mapping relationship between the target elements of engineering activities in remote sensing images and the actual land development and utilization process, so as to provide a solid foundation for the identification of "supply but not use"; 2. Introducing pile foundation and foundation pit samples in the model training stage of the supply but not use land change monitoring model, so as to quickly identify the pile foundation and foundation pit, effectively combine the morphological characteristics and spectral characteristics of the pile foundation and foundation pit, improve the recognition efficiency and accuracy of the remote sensing model, reduce the workload of manual verification, and solve the problem of false reporting of the current land construction status; 3. By comparing the images of the land at different times, extracting the changed parts, and establishing a set of normalized monitoring mechanisms, it is possible to grasp the progress of the construction in a timely manner. It can be seen that the embodiment of the present application realizes the long-term tracking and active monitoring of the land development and utilization status, improves the timeliness and authenticity of the supervision information, and solves the various problems existing in the existing monitoring and supervision methods of supply but not use land under the normalized mode.

[0141] It should be noted that, for the purpose of simple description, the method embodiments are expressed as a series of action combinations, but those skilled in the art should be aware that the embodiments of the present application are not limited to the described order of actions, because according to the embodiments of the present application, certain steps may be performed in other orders or simultaneously.

[0142] like Figure 8 As shown, the embodiment of the present application further provides a device 800 for monitoring and supervising supplied but unused land, comprising:

[0143] A supplied land determination module 810 is used to determine the supplied land to be monitored;

[0144] The multi-source data acquisition module 820 is used to acquire the land supply red line vector data, land supply project business materials and multi-source image data of the supplied land;

[0145] The classification monitoring processing module 830 is used to perform classification monitoring processing based on the multi-source image data, combined with the land supply red line vector data and the land supply project business materials, to obtain the working condition classification status and related monitoring data of the supplied land;

[0146] The normalized monitoring module 840 is used to monitor the changes of the working condition classification status and the related monitoring data according to the normalized supervision mechanism using the preset supplied but unused land change monitoring model to obtain the land monitoring results of the supplied land.

[0147] Optionally, the multi-source data acquisition module 820 includes:

[0148] The first data acquisition submodule acquires the land supply red line vector data and land supply project business materials of the supplied land;

[0149] A range locking submodule, used to lock the monitoring range of the supplied land in the supervision process based on the land supply red line vector data;

[0150] The image acquisition submodule is used to continuously acquire optical images of the supplied land within the monitoring range to obtain the multi-source image data, and the multi-source image data is used to characterize the actual working condition of the supplied land.

[0151] Optionally, the classification monitoring processing module 830 includes:

[0152] A data comparison submodule is used to compare the data according to the land supply project business materials and the land supply red line vector data to obtain a data comparison result;

[0153] A time statistics submodule is used to determine the working condition classification status based on the data comparison result and the actual start and completion information in the land supply project business materials, and, based on the working condition classification status, determine the time difference between the agreed start time in the land supply project business materials and the current time to obtain a time mark;

[0154] The overlay analysis submodule is used to perform overlay analysis based on the land supply red line vector data and the multi-source image data to obtain long-time series image data within the normalized monitoring range;

[0155] The classification monitoring submodule is used to perform classification monitoring on the working condition classification status based on the time mark and the long time series image data to obtain related monitoring data; wherein the working condition classification status includes a started state and a non-started state.

[0156] Optionally, the superposition analysis submodule includes:

[0157] A superposition analysis unit, used for performing superposition analysis based on the land supply red line vector data and the multi-source image data to obtain the image data of the supplied land at different periods;

[0158] The long time series image data extraction unit is used to extract the long time series image data matching the monitoring range from the image data.

[0159] Optionally, the classification monitoring submodule includes:

[0160] A first classification unit is used to extract image data corresponding to a current time period from the long time series image data according to the time identifier when the working condition classification state is the started state, so as to serve as the associated monitoring data;

[0161] The second classification unit is used to extract relevant monitoring data from the long time series image data according to a preset monthly monitoring cycle combined with a time mark when the working condition classification state is a non-operating state.

[0162] Optionally, the normalized monitoring module 840 includes:

[0163] A first monitoring result determination submodule is used for, when the working condition classification state is the started state, according to the preset supervision mechanism corresponding to the started state, using the supplied but unused land change monitoring model to identify buildings on the associated monitoring data, and obtaining the land monitoring result of the supplied land;

[0164] The second monitoring result determination submodule is used to extract the previous phase data and the later phase data from the associated monitoring data when the working condition classification state is the non-start state; according to the normalized supervision mechanism, based on the previous phase data and the later phase data, the change monitoring of the construction status is performed through the supplied but unused land change monitoring model to obtain the land monitoring result; wherein, the image data of the previous month of the month corresponding to the later phase data is the previous phase data.

[0165] Optionally, the unused land monitoring and supervision device 800 further includes:

[0166] A sample set acquisition module, used to acquire an engineering activity target recognition sample of a sample land, wherein the engineering activity target recognition sample includes a sample image and a sample vector;

[0167] A feature determination module, used to respectively determine the spectral features and morphological features of the pile foundation image and the foundation pit image in the construction change sample;

[0168] A cropping module, used for performing cropping processing based on the sample image and the sample vector to obtain a tile sample and a grid label tile corresponding to the tile sample;

[0169] A sample library construction module is used to construct a sample database based on the engineering activity target recognition sample, in combination with the tile sample and the grid label tile, and to construct a mixed sample library based on samples that simultaneously show the features of the foundation pit protection frame and the pile foundation in the engineering activity target recognition sample image;

[0170] A data preprocessing module, used to perform data preprocessing using the sample data in the unused sample database and the sample data in the mixed sample database to obtain a training set, a validation set and a test set;

[0171] A model training module is used to perform model training based on the training set to obtain the unused land change monitoring model; wherein the test set is used to perform model testing on the unused land change monitoring model, and the verification set is used to verify and evaluate the model accuracy of the unused land change monitoring model and the accuracy of the test set.

[0172] Optionally, the unused land monitoring and supervision device 800 further includes:

[0173] A monitoring monthly report determination module, used to determine a monitoring monthly report based on the land monitoring results;

[0174] A list building and early warning module, used to build an early warning list based on the monitoring monthly report, and to issue an early warning for the supplied land through the early warning list;

[0175] An inspection and supervision module is used to obtain field inspection results based on the multi-source image data combined with field inspections for the early warning list, and to strengthen supervision of the supplied land based on the field inspection results;

[0176] The regular monitoring module is used to carry out regular monitoring after land supply through preset regular monitoring applications, combined with collaborative supervision methods and supervision cycles.

[0177] It should be noted that the unused land monitoring and supervision device provided in the embodiment of the present application can execute the unused land monitoring and supervision method in the normalized mode provided in any embodiment of the present application, and has the corresponding functions and beneficial effects of the execution method.

[0178] In a specific implementation, the above-mentioned monitoring and supervision device for supplied but unused land can be integrated into a device, so that the device can perform classified monitoring based on multi-source image data of supplied land, combined with land supply red line vector data and land supply project business materials, to obtain land monitoring results. As an electronic device, it can realize long-term tracking and active monitoring of land development and utilization status, and improve the timeliness and authenticity of supervision information. The electronic device can be composed of two or more physical entities, or it can be composed of one physical entity. For example, the electronic device can be a personal computer (PC), a computer, a server, etc., and the embodiments of the present application do not impose specific restrictions on this.

[0179] like Fig. 9 As shown, an embodiment of the present application provides an electronic device, including a processor 111, a communication interface 112, a memory 113 and a communication bus 114, wherein the processor 111, the communication interface 112, and the memory 113 complete mutual communication through the communication bus 114; the memory 113 is used to store computer programs; the processor 111 is used to implement the steps of the monitoring and supervision method of unused land in the normalized mode provided by any of the above-mentioned method embodiments when executing the program stored in the memory 113. Exemplarily, the steps of the monitoring and supervision method of unused land in the normalized mode may include the following steps: determining the supplied land to be monitored; obtaining the land supply red line vector data, land supply project business materials and multi-source image data of the supplied land; according to the multi-source image data, combined with the land supply red line vector data and the land supply project business materials, classified monitoring and processing are performed to obtain the working condition classification status and related monitoring data of the supplied land; according to the normalized supervision mechanism, the preset unused land change monitoring model is used to monitor the changes of the working condition classification status and the related monitoring data to obtain the land monitoring results of the supplied land.

[0180] An embodiment of the present application also provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the steps of the method for monitoring and supervising unused land in a normalized mode provided in any one of the aforementioned method embodiments are implemented.

[0181] It should be noted that, in this article, relational terms such as "first" and "second" are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0182] The above description is only a specific implementation of the present application, so that those skilled in the art can understand or implement the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest range consistent with the principles and novel features applied for herein.

Claims

1. A method for monitoring and supervising unused land in a normalized mode, characterized in that: include: Identify the supplied land to be monitored; Obtaining land supply red line vector data, land supply project business materials and multi-source image data of the supplied land; Based on the multi-source image data, combined with the land supply red line vector data and the land supply project business materials, classified monitoring processing is performed to obtain the working condition classification status and related monitoring data of the supplied land; According to the normalized supervision mechanism, the preset supplied but unused land change monitoring model is used to monitor the changes of the working condition classification status and the related monitoring data to obtain the land monitoring results of the supplied land.

2. The method according to claim 1, characterized in that Obtain the land supply red line vector data, land supply project business materials and multi-source image data of the supplied land, including: Obtaining the land supply red line vector data and land supply project business materials of the supplied land; Based on the land supply red line vector data, the monitoring scope of the supplied land is locked in the supervision process; With respect to the monitoring range, continuous optical image acquisition is performed on the supplied land to obtain the multi-source image data, and the multi-source image data is used to characterize the actual working condition of the supplied land.

3. The method according to claim 2, characterized in that Based on the multi-source image data, combined with the land supply red line vector data and the land supply project business materials, classified monitoring processing is performed to obtain the working condition classification status and related monitoring data of the supplied land, including: Performing data comparison based on the business materials of the land supply project and the vector data of the land supply red line to obtain a data comparison result; Based on the data comparison result and the actual start and completion information in the land supply project business materials, determine the working condition classification status, and, based on the working condition classification status, determine the time difference between the agreed start time in the land supply project business materials and the current time to obtain a time mark; Performing overlay analysis on the land supply red line vector data and the multi-source image data to obtain long-time series image data within the normalized monitoring range; For the working condition classification state, classification monitoring is performed based on the time mark and the long time series image data to obtain related monitoring data; The operating condition classification status includes a started state and a not started state.

4. The method according to claim 3, characterized in that Based on the overlay analysis of the land supply red line vector data and the multi-source image data, long-time series image data within the normalized monitoring range is obtained, including: Performing overlay analysis based on the land supply red line vector data and the multi-source image data to obtain image data of the supplied land at different periods; Long time series image data matching the monitoring range is extracted from the image data.

5. The method according to claim 3, characterized in that: For the working condition classification state, classification monitoring is performed based on the time mark and the long time series image data to obtain related monitoring data, including: If the working condition classification state is the started state, then according to the time mark, the image data corresponding to the current time period is extracted from the long time series image data as the associated monitoring data; If the working condition classification status is a non-operating state, then the associated monitoring data is extracted from the long time series image data according to a preset monthly monitoring cycle combined with a time stamp.

6. The method according to claim 5, characterized in that According to the normalized supervision mechanism, the preset supplied but unused land change monitoring model is used to monitor the changes of the working condition classification status and the associated monitoring data to obtain the land monitoring results of the supplied land, including: If the working condition classification state is the started state, then according to the preset supervision mechanism corresponding to the started state, the associated monitoring data is used to identify buildings through the supplied but unused land change monitoring model to obtain the land monitoring result of the supplied land; If the working condition classification state is a non-operating state, extracting the previous phase data and the subsequent phase data from the associated monitoring data; According to the normalized supervision mechanism, based on the previous phase data and the later phase data, the change of construction status is monitored through the provided but unused land change monitoring model to obtain the land monitoring result; The image data of the previous month of the month corresponding to the later phase data is the earlier phase data.

7. The method according to any one of claims 1 to 6, characterized in that: Before monitoring the change of the working condition classification state and the associated monitoring data by using the preset unused land change monitoring model, the method further includes: Acquire an engineering activity target recognition sample of the sample land, wherein the engineering activity target recognition sample includes a sample image and a sample vector; Respectively determining the spectral characteristics and morphological characteristics of the pile foundation image and the foundation pit image in the construction change sample; Performing cropping processing based on the sample image and the sample vector to obtain a tile sample and a grid label tile corresponding to the tile sample; Based on the engineering activity target recognition samples, a database of unused samples is constructed in combination with the tile samples and the grid label tiles, and based on the samples showing the features of the foundation pit protection frame and the pile foundation in the engineering activity target recognition sample images, a mixed sample library is constructed; Using the sample data in the unused sample database and the sample data in the mixed sample database to perform data preprocessing to obtain a training set, a validation set, and a test set; Performing model training based on the training set to obtain the unused land change monitoring model; The test set is used to perform model testing on the unused land change monitoring model, and the verification set is used to verify and evaluate the model accuracy of the unused land change monitoring model and the accuracy of the test set.

8. The method according to claim 7, characterized in that After obtaining the land monitoring results of the supplied land, it also includes: Determine a monthly monitoring report based on the land monitoring results; Building an early warning list based on the monthly monitoring report, and issuing an early warning for the supplied land through the early warning list; For the early warning list, based on the multi-source image data combined with field inspections, field verification results are obtained, and based on the field verification results, the supplied land is subject to enhanced supervision; Through preset routine monitoring applications, combined with collaborative supervision methods and supervision cycles, regular monitoring after land supply will be carried out.

9. A device for monitoring and supervising unused land, characterized in that: include: A supplied land determination module, used to determine the supplied land to be monitored; A multi-source data acquisition module is used to acquire the land supply red line vector data, land supply project business materials and multi-source image data of the supplied land; A classification monitoring and processing module is used to perform classification monitoring and processing based on the multi-source image data, combined with the land supply red line vector data and the land supply project business materials, to obtain the working condition classification status and related monitoring data of the supplied land; The normalized monitoring module is used to monitor the changes of the working condition classification status and the related monitoring data according to the normalized supervision mechanism using the preset supplied but unused land change monitoring model to obtain the land monitoring results of the supplied land.

10. An electronic device, characterized in that: It includes a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other through the communication bus; Memory, used to store computer programs; The processor is used to implement the steps of the method for monitoring and supervising unused land in the normalized mode as described in any one of claims 1 to 8 when executing the program stored in the memory.

Images