Ecological system service capability evaluation model training and generating method

Abstract

The invention provides an ecological system-based service capability evaluation model training and generating method, and the method comprises the steps: dividing a pre-obtained training region into a plurality of sub-regions, and generating a corresponding multi-level associated region for any sub-region; and training the to-be-trained evaluation model according to the obtained related evaluation data of the plurality of sub-regions and the service capability parameters corresponding to the sub-regions, so that the error between the output of the training model and the real service capability parameters meets a preset condition. When the evaluation model is used for prediction, a corresponding region is divided into a plurality of sub-regions and a plurality of levels of associated regions, so that the difficulty of acquiring related evaluation data can be reduced to a certain extent, and a plurality of levels of service capability parameters can be output for a relatively large-scale research region; and the accuracy of an ecological system service capability evaluation result aiming at a relatively large-scale research area is improved.

Description

Technical Field

[0001] The present disclosure relates to the field of data processing technology, and in particular to a method for training and generating an ecosystem service capacity assessment model. Background Art

[0002] Quantitatively evaluating the contribution of ecosystems to human economic and social development is particularly important for correctly understanding and evaluating the relationship between humans and nature. At present, the urban ecosystem service capacity of urban agglomerations is mainly calculated and evaluated through the ecosystem service value evaluation system, which requires a large amount of spatial, meteorological, soil, topographic and other geographic data.

[0003] However, it is difficult to obtain data for larger study areas, and the quality of the data obtained is poor. Therefore, the results of ecosystem service capacity assessment in larger study areas are poor. Summary of the invention

[0004] The present invention proposes a method for training and generating an ecosystem service capacity assessment model.

[0005] The first embodiment of the present disclosure proposes an ecosystem service capacity assessment model training method, the method comprising:

[0006] Divide the pre-acquired training area into multiple sub-areas, and for any sub-area, generate multiple levels of associated areas of the sub-area;

[0007] Acquire relevant evaluation data of the multiple sub-areas and service capability parameters of multiple hierarchical areas respectively corresponding to the multiple sub-areas, wherein the relevant evaluation data includes image feature data, digital elevation data, population characteristics and point of interest data;

[0008] The relevant evaluation data is used as the input of the evaluation model to be trained, and the training service capability parameters corresponding to the multiple levels of regions are output. If the error between the training service capability parameters and the service capability parameters meets the preset conditions, the trained evaluation model is output.

[0009] In the embodiment of the present disclosure, for any sub-region, generating multiple levels of associated regions of the sub-region includes:

[0010] For any sub-region, taking the sub-region as a reference, generating a first neighborhood region of the sub-region; and for any first neighborhood region, taking the first neighborhood region as a reference, generating a second neighborhood region of the first neighborhood region;

[0011] For any sub-region, the sub-region is divided into at least one first microscopic sub-region; and for any first microscopic sub-region, the first microscopic region is divided into at least one second microscopic sub-region.

[0012] In the embodiment of the present disclosure, the relevant evaluation data includes image feature data, digital elevation data, population characteristics and point of interest data, and the obtaining of the relevant evaluation data of the multiple sub-areas includes:

[0013] Acquire remote sensing image data in the training area, and calculate image feature data of any sub-area according to the remote sensing image data, wherein the image feature data includes an average value of at least one band, an average value of a vegetation index, and an average value of a water index; wherein the at least one band includes at least one of red, blue, green, and near-infrared bands;

[0014] Acquire digital elevation data in the training area, and calculate the terrain features of any sub-area according to the digital elevation data, wherein the terrain features include an average elevation and an average slope;

[0015] Acquire mobile phone signaling data in the training area, and calculate population characteristic data of any sub-area according to the mobile phone signaling data, wherein the population characteristic data includes population quantity and population concentration;

[0016] Obtaining point of interest data in the training area, the point of interest data including multiple types of geographical locations; calculating geographical location data of any sub-area, the geographical location data including the number of points of interest and the concentration of points of interest;

[0017] The image feature data, terrain features, population feature data and geographical location data in the sub-area are combined to obtain relevant evaluation data of the sub-area.

[0018] In an embodiment of the present disclosure, the number of nodes in the first layer of the hidden layer of the evaluation model to be trained is half the number of nodes in the input layer, the number of nodes in the second layer is one quarter the number of nodes in the input layer, and the number of nodes in the output layer is the number of the levels.

[0019] The second aspect of the present disclosure provides a method for generating service capability parameters based on an ecosystem service capability assessment model, the method comprising:

[0020] Dividing the pre-acquired target area into a plurality of sub-areas to be evaluated, and relevant evaluation data corresponding to the plurality of sub-areas to be evaluated;

[0021] The relevant assessment data is input into the assessment model obtained by the ecosystem service capacity assessment model training method described in the first aspect or any optional implementation scheme of the first aspect to obtain the predicted service capacity parameters corresponding to the target area at multiple levels.

[0022] In an embodiment of the present disclosure, the predicted service capability parameters of the multiple levels include predicted service capability parameters of each sub-area, a first neighborhood service capability parameter, a second neighborhood service capability parameter, a first micro service capability parameter, and a second micro service capability parameter. The method further includes:

[0023] Determining a first modified service capability parameter for each sub-area according to the first neighborhood service capability parameter and the second neighborhood service capability parameter;

[0024] Determining a second modified service capability parameter for each sub-area according to the first micro service capability parameter and the second micro service capability parameter;

[0025] Based on the relationship between the first modified service capability parameter, the second modified service capability parameter and the predicted service capability parameter corresponding to each sub-area, obtaining the first target service capability parameter of the target area in each sub-area;

[0026] The second target service capability parameter is determined according to the relationship between the first target service capability parameter of the sub-area and each first neighborhood service capability parameter corresponding to the sub-area.

[0027] In the embodiment of the present disclosure, determining the first modified service capability parameter of each sub-area according to the first neighborhood service capability parameter and the second neighborhood service capability parameter includes:

[0028] Using the second neighborhood service capability parameter as a modified neighborhood capability parameter of each first neighborhood area corresponding to the second neighborhood service parameter;

[0029] Comparing the modified neighborhood capability parameters of each of the first neighborhood areas with the first neighborhood capability parameters of each of the first neighborhood areas to obtain a first neighborhood comparison result;

[0030] If the first neighborhood comparison result is within the neighborhood confidence interval, the corrected neighborhood capability parameters of each first neighborhood area are used as first corrected service capability parameters corresponding to each sub-area.

[0031] In the embodiment of the present disclosure, determining the second modified service capability parameter of each sub-area according to the first micro service capability parameter and the second micro service capability parameter includes:

[0032] Merging the second micro service capability parameters belonging to the same first micro sub-region to obtain a first modified micro service capability parameter of the first micro sub-region;

[0033] Comparing the first modified micro service capability parameter of the first micro sub-region with the first micro service capability parameter of the first micro sub-region to obtain a first micro comparison result;

[0034] If the first micro comparison result is within the micro confidence interval, the first corrected micro service capability parameters belonging to the same sub-region are merged to obtain the second corrected service capability parameters of the sub-region.

[0035] In the embodiment of the present disclosure, for any sub-region, obtaining the first target service capability parameter of the target region in each sub-region based on the relationship between the first modified service capability parameter, the second modified service capability parameter, and the predicted service capability parameter corresponding to each sub-region includes:

[0036] If a first difference between the first modified service capability parameter and the predicted service capability parameter is within a preset neighborhood trust region, using the first modified service capability parameter as a first target service capability parameter for the sub-region;

[0037] If a second difference between the second modified service capability parameter and the predicted service capability parameter is within a preset micro confidence region, using the second modified service capability parameter as the first target service capability parameter of the sub-region;

[0038] If the first difference between the first corrected service capability parameter and the predicted service capability parameter is not within the preset neighborhood trust region, and the second difference between the second corrected service capability parameter and the predicted service capability parameter is not within the preset micro trust region, the predicted service capability parameter is used as the first target service capability parameter.

[0039] In the embodiment of the present disclosure, determining the second target service capability parameter according to the relationship between the first target service capability parameter of the sub-area and each first neighborhood service capability parameter corresponding to the sub-area further includes:

[0040] Calculating a mean value of a first neighborhood service parameter corresponding to the sub-area;

[0041] If the difference between the first target service capability parameter of the sub-region and the mean value is greater than a preset threshold, taking the mean value as the second target service capability parameter of the sub-region;

[0042] If the difference between the first target service capability parameter of the sub-area and the mean value is less than or equal to a preset threshold, the first target service capability parameter is used as the second target service capability parameter.

[0043] The technical solution provided in the embodiments of the present disclosure has at least the following technical effects or advantages:

[0044] The training and generation method of the ecosystem service capacity assessment model in the disclosed embodiment includes dividing the pre-acquired training area into multiple sub-areas, generating multiple levels of corresponding associated areas for any sub-area; training the assessment model to be trained based on the relevant assessment data of the multiple sub-areas acquired, and the service capacity parameters corresponding to each sub-area, so that the error between the output of the training model and the actual service capacity parameters meets the preset conditions. When using the assessment model for prediction, dividing the corresponding area into multiple sub-areas and multiple levels of associated areas can reduce the difficulty of obtaining relevant assessment data to a certain extent, and can output multiple levels of service capacity parameters for a larger research area, thereby improving the accuracy of the results of the ecosystem service capacity assessment for the larger research area.

[0045] Additional aspects and advantages of the present disclosure will be given in part in the following description and in part will be obvious from the following description or learned through practice of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] By reading the detailed description of the preferred embodiment below, various other advantages and benefits will become clear to those of ordinary skill in the art. The accompanying drawings are only used for the purpose of illustrating the preferred embodiment and are not considered to be limitations of the present disclosure. In addition, the same reference symbols are used to represent the same components throughout the accompanying drawings.

[0047] In the attached picture:

[0048] Figure 1 A flowchart of an ecosystem service capability assessment model training method provided by an embodiment of the present disclosure is shown;

[0049] Figure 2 A flowchart of a method for generating service capability parameters based on an ecosystem service capability assessment model provided by another embodiment of the present disclosure is shown;

[0050] Figure 3 A schematic diagram of the structure of an electronic device provided by an embodiment of the present disclosure is shown;

[0051] Figure 4 A schematic diagram of a storage medium provided by an embodiment of the present disclosure is shown. DETAILED DESCRIPTION

[0052] The exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided in order to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0053] It should be noted that, unless otherwise specified, the technical terms or scientific terms used in the present disclosure should have the common meanings understood by those skilled in the art to which the present disclosure belongs.

[0054] Ecosystem service capacity assessment refers to the quantitative or qualitative research and evaluation of the various beneficial functions that ecosystems provide to humans. These functions include but are not limited to food supply, water resource regulation, climate regulation, disease control, soil fertility maintenance, flood prevention and disaster reduction, biodiversity protection, and spiritual, cultural and educational benefits to humans.

[0055] Ecosystem service capacity assessment usually includes the following aspects: Supply services: assess the quantity and quality of food, water, energy, medicine and other material products provided by ecosystems to humans. Regulating services: assess the role and value of ecosystems in climate regulation, water purification, soil fertility maintenance, flood prevention and disaster reduction, etc. Support services: assess the contribution of ecosystems in maintaining biodiversity, providing genetic resources, maintaining ecological processes, etc. Cultural services: assess the spiritual, cultural and educational benefits that humans obtain from the natural environment, such as aesthetic appreciation, leisure and entertainment, etc.

[0056] Ecosystem service capacity assessment is particularly important for correctly understanding and evaluating the relationship between humans and nature, and is also an important means of coordinating economic development and ecological protection. The assessment results can provide a strong basis for decision-making for policymakers and help formulate more scientific and reasonable ecological protection policies and resource management strategies. In addition, the assessment results can also enhance the public's understanding of the value of ecosystems and promote awareness of ecological protection and sustainable development.

[0057] However, when evaluating ecosystem service capabilities, there are problems such as incomplete data collection, poor data quality, and inconsistent evaluation methods. In order to solve these problems, the present disclosure proposes an ecosystem service capability evaluation model training method to solve the above problems. Figure 1 As shown, the ecosystem service capability assessment model training method of the embodiment of the present disclosure, specifically, the method may include the following steps:

[0058] In step S101 , the pre-acquired training region is divided into a plurality of sub-regions, and for any sub-region, a plurality of levels of associated regions of the sub-region are generated.

[0059] For example, for a larger-scale training area, the data in the training area has many types and a large amount of data. Dividing the training area into multiple sub-areas can partition and identify the large-scale training area and collect data sets.

[0060] The criterion for dividing the sub-regions can be to divide the training area into grids of fixed size according to the geographical location of the training area and the accuracy of the evaluation of the training area. For example, if the training area is a two-dimensional space, it can be divided into several square or rectangular sub-regions. It can also be divided according to features (such as terrain, landform, population density, etc.) so that each sub-region has a certain uniformity or representativeness in terms of features.

[0061] After the sub-regions are divided, it is necessary to determine the associated regions of each sub-region at each level. Specifically, each sub-region can have an associated region of a fixed size, which can be in the shape of concentric circles, squares, rectangles, etc. The size of the associated region can be adjusted according to some characteristics of the sub-region (such as population density, terrain complexity, etc.). Generate multiple levels of associated regions for each sub-region, for example, the first layer is the adjacent sub-region, the second layer is the next adjacent sub-region, and so on. Generate the associated region of each sub-region according to the determined level and range. For each sub-region, calculate the coordinate range of its associated region according to its coordinates and size. Spatial indexes (such as R-trees, quadtrees, etc.) can also be used to efficiently store and query associated regions. In addition, for each sub-region, calculate its distance from other sub-regions, and determine which sub-regions belong to its associated region based on the distance.

[0062] In an optional embodiment, the generation of multiple levels of associated regions of sub-regions can also be achieved in the following manner. For any sub-region, taking the sub-region as a reference, a first neighborhood region of the sub-region is generated; and for any first neighborhood region, taking the first neighborhood region as a reference, a second neighborhood region of the first neighborhood region is generated; for any sub-region, the sub-region is divided into at least one first microscopic sub-region; and for any first microscopic sub-region, the first microscopic region is divided into at least one second microscopic sub-region.

[0063] Exemplarily, for any sub-region, the first neighborhood region of the sub-region is generated by using 8 neighborhoods, and the first neighborhood region is an 8-neighborhood macro grid; for the first neighborhood region, the second neighborhood region of the first neighborhood region is generated by using 8 neighborhoods again, and the second neighborhood region is also an 8-neighborhood macro grid. Thus, two levels of neighborhood regions of the sub-region are generated;

[0064] On the other hand, for any sub-region, the sub-region can be divided into multiple micro-regions, specifically, the sub-region can be divided into 3×3 micro-regions to obtain a first microscopic sub-region; the first microscopic sub-region can be further divided into 3×3 micro-regions to obtain a second microscopic sub-region. Thus, two levels of microscopic regions of the sub-region are generated.

[0065] In summary, for the training area, five levels of regions have been generated, including sub-regions, two levels of neighborhood regions corresponding to the sub-regions, and two levels of micro-regions.

[0066] In step S102, relevant evaluation data of the plurality of sub-regions and service capability parameters of the plurality of hierarchical regions corresponding to the plurality of sub-regions are obtained.

[0067] Exemplarily, after the training area is divided into sub-areas, relevant evaluation data related to the service capability parameters in each sub-area is obtained. Specifically, it can be spatial distribution data, meteorological data, soil data, terrain data, vegetation coverage data, species distribution data and other data. When obtaining relevant evaluation data, the disclosed embodiment can select a sub-area that is convenient for obtaining data from all sub-areas according to the difficulty of obtaining the relevant evaluation data, so that when the evaluation model is subsequently trained, only the sub-area where the data is obtained can be trained. This can be done in the training model stage. If the target area is evaluated later, all the data of the target area needs to be obtained.

[0068] In an optional embodiment, the relevant evaluation data of each sub-area can be extracted through remote sensing images, digital elevation data, mobile phone signaling data and point of interest data, and can be specifically obtained in the following manner: obtain remote sensing image data in the training area, calculate image feature data of any sub-area based on the remote sensing image data, the image feature data including the average value of at least one band, the average value of vegetation index and the average value of water index; wherein at least one band includes at least one of red, blue, green and near-infrared bands; obtain digital elevation data in the training area, calculate terrain features of any sub-area based on the digital elevation data, the terrain features including the average elevation and the average slope; obtain mobile phone signaling data in the training area, calculate population feature data of any sub-area based on the mobile phone signaling data, the population feature data including population size and population concentration; obtain point of interest data in the training area, the point of interest data including multiple types of geographical locations; calculate geographical location data of any sub-area, the geographical location data including the number of points of interest and the concentration of points of interest; combine the image feature data, terrain features, population feature data and geographical location data in the sub-area to obtain relevant evaluation data of the sub-area.

[0069] Exemplarily, remote sensing image data in the training area is obtained, and the remote sensing image data is image data converted from electromagnetic wave information received by the remote sensing sensor. These electromagnetic wave information can come from electromagnetic waves reflected or emitted by the surface, such as visible light, infrared, microwave and other bands. Remote sensing image data records the spectral characteristics of the surface, which are received by the sensor after propagating through the atmosphere. Different types of surface coverage are manifested as different ground features, and ultimately reflected as remote sensing data with different chromaticity values ​​and brightness values. Therefore, different surface coverage can be determined by distinguishing the values ​​of each band in the remote sensing image data. Calculate the average value of each band included in the remote sensing image data corresponding to each sub-area, which can be the average value corresponding to the red, blue, green and near-infrared bands. The remote sensing data features are as follows: RSFetureGrid = (R, G, B, NIR, NDVI, NDWI).

[0070] You can also obtain digital elevation data in the training area, which can represent and store terrain elevation information. Calculate the terrain feature data of each sub-area, including the average value of elevation and slope. DEMFeatureGrid = (H, S) where H is the average elevation of the sub-area and S is the average slope of the sub-area.

[0071] The relevant evaluation data in the training area can also obtain the corresponding population characteristics. Specifically, the population characteristics can be extracted by obtaining mobile phone signaling data. The population characteristic data of each sub-area is extracted according to the mobile phone signaling data, mainly the population number and population concentration. RKFeatureGrid = (N, D) where N is the population number in the grid, and D is the population concentration in the grid. The training area is split into sub-areas and the population density is calculated. The maximum population density is used to characterize the population concentration.

[0072] Point of interest data within the training area can also be determined, where the point of interest data is a specific location or geographic entity that the user may be interested in, such as a restaurant, store, museum, park, bus stop, hospital, etc. The corresponding point of interest data may include the name of the point of interest, such as "park", "coffee shop". The type or classification of the point of interest, such as catering, entertainment, education, medical care, etc. The geographic coordinates of the point of interest, usually a longitude and latitude pair. A detailed description of the point of interest, which may include the address, opening hours, service content, etc. User evaluation and comments on the point of interest. The number of visitors or traffic at the point of interest, etc.

[0073] After calculating the relevant evaluation data in each sub-area, in order to train the model based on these data, these data need to be processed so that the data conforms to the input of the model. Specifically, each data can be normalized and then feature combined to obtain a complete data set so that the data set can be input into the model for training.

[0074] For example, the four data features of image feature data, terrain feature data, population feature data, and geographic location data are combined to obtain the feature data set of the evaluation object. That is, for each sub-area, a feature vector is constructed to express the characteristics of the grid.

[0075] FeatureGrid={RSFeatureGrid, DEMFeatureGrid, RKFeatureGrid, POIFeatureGrid}

[0076] In step S103, the relevant evaluation data is used as the input of the evaluation model to be trained, and the training service capability parameters corresponding to the regions at multiple levels are output. If the error between the training service capability parameters and the service capability parameters meets the preset conditions, the trained evaluation model is output.

[0077] Exemplarily, in the disclosed embodiment, the input of the evaluation model needs to match the features in the data set. Specifically, taking 4-band remote sensing images and 10 types of interest point data as an example, the input layer includes 32*5 nodes. In order to minimize the impact of data magnitude on model training, ensure good convergence, mapping ability and calculation accuracy of the network, and prevent data from being highly concentrated on certain neurons and weights during the calculation process, the input values ​​of the training data are normalized to ensure that the value of each variable is in the range of 0 to 1.

[0078] Ds=(D-Dmin) / (Dmax-Dmin)

[0079] Where D is the original data, Ds is the normalized input parameter, and Dmax and Dmin are the maximum and minimum values ​​of the corresponding feature data within all grid ranges at the same scale.

[0080] The number of hidden layers and nodes is crucial to the performance of the neural network. Too few layers will lead to underfitting problems, and too many layers will lead to overfitting and the model will not converge. Considering the actual complexity of the evaluation of the ecological service system, the present invention uses two hidden layers to construct a neural network.

[0081] When determining the number of hidden layer nodes, the number of nodes in the first layer of the hidden layer is half of the number of nodes in the input layer, and the number of nodes in the second layer is one quarter of the number of nodes in the input layer. Taking 4-band remote sensing images and 10 types of interest point data as an example, the input layer includes 32*5 nodes, then the first hidden layer is 16*5 nodes, and the second hidden layer is 8*5 nodes.

[0082] The number of output layer nodes is determined according to the level corresponding to the training area. For the embodiment of the present disclosure, there are 5 nodes, which correspond to the service capacity parameters of the ecosystems of different levels of areas.

[0083] During the model training process, the error between the training service capability parameter and the service capability parameter is determined according to a preset loss function, wherein the loss function may be a mean square error (MSE) or a mean absolute error (MAE).

[0084] Choose an appropriate optimizer, such as Adam, SGD, etc., to adjust the model parameters to minimize the loss function.

[0085] Input the preprocessed data into the model. Calculate the training service capacity parameters of the output layer. Calculate the error using the loss function. Update the model parameters through the back propagation algorithm. Iterate the above process until the preset number of iterations is reached or the error converges.

[0086] Use a dataset that was not used in the training (validation set) to evaluate the model performance. Calculate the error between the training service capacity parameters on the validation set and the actual service capacity parameters. Set an error threshold or condition, such as MSE is less than a certain value or MAE is less than a certain value. If the error on the validation set meets the preset conditions, the model training is complete. Save the trained model for subsequent use.

[0087] The ecosystem service capacity assessment model training method provided by the embodiment of the present disclosure divides the pre-acquired training area into multiple sub-areas, and generates corresponding multiple levels of associated areas for any sub-area; the assessment model to be trained is trained according to the relevant assessment data of the multiple sub-areas obtained, and the service capacity parameters corresponding to each sub-area, so that the error between the output of the training model and the actual service capacity parameters meets the preset conditions. When using the assessment model for prediction, the corresponding area is divided into multiple sub-areas and multiple levels of associated areas, which can reduce the difficulty of obtaining relevant assessment data to a certain extent, and can output multiple levels of service capacity parameters for a larger research area, thereby improving the accuracy of the results of the ecosystem service capacity assessment for the larger research area.

[0088] Corresponding to the above-mentioned ecosystem service capacity assessment model training method, the present disclosure also includes a method for generating service capacity parameters by applying the above-mentioned trained ecosystem service capacity assessment model, specifically as follows Figure 2 As shown, the method for generating service capacity parameters based on the ecosystem service capacity assessment model includes the following steps:

[0089] In step S201, the pre-acquired target area is divided into a plurality of sub-areas to be evaluated, and relevant evaluation data corresponding to the plurality of sub-areas to be evaluated.

[0090] In step S202, the relevant assessment data is input into the assessment model obtained by the ecosystem service capacity assessment model training method in the above embodiment to obtain the predicted service capacity parameters corresponding to the target area at multiple levels.

[0091] For example, when using the trained evaluation model to predict the service capacity parameters of the target area, in order to meet the requirements of the input of the evaluation model, the target area is divided into multiple sub-areas to be evaluated. The division method is shown in the above embodiment and will not be repeated here. After the multiple sub-areas to be evaluated are divided, the relevant evaluation data of each sub-area to be evaluated is obtained respectively. The acquisition of the relevant evaluation data is also shown in the acquisition method in the above embodiment and will not be repeated here.

[0092] The relevant evaluation data of each sub-area to be evaluated are input into the evaluation model to obtain the corresponding predicted service capacity parameters of the target area at multiple levels.

[0093] After the predicted service capability parameters of multiple levels are obtained according to the evaluation model, the service capability parameters may be corrected to further ensure their accuracy, including corrections between levels and corrections between spaces.

[0094] Specifically, the correction between each level can be achieved in the following ways: determine the first corrected service capacity parameter of each sub-area based on the first neighborhood service capacity parameter and the second neighborhood service capacity parameter; determine the second corrected service capacity parameter of each sub-area based on the first micro service capacity parameter and the second micro service capacity parameter; obtain the first target service capacity parameter of the target area in each sub-area based on the relationship between the first corrected service capacity parameter, the second corrected service capacity parameter and the predicted service capacity parameters corresponding to each sub-area; determine the second target service capacity parameter based on the relationship between the first target service capacity parameter of the sub-area and the first neighborhood service capacity parameters corresponding to the sub-area.

[0095] Exemplarily, the first neighborhood service capability parameters and the second neighborhood service capability parameters corresponding to each sub-area to be evaluated are used to determine the first modified service capability parameters corresponding to each sub-area to be evaluated included in each neighborhood, wherein the first modified capability parameters are to assign the service capability parameters of each neighborhood to the corresponding sub-area to be evaluated.

[0096] Similarly, the second modified service capacity parameter is obtained by summarizing the service capacity parameters of each micro-region to obtain the corresponding sub-region to be evaluated. The specific summarization method can be the average method.

[0097] Finally, the first target service capacity parameter corrected at the hierarchical level for the sub-area to be evaluated is determined by the relationship between the first corrected service capacity parameter, the second corrected service capacity parameter and the predicted service capacity parameter. Specifically, if the first difference between the first corrected service capacity parameter and the predicted service capacity parameter is within the preset neighborhood trust region, the first corrected service capacity parameter is used as the first target service capacity parameter for the sub-area to be evaluated; if the second difference between the second corrected service capacity parameter and the predicted service capacity parameter is within the preset micro-trust region, the second corrected service capacity parameter is used as the first target service capacity parameter for the sub-area to be evaluated; if the first difference between the first corrected service capacity parameter and the predicted service capacity parameter is not within the preset neighborhood trust region, and the second difference between the second corrected service capacity parameter and the predicted service capacity parameter is not within the preset micro-trust region, the predicted service capacity parameter is used as the first target service capacity parameter.

[0098] In the above embodiment, the specific assignment method of the first corrected service capability parameter may be: merging the second micro service capability parameters belonging to the same first micro sub-area to be evaluated to obtain the first corrected micro service capability parameter of the first micro sub-area to be evaluated; comparing the first corrected micro service capability parameter of the first micro sub-area to be evaluated with the first micro service capability parameter of the first micro sub-area to be evaluated to obtain the first micro comparison result; if the first micro comparison result is within the micro confidence interval, merging the first corrected micro service capability parameters belonging to the same sub-area to be evaluated to obtain the second corrected service capability parameter of the sub-area to be evaluated.

[0099] The specific assignment method of the second corrected service capability parameter can be: if the first difference between the first corrected service capability parameter and the predicted service capability parameter is within the preset neighborhood trust region, the first corrected service capability parameter is used as the first target service capability parameter of the sub-area to be evaluated; if the second difference between the second corrected service capability parameter and the predicted service capability parameter is within the preset micro-trust region, the second corrected service capability parameter is used as the first target service capability parameter of the sub-area to be evaluated; if the first difference between the first corrected service capability parameter and the predicted service capability parameter is not within the preset neighborhood trust region, and the second difference between the second corrected service capability parameter and the predicted service capability parameter is not within the preset micro-trust region, the predicted service capability parameter is used as the first target service capability parameter.

[0100] In addition, the correction between spaces can be achieved in the following way: calculate the mean of the first neighborhood service parameter corresponding to the sub-area; if the difference between the first target service capacity parameter of the sub-area and the mean is greater than a preset threshold, the mean is used as the second target service capacity parameter of the sub-area; if the difference between the first target service capacity parameter of the sub-area and the mean is less than or equal to the preset threshold, the first target service capacity parameter is used as the second target service capacity parameter.

[0101] For example, considering the continuity of ecosystem service value in space, correction can be made based on spatial correlation. The difference between the mean of the service capacity parameter of the sub-region to be evaluated and the sub-region to be evaluated in the corresponding neighborhood is calculated. If the difference is greater than a preset threshold, it is considered that the service capacity parameter of the sub-region to be evaluated is abnormal.

[0102] The present disclosure also provides an electronic device for executing the above data transmission method. Figure 3 , which shows a schematic diagram of an electronic device provided by some embodiments of the present disclosure. Figure 3 As shown, the electronic device includes: a processor 500, a memory 501, a bus 502 and a communication interface 503, the processor 500, the communication interface 503 and the memory 501 are connected through the bus 502; the memory 501 stores a computer program that can be run on the processor 500, and the processor 500 executes the aforementioned Figure 1 or Figure 2 The method provided.

[0103] The memory 501 may include a high-speed random access memory (RAM), and may also include a non-volatile memory (non-volatile memory), such as at least one disk storage. The communication connection between the system network element and at least one other network element is realized through at least one communication interface 503 (which may be wired or wireless), and the Internet, wide area network, local area network, metropolitan area network, etc. may be used.

[0104] The bus 502 may be an ISA bus, a PCI bus, or an EISA bus. The bus may be divided into an address bus, a data bus, a control bus, etc. Among them, the memory 501 is used to store programs. After receiving the execution instruction, the processor 500 executes the program. Figure 1 or Figure 2 The disclosed method may be applied in the processor 500 or implemented by the processor 500 .

[0105] The processor 500 may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method can be completed by the hardware integrated logic circuit in the processor 500 or the instruction in the form of software. The above processor 500 can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The methods, steps and logic block diagrams disclosed in the embodiments of the present disclosure can be implemented or executed. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor. The steps of the method disclosed in conjunction with the embodiments of the present disclosure can be directly embodied as a hardware decoding processor to be executed, or the hardware and software modules in the decoding processor can be executed. The software module can be located in a mature storage medium in the field such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory or an electrically erasable programmable memory, a register, etc. The storage medium is located in the memory 501, and the processor 500 reads the information in the memory 501 and completes the steps of the above method in combination with its hardware.

[0106] The electronic device provided by the embodiment of the present disclosure and the data transmission method provided by the embodiment of the present disclosure are based on the same inventive concept and have the same beneficial effects as the methods adopted, operated or implemented by them.

[0107] The present disclosure also provides a computer-readable storage medium corresponding to the data transmission method provided in the above embodiment. Figure 4 The computer-readable storage medium shown is a CD 30 on which a computer program (ie, a program product) is stored. When the computer program is run by a microprocessor, it will execute the data transmission method provided by any of the aforementioned embodiments.

[0108] It should be noted that examples of computer-readable storage media may also include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other optical or magnetic storage media, which are not listed here one by one.

[0109] The computer-readable storage medium provided by the above-mentioned embodiments of the present disclosure and the data transmission method provided by the embodiments of the present disclosure are based on the same inventive concept and have the same beneficial effects as the methods adopted, run or implemented by the application programs stored therein.

[0110] It should be noted that:

[0111] In the description provided herein, a large number of specific details are described. However, it is understood that the embodiments of the present disclosure can be practiced without these specific details. In some instances, well-known structures and technologies are not shown in detail so as not to obscure the understanding of this description.

[0112] Similarly, it should be understood that in order to streamline the present disclosure and aid in understanding one or more of the various inventive aspects, in the above description of the exemplary embodiments of the present disclosure, the various features of the present disclosure are sometimes grouped together into a single embodiment, figure, or description thereof. However, the disclosed method should not be interpreted as reflecting the following schematic diagram: the claimed disclosure requires more features than the features explicitly recited in each claim. More specifically, as reflected in the claims below, the inventive aspects lie in less than all the features of the single embodiment disclosed above. Therefore, the claims that follow the specific embodiment are hereby expressly incorporated into the specific embodiment, with each claim itself serving as a separate embodiment of the present disclosure.

[0113] In addition, those skilled in the art will appreciate that, although some embodiments herein include certain features included in other embodiments but not other features, the combination of features of different embodiments is meant to be within the scope of the present disclosure and form different embodiments. For example, in the claims below, any one of the claimed embodiments may be used in any combination.

[0114] The above are only preferred specific embodiments of the present disclosure, but the protection scope of the present disclosure is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed in the present disclosure should be included in the protection scope of the present disclosure. Therefore, the protection scope of the present disclosure should be based on the protection scope of the claims.

Claims

1. A method for training an ecosystem service capacity assessment model, characterized in that: The method comprises: Divide the pre-acquired training area into multiple sub-areas, and for any sub-area, generate multiple levels of associated areas of the sub-area; Acquire relevant evaluation data of the multiple sub-areas and service capability parameters of multiple hierarchical areas respectively corresponding to the multiple sub-areas, wherein the relevant evaluation data includes image feature data, digital elevation data, population characteristics and point of interest data; The relevant evaluation data is used as the input of the evaluation model to be trained, and the training service capability parameters corresponding to the multiple levels of regions are output. If the error between the training service capability parameters and the service capability parameters meets the preset conditions, the trained evaluation model is output.

2. The method according to claim 1, characterized in that For any of the sub-regions, generating multiple levels of associated regions of the sub-region includes: For any sub-region, taking the sub-region as a reference, generating a first neighborhood region of the sub-region; and for any first neighborhood region, taking the first neighborhood region as a reference, generating a second neighborhood region of the first neighborhood region; For any sub-region, the sub-region is divided into at least one first microscopic sub-region; and for any first microscopic sub-region, the first microscopic region is divided into at least one second microscopic sub-region.

3. The method according to claim 1, characterized in that: The obtaining of relevant evaluation data of the plurality of sub-areas includes: Acquire remote sensing image data in the training area, and calculate image feature data of any sub-area according to the remote sensing image data, wherein the image feature data includes an average value of at least one band, an average value of a vegetation index, and an average value of a water index; wherein the at least one band includes at least one of red, blue, green, and near-infrared bands; Acquire digital elevation data in the training area, and calculate the terrain features of any sub-area according to the digital elevation data, wherein the terrain features include an average elevation and an average slope; Acquire mobile phone signaling data in the training area, and calculate population characteristic data of any sub-area according to the mobile phone signaling data, wherein the population characteristic data includes population quantity and population concentration; Obtaining point of interest data in the training area, the point of interest data including multiple types of geographical locations; calculating geographical location data of any sub-area, the geographical location data including the number of points of interest and the concentration of points of interest; The image feature data, terrain features, population feature data and geographical location data in the sub-area are combined to obtain relevant evaluation data of the sub-area.

4. The method according to claim 1, characterized in that: The number of nodes in the first layer of the hidden layer of the evaluation model to be trained is half of the number of nodes in the input layer, the number of nodes in the second layer is one quarter of the number of nodes in the input layer, and the number of nodes in the output layer is the number of the levels.

5. A method for generating service capacity parameters based on an ecosystem service capacity assessment model, characterized in that: The method comprises: Dividing the pre-acquired target area into a plurality of sub-areas to be evaluated, and acquiring relevant evaluation data corresponding to the plurality of sub-areas to be evaluated; The relevant assessment data is input into the assessment model obtained by the ecosystem service capacity assessment model training method according to any one of claims 1 to 4 to obtain the predicted service capacity parameters corresponding to the target area at multiple levels.

6. The method according to claim 5, characterized in that The multiple levels of predicted service capability parameters include predicted service capability parameters of each sub-area to be evaluated, a first neighborhood service capability parameter, a second neighborhood service capability parameter, a first micro service capability parameter, and a second micro service capability parameter. The method further includes: Determining a first modified service capability parameter of each sub-area to be evaluated according to the first neighborhood service capability parameter and the second neighborhood service capability parameter; Determining a second modified service capability parameter of each sub-area to be evaluated according to the first micro service capability parameter and the second micro service capability parameter; Based on the relationship between the first modified service capability parameter, the second modified service capability parameter and the predicted service capability parameter corresponding to each sub-area to be evaluated, obtaining a first target service capability parameter of the target area in each sub-area to be evaluated; The second target service capability parameter is determined according to the relationship between the first target service capability parameter of the sub-region to be evaluated and each first neighborhood service capability parameter corresponding to the sub-region to be evaluated.

7. The method according to claim 6, characterized in that The determining, according to the first neighborhood service capability parameter and the second neighborhood service capability parameter, a first modified service capability parameter of each sub-area to be evaluated includes: Using the second neighborhood service capability parameter as a modified neighborhood capability parameter of each first neighborhood area corresponding to the second neighborhood service parameter; Comparing the modified neighborhood capability parameters of each of the first neighborhood areas with the first neighborhood capability parameters of each of the first neighborhood areas to obtain a first neighborhood comparison result; If the first neighborhood comparison result is within the neighborhood confidence interval, the corrected neighborhood capability parameters of each first neighborhood area are used as first corrected service capability parameters corresponding to each sub-area to be evaluated.

8. The method according to claim 6, characterized in that The step of determining the second modified service capability parameter of each sub-area to be evaluated according to the first micro service capability parameter and the second micro service capability parameter includes: Merging the second micro service capability parameters belonging to the same first micro sub-region to be evaluated to obtain a first revised micro service capability parameter of the first micro sub-region to be evaluated; Comparing the first corrected micro service capability parameter of the first micro sub-region to be evaluated with the first micro service capability parameter of the first micro sub-region to be evaluated to obtain a first micro comparison result; If the first micro comparison result is within the micro confidence interval, the first corrected micro service capability parameters belonging to the same sub-area to be evaluated are merged to obtain the second corrected service capability parameters of the sub-area to be evaluated.

9. The method according to claim 6, characterized in that For any sub-region to be evaluated, obtaining the first target service capability parameter of the target region in each sub-region to be evaluated based on the relationship between the first modified service capability parameter, the second modified service capability parameter, and the predicted service capability parameter corresponding to each sub-region to be evaluated includes: If a first difference between the first corrected service capability parameter and the predicted service capability parameter is within a preset neighborhood trust region, using the first corrected service capability parameter as a first target service capability parameter of the sub-area to be evaluated; If a second difference between the second corrected service capability parameter and the predicted service capability parameter is within a preset micro confidence region, using the second corrected service capability parameter as the first target service capability parameter of the sub-region to be evaluated; If the first difference between the first corrected service capability parameter and the predicted service capability parameter is not within the preset neighborhood trust region, and the second difference between the second corrected service capability parameter and the predicted service capability parameter is not within the preset micro trust region, the predicted service capability parameter is used as the first target service capability parameter.

10. The method according to claim 6, characterized in that The determining of the second target service capability parameter according to the relationship between the first target service capability parameter of the sub-region to be evaluated and each first neighborhood service capability parameter corresponding to the sub-region to be evaluated further includes: Calculating the mean value of the first neighborhood service parameter corresponding to the sub-area to be evaluated; If the difference between the first target service capability parameter of the sub-area to be evaluated and the mean value is greater than a preset threshold, the mean value is used as the second target service capability parameter of the sub-area to be evaluated; If the difference between the first target service capability parameter of the sub-area to be evaluated and the mean value is less than or equal to a preset threshold, the first target service capability parameter is used as the second target service capability parameter.

Images