Project supervision system and method for natural resource and ecological protection and restoration project

Abstract

The invention relates to the technical field of ecological protection and restoration project supervision, and discloses a project supervision system and method for natural resources and ecological protection and restoration projects. The project supervision method for the natural resource and ecological protection and restoration project comprises the following steps: identifying project elements from a construction image and a construction drawing of the natural resource and ecological protection and restoration project; collecting ecological species information in the construction process; structuring the collected information to obtain first structured data, second structured data and ecological flow data, then inputting the data into a project dynamic monitoring model, and decoding to obtain problems existing in project elements in sub-projects, total projects and construction images; according to the invention, machine learning and artificial intelligence are utilized to carry out long-time-sequence dynamic monitoring on the constructed sub-project and the engineering elements of the sub-project, and problems existing in the constructed project are evaluated through correlation analysis of the engineering elements and introduction of ecological flow data, and solution suggestions are corrected through corridor increase.

Description

Technical Field

[0001] The present invention relates to the technical field of ecological protection and restoration project supervision, and more specifically, to a project supervision system and method for natural resource and ecological protection and restoration projects. Background Art

[0002] Due to the large amount of work, complex systems, and high funding requirements, large-scale ecological restoration projects usually take about 10 years to construct, and some projects even take longer. Long-term persistence and continuous optimization are the key to the success of ecological restoration. Although the project will undergo various expert evaluations in the early stages, the long time span and dynamic changes in the ecological environment will still bring uncontrollable factors. Therefore, it is necessary to monitor the construction process of each sub-project. Due to the long time span and personnel mobility, it is difficult for experts to monitor the entire process manually. The general measure is to design phased evaluation indicators, and then judge the effect of the sub-project construction process by collecting data. It is impossible to evaluate the existing problems and improvement measures through the dynamic changes of the sub-projects after construction. Summary of the invention

[0003] The present invention provides a project supervision system and method for natural resource and ecological protection and restoration projects, which solves the technical problem in related technologies that existing problems and improvement measures cannot be evaluated through dynamic changes of sub-projects after construction.

[0004] The present invention provides a project supervision method for a natural resource and ecological protection and restoration project, comprising: identifying project elements from construction images and construction drawings of the natural resource and ecological protection and restoration project;

[0005] Collect ecological species information during the construction process;

[0006] Constructing first structured data based on the engineering elements in the construction drawings, the first structured data including a structure dimension, and the basic structure elements included in the structure dimension are objects;

[0007] The second structured data is constructed based on the engineering elements in the construction image, wherein the second structured data includes two structure dimensions, wherein the basic structure elements included in the first structure dimension are objects, and the basic structure elements included in the second structure dimension are time series units;

[0008] constructing ecological flow data based on the second structured data, the ecological flow data including ecological flow units, and one ecological flow unit includes ecological flow information between two specific objects in the second structured data;

[0009] The first structured data, the second structured data, and the ecological flow data are input into the engineering dynamic monitoring model, and the engineering dynamic monitoring model includes a first branch, a second branch, a first fusion branch, and a second fusion branch, wherein the first branch includes a first encoder, the first encoder inputs the first structured data, and outputs the first latent vectors of all objects of the first structured data; the second branch includes a second encoder and a third encoder, the second encoder sequentially inputs the associated specific features of the objects of the second structured data, and sequentially outputs the second latent vectors of the objects of the second structured data to the third encoder, the third encoder inputs the ecological flow data, and the third encoder outputs the third latent vectors of the objects of the second structured data;

[0010] The first fusion branch is used to concatenate the first latent vector and the third latent vector and then decode them to obtain the problems existing in the engineering elements in the sub-project, the overall project and the construction image; the second fusion branch is used to concatenate the first latent vector and the third latent vector and then decode them to obtain the result of the corridor that needs to be added between the engineering elements in the construction image.

[0011] Furthermore, the specific object in the second structured data refers to the engineering element identified in the construction image.

[0012] Furthermore, the objects of the first structured data include sub-projects, general projects, and engineering elements identified in construction drawings.

[0013] Furthermore, objects of the second structured data include: sub-projects, overall projects, and engineering elements identified in the construction images.

[0014] Furthermore, the method for constructing the first structured data includes:

[0015] extracting a first structured data object;

[0016] Establishing a mapping between the extracted objects and their associated specific features;

[0017] For any two objects, if there is a correlation between the associated specific features related to ecological protection and restoration or there is a correlation between the objects related to ecological protection and restoration, an object relationship is established between the two objects.

[0018] Furthermore, the object of the first structured data and its associated specific features are as follows:

[0019] Specific characteristics associated with the engineering elements identified in the construction drawings include: functional descriptions, areas, and construction instructions of the engineering elements identified in the construction drawings;

[0020] Specific features associated with subprojects include: overview and construction drawings of subprojects of natural resource and ecological protection and restoration projects;

[0021] Specific characteristics associated with the overall project include: the overall overview and objectives of the natural resources and ecological protection and restoration project.

[0022] Furthermore, the method for constructing the second structured data includes:

[0023] extracting an object of second structured data;

[0024] Extract specific features associated with the object, and divide the extracted specific features into multiple parts according to the collection period, and map one part to a time sequence unit. The order of the time sequence units is consistent with the time order of the collection period of the specific features;

[0025] Establishing a mapping between the extracted objects and their associated specific features;

[0026] For any two objects, if there is a correlation between the associated specific features related to ecological protection and restoration or there is a correlation between the objects related to ecological protection and restoration, an object relationship is established between the two objects.

[0027] Furthermore, the object of the second structured data and its associated specific features are as follows:

[0028] The specific features associated with the engineering elements identified in the construction image include: information on the species and area of ​​the engineering elements identified in the construction image;

[0029] Specific characteristics associated with subprojects include: the construction schedule of the subproject;

[0030] Specific characteristics associated with the master project include: the construction progress of the master project.

[0031] Furthermore, the relevance to ecological protection and restoration includes:

[0032] Subprojects belong to the overall project;

[0033] There are dependencies between subprojects;

[0034] Engineering elements identified in construction drawings or engineering elements identified in construction images belong to sub-items;

[0035] The engineering elements identified in the construction drawings are adjacent on the construction drawings;

[0036] The engineering elements identified in the construction image are adjacent to each other on the construction image;

[0037] The functions of the engineering elements identified in the construction drawings are the same;

[0038] The information of species contained in the engineering elements identified in the construction images is similar.

[0039] The present invention provides a project supervision system for natural resource and ecological protection and restoration projects, comprising:

[0040] An element recognition module is used to identify engineering elements from construction images and construction drawings of natural resource and ecological protection and restoration projects;

[0041] Ecological information collection module, which is used to collect ecological species information during the construction process;

[0042] A first data structuring module, which constructs first structured data based on engineering elements in the construction drawings, wherein the first structured data includes a structure dimension, and the basic structure elements included in the structure dimension are objects;

[0043] A second data structuring module, which constructs second structured data based on the engineering elements in the construction image, wherein the second structured data includes two structure dimensions, wherein the basic structure elements included in the first structure dimension are objects, and the basic structure elements included in the second structure dimension are time sequence units;

[0044] A flow data module, which is used to construct ecological flow data based on the second structured data, the ecological flow data including ecological flow units, and an ecological flow unit contains ecological flow information between two specific objects in the second structured data;

[0045] A dynamic monitoring module, which is used to input the first structured data, the second structured data, and the ecological flow data into the engineering dynamic monitoring model, the engineering dynamic monitoring model includes a first branch, a second branch, a first fusion branch, and a second fusion branch, wherein the first branch includes a first encoder, the first encoder inputs the first structured data, and outputs the first latent vectors of all objects of the first structured data; the second branch includes a second encoder and a third encoder, the second encoder sequentially inputs the associated specific features of the objects of the second structured data, and sequentially outputs the second latent vectors of the objects of the second structured data to the third encoder, the third encoder inputs the ecological flow data, and the third encoder outputs the third latent vectors of the objects of the second structured data;

[0046] The first fusion branch is used to concatenate the first latent vector and the third latent vector and then decode them to obtain the problems existing in the engineering elements in the sub-project, the overall project and the construction image; the second fusion branch is used to concatenate the first latent vector and the third latent vector and then decode them to obtain the result of the corridor that needs to be added between the engineering elements in the construction image.

[0047] The beneficial effects of the present invention are: using machine learning and artificial intelligence to conduct long-term dynamic monitoring of sub-projects that have been constructed and the engineering elements of the sub-projects, and evaluating the problems existing in the constructed projects through correlation analysis of engineering elements and the introduction of ecological flow data, and proposing solutions by correcting them through the addition of corridors. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 It is a flow chart of the project supervision method of the natural resource and ecological protection and restoration project of the present invention.

[0049] Figure 2 It is a schematic diagram of the engineering element identification results.

[0050] Figure 3 This is a schematic diagram of ecological species monitoring data.

[0051] Figure 4 It is a schematic diagram of structured data.

[0052] Figure 5 This is a schematic diagram of ecological flow data. DETAILED DESCRIPTION

[0053] The subject matter described herein will now be discussed with reference to example implementations. It should be understood that the discussion of these implementations is only to enable those skilled in the art to better understand and implement the subject matter described herein, and the functions and arrangements of the elements discussed may be changed without departing from the scope of protection of the present specification. Various examples may omit, replace, or add various processes or components as needed. In addition, the features described in some examples may also be combined in other examples.

[0054] At least one embodiment of the present invention discloses a project supervision method for natural resource and ecological protection and restoration projects, such as Figure 1 As shown, the following steps are included:

[0055] Step 101, identifying engineering elements from a construction image of a natural resource and ecological protection and restoration project, where the engineering elements include patches, corridors, and matrix;

[0056] Identify engineering elements from construction drawings of natural resource and ecological protection and restoration projects;

[0057] It should be noted that the construction image is a presentation of the interim results of the unfinished construction, so the engineering elements and combinations of engineering elements contained in the construction image are different from the engineering elements and combinations of engineering elements in the construction drawings.

[0058] Step 102, collecting ecological species information during the construction process, the ecological species information includes species distribution and species flow information;

[0059] The movement of species includes the diffusion and migration of species.

[0060] Step 103, constructing first structured data based on the engineering elements in the construction drawings, the first structured data including a structure dimension, the basic structural elements included in the structure dimension are objects, and the objects of the first structured data include sub-projects, general projects, and engineering elements identified in the construction drawings;

[0061] Step 104, constructing second structured data based on the engineering elements in the construction image, the second structured data includes two structure dimensions, the basic structure elements included in the first structure dimension are objects, and the objects of the second structured data include: sub-projects, general projects, and engineering elements identified in the construction image; the basic structure elements included in the second structure dimension are time series units;

[0062] Step 105, constructing ecological flow data based on the second structured data, the ecological flow data including ecological flow units, one ecological flow unit including ecological flow information between two specific objects in the second structured data, the ecological flow information including species diffusion information and species migration information;

[0063] The specific objects in the second structured data refer to engineering elements identified in the construction image;

[0064] Step 106, inputting the first structured data, the second structured data, and the ecological flow data into a dynamic monitoring model for an engineering project, wherein the dynamic monitoring model for an engineering project includes a first branch, a second branch, a first fusion branch, and a second fusion branch, wherein the first branch includes a first encoder, the first encoder inputs the first structured data, and outputs a first latent vector of all objects of the first structured data;

[0065] The second branch includes a second encoder and a third encoder, the second encoder sequentially inputs the associated specific features of the object of the second structure data, and sequentially outputs the second latent vector of the object of the second structure data to the third encoder, the third encoder inputs the ecological flow data, and the third encoder outputs the third latent vector of the object of the second structure data;

[0066] The first fusion branch is used to decode the first latent vector and the third latent vector after splicing them, and decode to obtain the problems existing in the engineering elements in the sub-project, the overall project and the construction image;

[0067] The second fusion branch is used to decode the first latent vector and the third latent vector after splicing them, and decode to obtain the result of the corridor that needs to be added between the engineering elements in the construction image.

[0068] In some embodiments of the present invention, the first encoder is a multi-layer structure, and the calculation formula of the first layer is as follows:

[0069]

[0070] represents the first object identification feature of the vth object of the first structured data of the lth layer, represents the first object identification feature of the u-th object of the first structured data of the l-1-th layer, and Respectively represent the sets of objects that have object relationships with the vth object and the uth object of the first structured data, represents the cardinality of a set, represents the plane recognition weight matrix of the lth layer of the first encoder, , E is the total number of layers of the first encoder, when l=1 , Indicates the specific features associated with the u-th object of the first structured data, when l=E is equal to the first latent vector of the vth object of the first structured data, is the sigmoid function;

[0071] In some embodiments of the present invention, the calculation formula of the second encoder is as follows:

[0072]

[0073]

[0074]

[0075] in The t-th second latent vector of the v-th object representing the second structured data, Represents the t-1th second latent vector of the vth object of the second structured data, and Respectively represent the first and second intermediate features of the vth object of the second structured data, The t-th time sequence unit representing the associated specific feature of the v-th object of the second structured data, and are the first and second serial recognition weight matrices, is the serial identification bias parameter, Represents the Sigmoid function.

[0076] In some embodiments of the present invention, the third encoder is a multi-layer structure, and the calculation formula of the first layer is as follows:

[0077]

[0078]

[0079] in represents the aggregate representation of the vth object of the second structured data of the lth layer, and The second object identification features of the vth object of the second structured data of the lth layer and the l-1th layer, respectively, A set of objects representing an object relationship with the object v of the second structured data; Represents the universal quantifier symbol; The second object identification feature of the u-th object of the second structured data of the l-1-th layer, , E is the total number of layers of the third encoder (the data size is close to that of the first encoder input, and the structure is similar, so the two have the same number of layers, the default value is 3, and the maximum value is 5), when l=1 , , and Represents the last second latent vector of the u-th and v-th objects of the second structured data, when l=E is equal to the third latent vector of the vth object of the second structured data, represents the aggregation function (concatenation function or average pooling function) of the lth layer, represents the plane recognition weight matrix of the lth layer of the third encoder, σ represents the Sigmoid function, represents the concatenation function, Represents the ecological flow information between the objects v and u of the second structured data (if there is no ecological flow information, a vector with all component values ​​0 is input instead).

[0080] In some embodiments of the present invention, the calculation formula of the first fusion branch is as follows:

[0081]

[0082] in The question vector representing the vth object of the second structured data, one component of which represents the probability of a question type, and the question type with the largest probability value is selected as the output. a set of all objects representing the first structured data, Represents the first latent vector of the vth object of the first structured data, The third latent vector representing the vth object of the second structured data, represents the first decoding weight parameter, Represents the first decoding bias parameter.

[0083] In some embodiments of the present invention, the calculation formula of the second fusion branch is as follows:

[0084]

[0085] in The corridor reconstruction vector between the vth object and the uth object of the second structured data is represented. One component of the corridor reconstruction vector represents the probability of a corridor type. The corridor type with the largest probability value is selected as the output. The corridor type contains an empty type, which means that the corridor is not constructed. Represents the first latent vector of the vth object of the first structured data, and Respectively represent the third latent vectors of the vth and uth objects of the second structured data, represents the second decoding weight parameter, Represents the second decoding bias parameter.

[0086] Here, corridor transformation vectors are output only between objects of the second structured data corresponding to engineering elements that do not belong to the corridor.

[0087] The corridor type output by the second fusion branch is used as the corridor that needs to be added. A corridor type expresses some corridors that may be implemented. The type here can be expressed as an ID, and an ID is assigned to the optional corridor as the corridor type.

[0088] The first fusion branch mentioned above is used to analyze the problem type and can be trained through general supervised learning methods. The training samples can be obtained through expert labeling methods.

[0089] The aforementioned second fusion branch is used to output suggestions for corridor transformation. Here, it can also be trained through general supervised learning methods, but the standard training samples are more difficult. Further consideration can be given to using reinforcement learning methods for training. The environment is configured as a simulation environment, and the action is configured to transform the corridor through the output corridor type. The reward is the reduction in the number of problems of the object after the action is executed.

[0090] In some embodiments of the present invention, the method of constructing the first structured data includes:

[0091] extracting a first structured data object;

[0092] Establishing a mapping between the extracted objects and their associated specific features;

[0093] For any two objects, if there is a correlation between the associated specific features related to ecological protection and restoration or there is a correlation between the objects related to ecological protection and restoration, then an object relationship is established between the two objects;

[0094] In some embodiments of the present invention, the method of constructing the second structured data includes:

[0095] extracting an object of second structured data;

[0096] Extract specific features associated with the object, and divide the extracted specific features into multiple parts according to the collection period, and map one part to a time sequence unit. The order of the time sequence units is consistent with the time order of the collection period of the specific features;

[0097] Establishing a mapping between the extracted objects and their associated specific features;

[0098] For any two objects, if there is a correlation between the associated specific features related to ecological protection and restoration or there is a correlation between the objects related to ecological protection and restoration, then an object relationship is established between the two objects;

[0099] In some embodiments of the present invention, the object of the first structured data and the specific features associated therewith are as follows:

[0100] Specific features associated with the engineering elements identified in the construction drawings include: functional description, area, construction description, etc. of the engineering elements identified in the construction drawings.

[0101] Specific features associated with subprojects include: overviews and construction drawings of subprojects of natural resources and ecological protection and restoration projects.

[0102] Specific characteristics associated with the overall project include: the overall overview and objectives of the natural resources and ecological protection and restoration project.

[0103] In some embodiments of the present invention, the object of the second structured data and the specific features associated therewith are as follows:

[0104] The specific features associated with the engineering elements identified in the construction image include: information on the species (variety, quantity, etc.) and area of ​​the engineering elements identified in the construction image;

[0105] Specific characteristics associated with subprojects include: the construction schedule of the subproject;

[0106] Specific characteristics associated with the master project include: the construction progress of the master project.

[0107] In some embodiments of the present invention, the associations related to ecological protection and restoration include:

[0108] Subprojects belong to the overall project;

[0109] There are dependencies between subprojects;

[0110] Engineering elements identified in construction drawings or engineering elements identified in construction images belong to sub-items;

[0111] The engineering elements identified in the construction drawings are adjacent on the construction drawings;

[0112] The engineering elements identified in the construction image are adjacent to each other on the construction image;

[0113] The functions of the engineering elements identified in the construction drawings are the same;

[0114] The information of species contained in the engineering elements identified in the construction images is similar (the number of identical species is greater than 40% of the total number of species contained in the two).

[0115] In at least one embodiment of the present invention, a project supervision system for natural resource and ecological protection and restoration projects is disclosed, which includes:

[0116] An element recognition module is used to identify engineering elements from construction images and construction drawings of natural resource and ecological protection and restoration projects;

[0117] Ecological information collection module, which is used to collect ecological species information during the construction process;

[0118] A first data structuring module, which constructs first structured data based on engineering elements in the construction drawings, wherein the first structured data includes a structure dimension, and the basic structure elements included in the structure dimension are objects;

[0119] A second data structuring module, which constructs second structured data based on the engineering elements in the construction image, wherein the second structured data includes two structure dimensions, wherein the basic structure elements included in the first structure dimension are objects, and the basic structure elements included in the second structure dimension are time sequence units;

[0120] A flow data module, which is used to construct ecological flow data based on the second structured data, the ecological flow data including ecological flow units, and an ecological flow unit contains ecological flow information between two specific objects in the second structured data;

[0121] A dynamic monitoring module, which is used to input the first structured data, the second structured data, and the ecological flow data into the engineering dynamic monitoring model, the engineering dynamic monitoring model includes a first branch, a second branch, a first fusion branch, and a second fusion branch, wherein the first branch includes a first encoder, the first encoder inputs the first structured data, and outputs the first latent vectors of all objects of the first structured data; the second branch includes a second encoder and a third encoder, the second encoder sequentially inputs the associated specific features of the objects of the second structured data, and sequentially outputs the second latent vectors of the objects of the second structured data to the third encoder, the third encoder inputs the ecological flow data, and the third encoder outputs the third latent vectors of the objects of the second structured data;

[0122] The first fusion branch is used to concatenate the first latent vector and the third latent vector and then decode them to obtain the problems existing in the engineering elements in the sub-project, the overall project and the construction image; the second fusion branch is used to concatenate the first latent vector and the third latent vector and then decode them to obtain the result of the corridor that needs to be added between the engineering elements in the construction image.

[0123] At least one embodiment of the present invention discloses a project supervision platform for natural resource and ecological protection and restoration projects, which implements full-process supervision for all projects and sub-projects involved in the natural resource and ecological protection and restoration projects, and includes a project full-process management unit and an archive management unit, wherein the archive management unit is used to manage the digital archives of the project;

[0124] The project management unit includes:

[0125] Project information module, which is used to edit and manage information of projects and sub-projects;

[0126] Survey data module, which is used to edit ecological survey information for management projects;

[0127] Investment plan management module, which is used to edit and manage investment plans;

[0128] Survey and design management module, which is used to edit and manage survey and design plans;

[0129] Feasibility study report management module, which is used to edit and manage the feasibility report of the project;

[0130] The bidding document module is used to manage the bidding documents of the project;

[0131] Contract filing module, which is used to manage the contract documents of the project;

[0132] Schedule module, which is used to edit and manage project progress information;

[0133] Problem feedback module, which is used to edit and manage engineering problem information;

[0134] Acceptance report module, which is used to edit and manage project acceptance reports;

[0135] Performance evaluation module, which is used to edit management project performance evaluation information;

[0136] The restoration effect implementation monitoring system is realized based on the project supervision platform of the aforementioned natural resources and ecological protection and restoration projects. The restoration effect implementation monitoring system is to make an overall plan for the ecological restoration effectiveness evaluation system, set performance goals and monitoring plans for each indicator system.

[0137] According to the construction projects and regional ecological conditions, comprehensive data monitoring and compilation of regional ecological protection are achieved through satellite remote sensing, investigation and research, data compilation, on-site sampling, testing and analysis and other technologies or working methods.

[0138] At the same time, big data technology is used to conduct comprehensive analysis and display of the construction status of the project and its sub-projects, regional ecological conditions, survey data, and effectiveness evaluation to achieve overall visualization of the project.

[0139] Based on the aforementioned project supervision platform for natural resources and ecological protection and restoration projects, a real-time supervision system for the progress of the full process of project funds is implemented. The real-time supervision system for the progress of the full process of project funds mainly provides full-process technical monitoring services for the investment control, cost monitoring, economic indicator analysis, and compliance use of funds for each sub-project of the natural resources and ecological protection and restoration project. The monitoring service content mainly includes: project budget control, budget evaluation monitoring, fund tracking and performance evaluation, dynamic economic indicator analysis, project price review and supervision, etc. At the same time, big data technology is used to conduct a comprehensive analysis and display of the progress of funds for the project and sub-projects to achieve overall visualization of the project.

[0140] Based on the aforementioned project supervision platform for natural resources and ecological protection and restoration projects, the digital archives and review system for the entire life cycle of the project is realized. The digital archives and review system for the entire life cycle of the project is to establish a digital archive management system and a project review information construction plan. Digital archives ensure the integration and management of a large amount of data, technical achievements and other materials generated during the project. These data can be deeply analyzed through computational management to provide intelligent support for future planning and decision-making of the project. The review system includes sub-project acceptance, ecological protection and restoration unit evaluation, overall project acceptance and project review standardized review reports and processes. At the same time, digital archives are used to retain information throughout the entire process of each project to assist the full process review system.

[0141] In some embodiments of the present invention, a project supervision platform for natural resources and ecological protection and restoration projects also includes a system management unit, which is used for user management, role management, unit management, menu management, dictionary management and log management.

[0142] An example of a practical application scenario of a project supervision method for natural resource and ecological protection and restoration projects:

[0143] Take a wetland restoration project as an example. The project is located in a certain city in a certain province, with a total area of ​​about 5,000 hectares. The project goal is to restore the structure and function of the wetland ecosystem, increase biodiversity, and improve the quality of the regional ecological environment within 3-5 years.

[0144] The main sub-projects involved in this project include:

[0145] Returning farmland to wetlands: returning farmland to wetlands in the reclaimed areas, restoring natural wetlands of about 2,000 hectares;

[0146] Water system connectivity: dredging the internal water system of the wetland to increase connectivity with external water bodies and improve the water environment;

[0147] Vegetation restoration: Implement vegetation restoration on degraded wetlands and plant about 1,000 hectares of wetland plants such as reeds and cattails;

[0148] Wetland park construction: plan and build wetland parks, and carry out ecological tourism, science education and other activities.

[0149] Figure 2 It is a schematic diagram of the engineering element identification results.

[0150] Figure 3 This is a schematic diagram of ecological species monitoring data (excerpt).

[0151] Figure 4 It is a schematic diagram of structured data.

[0152] Figure 5 This is a schematic diagram of ecological flow data.

[0153] Taking the subproject of returning farmland to wetlands as an example, the detailed implementation process of the method of the present invention is explained.

[0154] Feature identification:

[0155] Input data: Construction drawings and construction site images of the returning farmland to wetland subproject.

[0156] Processing process:

[0157] Analyze the construction drawings to identify key information such as the location, area, and construction requirements of the patches in the returning farmland to wetlands area;

[0158] Perform semantic segmentation on construction site images to identify completed patches of returning farmland to wetlands.

[0159] Output results: vector boundaries and attribute information of patches in the returning farmland to wetland areas.

[0160] Key parameters: The confidence threshold of image segmentation is set to 0.9, that is, only areas exceeding 0.9 are judged as effective areas for returning farmland to wetlands.

[0161] Ecological information collection:

[0162] Using infrared cameras, sound monitoring and other equipment, we continuously collect information on the activities of birds, fish and other species in the areas of returning farmland to wetlands.

[0163] Data structuring:

[0164] Taking the sub-projects of returning farmland to wetlands and patches of returning farmland to wetlands as objects, their associated features are extracted and structured data is constructed, such as Figure 3 shown.

[0165] Organize species monitoring information according to time series and construct time-series structured data.

[0166] Ecological flow analysis:

[0167] Based on the temporal and spatial distribution patterns of species migration and diffusion, the ecological flow data between the returning farmland to wetland areas and other areas are constructed, such as Figure 4 shown.

[0168] Dynamic monitoring:

[0169] The above structured data and ecological flow data were input into the engineering dynamic monitoring model for analysis.

[0170] Output:

[0171] Identify problems in the areas of returning farmland to wetlands, such as poor vegetation recovery and insufficient water resources;

[0172] Optimization suggestions were put forward, such as building ecological corridors between the farmland conversion and wetland restoration areas and introducing more water sources.

[0173] Technical effect verification:

[0174] After three years of continuous monitoring and dynamic optimization, the wetland restoration project has achieved remarkable results.

[0175] The accuracy rate of identifying problems in areas where farmland has been returned to wetlands reached over 95%, and key issues hindering wetland restoration were discovered and resolved in a timely manner.

[0176] According to the optimization suggestions, five ecological corridors were built and 10 new water sources were introduced, effectively improving the quality of the wetland habitat.

[0177] The wetland vegetation coverage increased from 30% to 85%, the number of bird species increased from 150 to 230, and the amount of fish resources increased by 2 times.

[0178] The ecosystem in the project area has been fully restored, biodiversity has been significantly improved, and the beauty of the landscape has been greatly enhanced.

[0179] Compared with other wetland restoration projects that do not adopt the method of the present invention, the ecological restoration speed of this project is increased by 50% and the cost is reduced by 30%.

[0180] Expert evaluation opinion: The dynamic supervision method adopted in this project is scientific and effective, has good guiding significance for wetland ecological restoration, and is worthy of promotion and application in more projects.

[0181] The above describes an embodiment of the present invention, but this embodiment is not limited to the above-mentioned specific implementation mode. The above-mentioned specific implementation mode is merely illustrative and not restrictive. Under the guidance of this embodiment, ordinary technicians in this field can also make more forms of equivalent embodiments, all of which are within the protection of this embodiment.

Claims

1. A project supervision method for natural resource and ecological protection and restoration projects, characterized in that: include: Identify engineering elements from construction images and drawings of natural resource and ecological protection and restoration projects; Collect ecological species information during the construction process; Constructing first structured data based on the engineering elements in the construction drawings, the first structured data including a structure dimension, and the basic structure elements included in the structure dimension are objects; The second structured data is constructed based on the engineering elements in the construction image, wherein the second structured data includes two structure dimensions, wherein the basic structure elements included in the first structure dimension are objects, and the basic structure elements included in the second structure dimension are time series units; constructing ecological flow data based on the second structured data, the ecological flow data including ecological flow units, and one ecological flow unit includes ecological flow information between two specific objects in the second structured data; The first structured data, the second structured data, and the ecological flow data are input into the engineering dynamic monitoring model, and the engineering dynamic monitoring model includes a first branch, a second branch, a first fusion branch, and a second fusion branch, wherein the first branch includes a first encoder, the first encoder inputs the first structured data, and outputs the first latent vectors of all objects of the first structured data; the second branch includes a second encoder and a third encoder, the second encoder sequentially inputs the associated specific features of the objects of the second structured data, and sequentially outputs the second latent vectors of the objects of the second structured data to the third encoder, the third encoder inputs the ecological flow data, and the third encoder outputs the third latent vectors of the objects of the second structured data; The first fusion branch is used to concatenate the first latent vector and the third latent vector and then decode them to obtain the problems existing in the engineering elements in the sub-project, the overall project and the construction image; the second fusion branch is used to concatenate the first latent vector and the third latent vector and then decode them to obtain the result of the corridor that needs to be added between the engineering elements in the construction image.

2. A project supervision method for natural resource and ecological protection and restoration projects according to claim 1, characterized in that: The specific objects in the second structured data refer to engineering elements recognized in the construction image.

3. A project supervision method for natural resource and ecological protection and restoration projects according to claim 1, characterized in that: The objects of the first structured data include sub-projects, general projects, and engineering elements identified in construction drawings.

4. A project supervision method for natural resource and ecological protection and restoration projects according to claim 1, characterized in that: The objects of the second structured data include: sub-projects, overall projects, and engineering elements identified in the construction images.

5. A project supervision method for natural resource and ecological protection and restoration projects according to claim 1, characterized in that: The method of constructing the first structured data includes: extracting a first structured data object; Establishing a mapping between the extracted objects and their associated specific features; For any two objects, if there is a correlation between the associated specific features related to ecological protection and restoration or there is a correlation between the objects related to ecological protection and restoration, an object relationship is established between the two objects.

6. A project supervision method for natural resource and ecological protection and restoration projects according to claim 5, characterized in that: The first structured data object has specific features associated with it as follows: Specific characteristics associated with the engineering elements identified in the construction drawings include: functional descriptions, areas, and construction instructions of the engineering elements identified in the construction drawings; Specific features associated with subprojects include: overview and construction drawings of subprojects of natural resource and ecological protection and restoration projects; Specific characteristics associated with the overall project include: the overall overview and objectives of the natural resources and ecological protection and restoration project.

7. A project supervision method for natural resource and ecological protection and restoration projects according to claim 1, characterized in that: The method of constructing the second structured data includes: extracting an object of second structured data; Extract specific features associated with the object, and divide the extracted specific features into multiple parts according to the collection period, and map one part to a time sequence unit. The order of the time sequence units is consistent with the time order of the collection period of the specific features; Establishing a mapping between the extracted objects and their associated specific features; For any two objects, if there is a correlation between the associated specific features related to ecological protection and restoration or there is a correlation between the objects related to ecological protection and restoration, an object relationship is established between the two objects.

8. A project supervision method for natural resource and ecological protection and restoration projects according to claim 7, characterized in that: The specific features associated with the objects of the second structured data are as follows: The specific features associated with the engineering elements identified in the construction image include: information on the species and area of ​​the engineering elements identified in the construction image; Specific characteristics associated with subprojects include: the construction schedule of the subproject; Specific characteristics associated with the master project include: the construction progress of the master project.

9. A project supervision method for natural resource and ecological protection and restoration projects according to claim 5 or 7, characterized in that: The relevance to ecological protection and restoration includes: Subprojects belong to the overall project; There are dependencies between subprojects; Engineering elements identified in construction drawings or engineering elements identified in construction images belong to sub-items; The engineering elements identified in the construction drawings are adjacent on the construction drawings; The engineering elements identified in the construction image are adjacent to each other on the construction image; The functions of the engineering elements identified in the construction drawings are the same; The information of species contained in the engineering elements identified in the construction images is similar.

10. A project supervision system for natural resources and ecological protection and restoration projects, characterized in that: include: An element recognition module is used to identify engineering elements from construction images and construction drawings of natural resource and ecological protection and restoration projects; Ecological information collection module, which is used to collect ecological species information during the construction process; A first data structuring module, which constructs first structured data based on engineering elements in the construction drawings, wherein the first structured data includes a structure dimension, and the basic structure elements included in the structure dimension are objects; A second data structuring module, which constructs second structured data based on the engineering elements in the construction image, wherein the second structured data includes two structure dimensions, wherein the basic structure elements included in the first structure dimension are objects, and the basic structure elements included in the second structure dimension are time sequence units; A flow data module, which is used to construct ecological flow data based on the second structured data, the ecological flow data including ecological flow units, and an ecological flow unit contains ecological flow information between two specific objects in the second structured data; A dynamic monitoring module, which is used to input the first structured data, the second structured data, and the ecological flow data into the engineering dynamic monitoring model, the engineering dynamic monitoring model includes a first branch, a second branch, a first fusion branch, and a second fusion branch, wherein the first branch includes a first encoder, the first encoder inputs the first structured data, and outputs the first latent vectors of all objects of the first structured data; the second branch includes a second encoder and a third encoder, the second encoder sequentially inputs the associated specific features of the objects of the second structured data, and sequentially outputs the second latent vectors of the objects of the second structured data to the third encoder, the third encoder inputs the ecological flow data, and the third encoder outputs the third latent vectors of the objects of the second structured data; The first fusion branch is used to concatenate the first latent vector and the third latent vector and then decode them to obtain the problems existing in the engineering elements in the sub-project, the overall project and the construction image; the second fusion branch is used to concatenate the first latent vector and the third latent vector and then decode them to obtain the result of the corridor that needs to be added between the engineering elements in the construction image.

Images