Method and device for evaluating quality of biological living environment, and electronic device

Abstract

The application relates to the technical field of habitat quality evaluation, and discloses a method and device for evaluating the quality of a biological living environment and an electronic device, wherein the method comprises the following steps: acquiring spatialized environment data of a target biological living environment; the spatialized environment data comprises grid data of the spatialization of the environment data of the target biological living environment; inputting the spatialized environment data into a living environment quality evaluation model corresponding to the target biological living environment to obtain a nest probability of the target biological nest in the target biological living environment; the living environment quality evaluation model is obtained by training a machine learning algorithm based on a gradient boosting framework; and determining a habitat quality evaluation result of the target biological living environment according to the nest probability. The nest probability of the target biological nest in the target biological living environment is obtained by using the living environment quality evaluation model, and the habitat quality is analyzed according to the nest probability, so that the accuracy of the habitat quality evaluation result is improved.

Description

Technical Field

[0001] The present application relates to the technical field of habitat quality assessment, for example, to a method and device, and electronic equipment for assessing the quality of a biological living environment. Background Art

[0002] To protect rare and endangered species and improve habitat quality, my country has established numerous nature reserves dedicated to these species. Assessing habitat quality within these reserves can help develop targeted conservation measures and provide scientific guidance for conservationists. However, current habitat quality assessments for rare and endangered species are limited by a lack of real-time data and a single assessment model, resulting in inaccurate results.

[0003] Related technology discloses a habitat quality monitoring method, including: constructing a basic geographic database in a cloud platform based on attribute information of multiple monitoring areas and multi-source land cover products; constructing a habitat quality monitoring model in the cloud platform based on evaluation rule information in a desktop evaluation model; obtaining habitat quality evaluation parameters of the target monitoring area through public information; obtaining land cover products corresponding to the target monitoring area based on the geographic database; inputting the habitat quality evaluation parameters and corresponding land cover products of the target monitoring area into the habitat quality monitoring model to obtain the habitat quality of the target monitoring area. Among them, the desktop evaluation model is, for example, the InVEST model (Integrated Valuation of Ecosystem Services and Trade-offs) or the Forest-BGC model (forest biogeochemical cycles, process-based biogeochemical cycle model).

[0004] During the implementation of the embodiments of the present disclosure, it was found that at least the following problems exist in the related art:

[0005] In related technologies, a habitat quality monitoring model is constructed on a cloud platform based on the InVEST model or the Forest-BGC model. However, if the data quality is not high, there is still a problem of low accuracy of habitat quality assessment results.

[0006] It should be noted that the information disclosed in the above background technology section is only used to enhance the understanding of the background of this application, and therefore may include information that does not constitute prior art known to ordinary technicians in this field. Summary of the Invention

[0007] In order to provide a basic understanding of some aspects of the disclosed embodiments, a brief summary is given below. The summary is not an extensive review, nor is it intended to identify key / critical elements or delineate the scope of protection of these embodiments, but rather serves as a prelude to the detailed description that follows.

[0008] The embodiments of the present disclosure provide a method, an apparatus, and an electronic device for evaluating the quality of a biological living environment, so as to improve the accuracy of habitat quality evaluation results.

[0009] In some embodiments, a method for evaluating the quality of a living environment of an organism includes: obtaining spatialized environmental data of a living environment to be evaluated in which the target organism is located; wherein the spatialized environmental data includes spatialized raster data of environmental data of the living environment to be evaluated; inputting the spatialized environmental data into a living environment quality evaluation model corresponding to the target organism to obtain a nest probability of the target organism's nest appearing in the living environment to be evaluated; wherein the living environment quality evaluation model is obtained by training a machine learning algorithm based on a gradient boosting framework; and determining a habitat quality evaluation result of the living environment to be evaluated for the target organism based on the nest probability.

[0010] Optionally, a living environment quality assessment model corresponding to the target organism is obtained in the following manner: a training data set is obtained based on the environmental data of the living environment of the target organism and the target organism's nest distribution data; the training data set includes a training set and a test set; a machine learning algorithm based on a gradient boosting framework is used to train a model on the training set, determine model parameters, and obtain an initial living environment quality assessment model; error evaluation and optimization training are performed on the initial living environment quality assessment model based on the test set to obtain a trained living environment quality assessment model.

[0011] Optionally, a training data set is obtained based on the environmental data of the target organism's living environment and the target organism's nest distribution data, including: spatially gridding the environmental data to obtain environmental grid data, and spatially gridding the target organism's nest distribution data to obtain label grid data; using the environmental grid data and the label grid data as a grid data set; extracting the grid data set into the corresponding spatial grid to obtain a training data set.

[0012] Optionally, the environmental data includes water source information, soil information, elevation and land use type; the environmental data is spatially gridded to obtain environmental grid data, including: spatially processing the water source information and soil information separately to obtain raster data; uniformly projecting the raster data such as water source information, soil information, elevation, land use type, etc. to obtain projection data; dividing the living environment to be evaluated into spatial grids of a preset size, and assigning the projection data to the spatial grid corresponding to the projection data to obtain environmental grid data.

[0013] Optionally, the method further includes: normalizing the elevation, soil information and water source information; and performing classification variable processing on the land use type.

[0014] Optionally, a machine learning algorithm based on a gradient boosting framework is used to perform model training on the training set to determine model parameters, including: performing model training on the training set based on a machine learning algorithm based on a gradient boosting framework to determine multiple initial parameter combinations; determining a target parameter combination from multiple initial parameter combinations based on preset evaluation indicators; and optimizing and adjusting the target parameter combination to obtain model parameters.

[0015] Optionally, the initial living environment quality assessment model is optimized and trained based on the test set to obtain a trained living environment quality assessment model, including: inputting the test set into the initial living environment quality assessment model to obtain a test result; evaluating the error of the initial living environment quality assessment model based on the test result to obtain an error evaluation result; and optimizing and adjusting the initial living environment quality assessment model based on the error evaluation result to obtain a trained living environment quality assessment model.

[0016] Optionally, based on the nest probability, the habitat quality assessment result of the living environment to be assessed for the target organism is determined, including: dividing the nest probability according to preset habitat quality assessment levels; and generating a habitat quality assessment map based on the divided nest probability.

[0017] In some embodiments, an apparatus for evaluating the quality of a living environment for a living organism includes a processor and a memory storing program instructions, wherein the processor is configured to execute the above-described method for evaluating the quality of a living environment for a living organism when executing the program instructions.

[0018] In some embodiments, the electronic device includes: an electronic device body; and the above-mentioned device for evaluating the quality of a biological living environment, which is installed in the electronic device body.

[0019] The method, device, and electronic device for evaluating the quality of a living environment for organisms provided by the embodiments of the present disclosure can achieve the following technical effects:

[0020] In the disclosed embodiment, the spatialized environmental data of the target organism's living environment to be evaluated is obtained, and a living environment quality assessment model obtained by training a machine learning algorithm based on a gradient boosting framework is input to obtain the nest probability of the target organism's nest in the living environment to be evaluated output by the model. The habitat quality assessment result of the living environment to be evaluated for the target organism is analyzed by the nest probability. The nest probability of the target organism's nest appearing in the living environment to be evaluated is predicted. The higher the nest probability, the higher the possibility of the target organism building a nest in the area, and the more suitable the area is for the survival of the target organism. By analyzing the nest probability, the habitat quality of the living environment to be evaluated for the target organism can be accurately obtained, and the accuracy of the habitat quality assessment result can be improved. In addition, after obtaining the habitat quality assessment result, protection can be carried out according to the area with a higher probability of the target organism's nest in the living environment to be evaluated, thereby improving the quality of the living environment of the target organism.

[0021] The above general description and the following description are exemplary and explanatory only and are not intended to limit the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] One or more embodiments are exemplarily described by corresponding drawings. These exemplary descriptions and drawings do not limit the embodiments. Elements with the same reference numerals in the drawings are shown as similar elements. The drawings do not constitute a scale limitation. In addition,

[0023] Figure 1 is a schematic diagram of a method for evaluating the quality of a living environment of a biological organism provided by an embodiment of the present disclosure;

[0024] Figure 2 is a schematic diagram of a method for constructing a living environment quality assessment model provided by an embodiment of the present disclosure;

[0025] Figure 3 is a schematic diagram of another method for constructing a living environment quality assessment model provided by an embodiment of the present disclosure;

[0026] Figure 4 is a schematic diagram of another method for evaluating the quality of a living environment for organisms provided by an embodiment of the present disclosure;

[0027] Figure 5 is a schematic diagram of a habitat quality assessment map for black-headed gulls provided in an embodiment of the present disclosure;

[0028] Figure 6 Schematic diagram of a device for evaluating the quality of a living environment for organisms provided by an embodiment of the present disclosure. DETAILED DESCRIPTION

[0029] In order to be able to understand the features and technical content of the embodiments of the present disclosure in more detail, the implementation of the embodiments of the present disclosure is described in detail below in conjunction with the accompanying drawings. The accompanying drawings are for reference only and are not used to limit the embodiments of the present disclosure. In the following technical description, for the sake of convenience of explanation, a full understanding of the disclosed embodiments is provided through multiple details. However, one or more embodiments can still be implemented without these details. In other cases, to simplify the drawings, well-known structures and devices can be simplified for display.

[0030] In the description and claims of the embodiments of the present disclosure, as well as in the accompanying drawings, the terms "first," "second," and the like are used to distinguish similar items and are not necessarily used to describe a particular order or precedence. It should be understood that the terms used in this manner are interchangeable where appropriate to describe the embodiments of the present disclosure herein. In addition, the terms "including," "having," and any variations thereof are intended to cover non-exclusive inclusions.

[0031] Unless otherwise stated, the term "plurality" means two or more.

[0032] In the embodiment of the present disclosure, the character " / " indicates that the preceding and following objects are in an "or" relationship. For example, A / B means: A or B.

[0033] The term "and / or" describes an association between objects, indicating that three relationships can exist. For example, A and / or B means: A or B, or A and B.

[0034] The term "correspondence" may refer to an association relationship or a binding relationship. The correspondence between A and B means that there is an association relationship or a binding relationship between A and B.

[0035] Combine Figure 1 As shown, an embodiment of the present disclosure provides a method for evaluating the quality of a living environment for organisms. The method may be executed by a processor and includes:

[0036] S001, the processor obtains spatialized environmental data of the living environment to be evaluated where the target organism is located; wherein the spatialized environmental data includes spatialized raster data of the environmental data of the living environment to be evaluated.

[0037] S002, the processor inputs the spatialized environmental data into the living environment quality assessment model corresponding to the target organism to obtain the nest probability of the target organism's nest in the living environment to be assessed; wherein, the living environment quality assessment model is obtained by training the machine learning algorithm based on the gradient boosting framework.

[0038] S003: The processor determines the habitat quality assessment result of the living environment to be assessed for the target organism based on the nest probability.

[0039] In the disclosed embodiment, the spatialized environmental data of the target organism's living environment to be evaluated is obtained, and a living environment quality assessment model obtained by training a machine learning algorithm based on a gradient boosting framework is input to obtain the nest probability of the target organism's nest in the living environment to be evaluated output by the model. The habitat quality assessment result of the living environment to be evaluated for the target organism is analyzed by the nest probability. The nest probability of the target organism's nest appearing in the living environment to be evaluated is predicted. The higher the nest probability, the higher the possibility of the target organism building a nest in the area, and the more suitable the area is for the survival of the target organism. By analyzing the nest probability, the habitat quality of the living environment to be evaluated for the target organism can be accurately obtained, and the accuracy of the habitat quality assessment result can be improved. In addition, after obtaining the habitat quality assessment result, protection can be carried out according to the area with a higher probability of the target organism's nest in the living environment to be evaluated, thereby improving the quality of the living environment of the target organism.

[0040] Optionally, a living environment quality assessment model corresponding to the target organism is obtained in the following manner: a training data set is obtained based on the environmental data of the living environment of the target organism and the target organism's nest distribution data; the training data set includes a training set and a test set; a machine learning algorithm based on a gradient boosting framework is used to train a model on the training set, determine model parameters, and obtain an initial living environment quality assessment model; error evaluation and optimization training are performed on the initial living environment quality assessment model based on the test set to obtain a trained living environment quality assessment model.

[0041] Combine Figure 2 As shown, an embodiment of the present disclosure provides a method for constructing a living environment quality assessment model. The execution subject of the method may be a processor, and the method includes:

[0042] S101, the processor obtains a training data set based on the environmental data of the target organism's living environment and the target organism's nest distribution data; the training data set includes a training set and a test set.

[0043] S102, the processor performs model training on the training set based on the machine learning algorithm of the gradient boosting framework, determines the model parameters, and obtains the initial living environment quality assessment model.

[0044] S103, the processor optimizes and trains the initial living environment quality assessment model according to the test set to obtain a trained living environment quality assessment model.

[0045] The training process of the living environment quality assessment model includes obtaining a training dataset, training the model, and optimizing the model. The training dataset is divided into a training set and a test set, with the training set and the test set accounting for 80% and 20%, respectively. A machine learning algorithm based on the gradient boosting framework is used to initially train the model using the training set, and then the model is optimized using the test set to obtain the trained living environment quality assessment model.

[0046] Optionally, a training data set is obtained based on the environmental data of the target organism's living environment and the target organism's nest distribution data, including: spatially gridding the environmental data to obtain environmental grid data, and spatially gridding the target organism's nest distribution data to obtain label grid data; using the environmental grid data and the label grid data as a grid data set; extracting the grid data set into the corresponding spatial grid to obtain a training data set.

[0047] In this embodiment, the environmental data and the target organism nest distribution data are spatially gridded, respectively, so that the living environment of the target organism can be divided into multiple spatial grids. In each spatial grid, the environmental data and the organism nest distribution data corresponding to the spatial grid are determined.

[0048] Optionally, the environmental data includes water source information, soil information, elevation and land use type; the environmental data is spatially gridded to obtain environmental grid data, including: spatially processing the water source information and soil information separately to obtain raster data; uniformly projecting the raster data such as water source information, soil information, elevation, land use type, etc. to obtain projection data; dividing the living environment to be evaluated into spatial grids of a preset size, and assigning the projection data to the spatial grid corresponding to the projection data to obtain environmental grid data.

[0049] In this embodiment, the environmental data that influences biological nesting includes water source information, soil information, elevation, and land use type. Water source and soil information vary in each area of ​​the target biological habitat, so each needs to be spatially processed to obtain corresponding raster data. The raster data is then projected together with the elevation and land use type to obtain projection data. Finally, the projection data is assigned to the corresponding spatial grid.

[0050] Optionally, the water source information includes the distance to the coastline and the distance to the riverbank.

[0051] Optionally, the soil information includes soil copper content, soil electrical conductivity, soil lead content, soil pH, soil temperature, normalized vegetation index, and vegetation height.

[0052] Optionally, the water source information is spatially processed, including: using ARCGIS (Advanced Raster and Geographic Information System) distance analysis to calculate the distance from the target biological habitat to the coastline and riverbank, and outputting it as raster data.

[0053] ARCGIS is a comprehensive platform for building and applying geographic information systems. Using ARCGIS's distance analysis capabilities, we can quickly determine the distance from the target organism's habitat to coastlines and riverbanks, and output it as raster data.

[0054] Optionally, the soil information is spatially processed, including: spatializing data such as vegetation height, soil copper content, soil electrical conductivity, soil lead content, soil pH, and soil temperature in the target biological living environment into raster data based on the ARCGIS Kriging interpolation method.

[0055] Kriging interpolation is a mathematical interpolation method based on the theory of spatial autocorrelation. It determines the location of the interpolated point, calculates the spatial weights of the surrounding data points to the interpolated point, and then uses this spatial weight to calculate the contribution of the surrounding data points to the interpolated point. Finally, the contribution of the surrounding data points to the interpolated point is used to estimate the value of the interpolated point. For soil data types, the soil information of adjacent data points often affects each other. Therefore, kriging interpolation can be used to interpolate the soil information within the target organism's habitat to obtain richer soil information data.

[0056] Optionally, the raster data, the elevation, and the land use type are projected, including: uniformly projecting the raster data, the elevation, and the land use type using a projection tool of ARCGIS.

[0057] The ArcGIS projection tool can convert geographic data from one coordinate system to another. In this embodiment, the raster data, elevation, and land use type need to be projected onto the same coordinate system to unify the different environmental data and facilitate the subsequent assignment of the spatial grid of the target biological habitat.

[0058] Optionally, the living environment to be evaluated is divided into spatial grids of a preset size, including: using a fishing net tool of ARCGIS to divide the space of the living environment of the target organism into spatial grids of a preset size.

[0059] In the ARCGIS Fishnet tool, by setting the parameters of the fishnet, the target organism's habitat can be divided into spatial grids. The preset spatial grid sizes include: 10m x 10m.

[0060] Optionally, assigning the projection data to a spatial grid corresponding to the projection data includes: using a spatial statistical tool to assign the projection data to the corresponding spatial grid using a grid mean.

[0061] In a spatial grid, there may be multiple environmental data collection points. The mean value of each environmental data of all collection points in the spatial grid is calculated as the environmental data of the spatial grid.

[0062] Optionally, the target organism nest distribution data is spatially gridded, including: after dividing the space of the target organism's living environment into spatial grids of a preset size, for each spatial grid, if there is a target organism nest, the label of the spatial grid is recorded as 1, otherwise the label of the spatial grid is recorded as 0.

[0063] Optionally, the method further includes: normalizing the elevation, soil information and water source information; and performing classification variable processing on the land use type.

[0064] In this example, elevation, soil information, and water source information vary significantly across different areas of the target organism's habitat. Directly using these data for model training can result in significant errors, so normalization is necessary. Furthermore, land use type is text data and can be processed as a categorical variable to classify different types of land use.

[0065] Alternatively, taking elevation as an example, the elevation can be normalized using the following formula:

[0066]

[0067] Among them, x inew is the normalized value of the ith elevation, x i is the value of the i-th elevation, x min is the minimum value of elevation, x max is the maximum value of the elevation.

[0068] Optionally, a machine learning algorithm based on a gradient boosting framework is used to perform model training on the training set to determine model parameters, including: performing model training on the training set based on a machine learning algorithm based on a gradient boosting framework to determine multiple initial parameter combinations; determining a target parameter combination from multiple initial parameter combinations based on preset evaluation indicators; and optimizing and adjusting the target parameter combination to obtain model parameters.

[0069] Combine Figure 3 As shown, the embodiment of the present disclosure provides another method for constructing a living environment quality assessment model, including:

[0070] S201, the processor obtains a training data set based on the environmental data of the target organism's living environment and the target organism's nest distribution data; the training data set includes a training set and a test set.

[0071] S202: The processor performs model training on the training set based on a machine learning algorithm of a gradient boosting framework to determine multiple initial parameter combinations.

[0072] S203: The processor determines a target parameter combination from a plurality of initial parameter combinations according to a preset evaluation index.

[0073] S204: The processor optimizes and adjusts the target parameter combination, obtains model parameters, and determines an initial living environment quality assessment model.

[0074] S205: The processor optimizes and trains the initial living environment quality assessment model according to the test set to obtain a trained living environment quality assessment model.

[0075] By training the model on the training set, multiple initial parameter combinations can be determined. An evaluation result is obtained for each initial parameter combination. Based on pre-set evaluation metrics, a target parameter combination can be determined from these multiple initial parameter combinations. Finally, the target parameter combination is optimized and adjusted to determine the optimal parameter combination, which serves as the model parameters.

[0076] Optionally, the machine learning algorithm based on the gradient boosting framework includes: LightGBM (Light GradientBoosting Machine).

[0077] LightGBM is a fast, distributed, and high-performance gradient boosting framework based on the gradient boosting framework. It is used for sorting, classification, and many other machine learning tasks. It is based on a tree algorithm and uses a histogram-based decision tree algorithm for optimization. LightGBM has faster training speed and lower memory consumption while maintaining high accuracy. Using the LightGBM algorithm, it can quickly process large amounts of data in the training set and determine the required multiple initial parameter combinations.

[0078] Optionally, according to preset evaluation indicators, a target parameter combination is determined among multiple initial parameter combinations, including: performing a search process on all parameters separately, and determining the optimal parameter of each parameter according to the preset evaluation indicators as the target parameter combination; wherein, performing a search process on the target parameter includes: using GridSearchCV (grid search cross validation) to control all parameters other than the target parameter, and determining the optimal parameter of the target parameter within the search range of the target parameter.

[0079] In this embodiment, in GridSearchCV, the search range of the target parameter can be determined by setting the parameter grid, and the preset evaluation index can be set by the ROC (Receiver Operating Characteristic) curve. The ROC curve uses the size of the area under the curve (AUC) to evaluate the model. The value range of AUC is between 0.5 and 1. The larger the AUC is, the closer it is to 1, and the better the diagnostic or predictive effect of the model. When AUC is between 0.5 and 0.7, the accuracy is low; when it is between 0.7 and 0.9, there is a certain accuracy; when AUC is above 0.9, the accuracy is high. GridSearchCV can perform an exhaustive search for parameter combinations. First, the search range for each parameter value is determined based on the parameter grid, and a list of possible values ​​for the parameter is obtained. Then, all possible parameter combinations composed of these values ​​are evaluated, and the best combination is determined based on the ROC curve. In addition, evaluating model performance through cross-validation helps to reduce the effects of overfitting or underfitting.

[0080] Optionally, the initial living environment quality assessment model is optimized and trained based on the test set to obtain a trained living environment quality assessment model, including: inputting the test set into the initial living environment quality assessment model to obtain test results; performing error evaluation on the initial living environment quality assessment model based on the test results; and optimizing and adjusting the initial living environment quality assessment model based on the error evaluation results to obtain a trained living environment quality assessment model.

[0081] According to the test set, the error evaluation of the initial living environment quality assessment model is carried out, and the nest probability output by the model is compared with the distribution probability of the target organism nest in the label grid data to determine the error evaluation result. Based on the error evaluation result, the initial living environment quality assessment model is optimized and adjusted, and the optimal model is saved as the living environment quality assessment model.

[0082] Optionally, based on the nest probability, the habitat quality assessment result of the living environment to be assessed for the target organism is determined, including: dividing the nest probability according to preset habitat quality assessment levels; and generating a habitat quality assessment map based on the divided nest probability.

[0083] Combine Figure 4 As shown, the embodiment of the present disclosure provides another method for evaluating the quality of a living environment of a biological organism, comprising:

[0084] S301, the processor obtains spatialized environmental data of the living environment to be evaluated where the target organism is located; wherein the spatialized environmental data includes spatialized raster data of the environmental data of the living environment to be evaluated.

[0085] In step S302, the processor inputs the spatialized environmental data into a living environment quality assessment model corresponding to the target organism to obtain the nest probability of the target organism's nest in the living environment to be assessed; wherein, the living environment quality assessment model is obtained by training a machine learning algorithm based on a gradient boosting framework.

[0086] S303: The processor divides the nest probability according to the preset habitat quality assessment level.

[0087] S304: The processor generates a habitat quality assessment map based on the divided nest probabilities.

[0088] The nest probability output by the habitat quality assessment model is a random value between 0 and 1. A higher nest probability indicates a higher likelihood that the target organism will nest in that area, and therefore a more suitable area for its survival. Therefore, we pre-set habitat quality assessment levels and categorize nest probabilities. For example, using four habitat quality assessment levels, areas with a nest probability of [0, 0.5] are classified as average habitat, [0.5, 0.7] as intermediate habitat, [0.7, 0.9] as good habitat, and [0.9, 0.1] as high-quality habitat. Based on the resulting nest probability, areas with different habitat quality assessment levels are marked with different colors to construct a habitat quality assessment map.

[0089] Optionally, the nest probabilities are divided according to preset habitat quality assessment levels, including dividing the nest probabilities according to the following formula:

[0090]

[0091] Among them, F(i) is the target habitat quality assessment level, lgbm(i) is the nest probability output by the living environment quality assessment model, i is the grid number, w m is the weighted number of habitat quality grades, and m is the grade threshold parameter.

[0092] In a specific embodiment, the method for evaluating the quality of a biological living environment provided by the embodiment of the present disclosure is described by taking the habitat quality evaluation of a black-headed gull living environment as an example.

[0093] First, we need to train a model for assessing the quality of the habitat of black-headed gulls. This model is based on the known environmental data and nest distribution data of black-headed gulls. Based on the above method, we set the objective function and the appropriate range of each LightGBM model parameter. The optimal number of training times is determined based on the model runtime. The optimal parameter combination is automatically searched based on the target value to obtain the model for assessing the quality of the habitat.

[0094] In an optional embodiment, the optimal parameter combination includes: n_estimators=100; learning_rate=0.1; num_leaves=31; max_depth=9; boosting_type=gbdt; min_child_weight=0.0001; min_gain_to_split=0; subsample_for_bin=200000; reg_alpha=0; reg_lambda=0.

[0095] Among them, n_estimators is the number of iterations of the LightGBM algorithm, which is set to 100 by default. Depending on the environment data, n_estimators can be selected between 100 and 1000. In addition, you can set a larger value and use it with early_stopping_round to allow the model to automatically select the best number of iterations based on performance. Choosing a larger number of iterations will achieve better performance on the training set but may easily lead to overfitting and degraded performance on the test set. learning_rate is the learning rate, which is set to 0.1 by default and is generally set between 0.05 and 0.1. Choosing a smaller learning rate can achieve stable and better model performance. Num_leaves is the number of leaf nodes on a tree in the tree algorithm. The default setting is 31. It is used with max_depth to determine the shape of the null value tree. It is generally set to a value in the range of (0, 2^max_depth-1]. Max_depth is the maximum depth of the tree model. It is generally limited to 3 to 5 and is an important parameter to prevent overfitting. Boosting_type is used to specify the type of weak learner. The default value is "gbdt", which means that a tree-based model is used for calculation. Min_child_weight is the minimum Hessian sum on a leaf. The default setting is 0.001. Min_gain_to_split is the amount of branching for a leaf node. The minimum loss reduction required, with a default value of 0. The larger the value, the more conservative the model. subsample_for_bin is the maximum number of samples that can be accommodated in a single box of a single feature. When constructing a column histogram, each feature will be binned to accommodate the samples. reg_alpha is the L1 regularization parameter, which is set to 0 by default. This parameter will not change significantly after feature selection. If it is found to be large, it means that there are some features with lower impact in the model, and the model needs to be adjusted to control overfitting. reg_lambda is the L2 regularization parameter, which is set to 0 by default. A larger value will make the influence of each feature on the model tend to be uniform, and no single feature will dominate the performance of the entire model.

[0096] Obtain spatialized environmental data for the to-be-assessed living environment of the black-headed gull. This spatialized environmental data includes raster data of the spatialized environmental data of the to-be-assessed living environment. In conjunction with the aforementioned description, raster data can also be obtained through various functions in ARCDIS. The to-be-assessed living environment is divided into multiple spatial grids, and the raster data for each spatial grid is determined using ARCDIS. For example, the raster data corresponding to one of the spatial grids is (17, 11, 2.081237, 798.4885, 1.448566, 7.14006, 31.92602, 45, 0.063204, 75.53122, 3791.695, 3791.695). Here, 17 is the grid cell number, 11 is the elevation, 2.081237 is the soil copper content, 798.4885 is the soil electrical conductivity, 1.448566 is the soil lead content, 7.14006 is the soil pH, 31.92602 is the soil temperature, 45 is the land use type, 0.063204 is the normalized difference vegetation index, 75.53122 is the vegetation height, 3791.695 is the distance from the coastline, and 3791.695 is the distance from the riverbank. The grid data for this grid cell was input into the habitat quality assessment model for black-headed gulls, resulting in a nest probability of 0.021994 for this grid cell. This indicates that the area in this grid cell is not suitable for black-headed gull nesting and is considered a general habitat.

[0097] A corresponding nest probability is obtained for each grid data of all spatial grids, and the nest probability is divided into habitat quality assessment levels. According to the divided nest probability, the areas with different habitat quality assessment levels are marked with different colors, so as to construct a Figure 5 The habitat quality assessment map shown. The blue part represents the general habitat area, the green part represents the medium habitat area, the orange part represents the good habitat area, and the red part represents the high-quality habitat area.

[0098] Combine Figure 6 As shown, an embodiment of the present disclosure provides a device 300 for evaluating the quality of a living environment for organisms, comprising a processor 100 and a memory 101. Optionally, the device may further comprise a communication interface 102 and a bus 103. The processor 100, the communication interface 102, and the memory 101 may communicate with each other via the bus 103. The communication interface 102 may be used for information transmission. The processor 100 may call the logic instructions in the memory 101 to execute the method for evaluating the quality of a living environment for organisms of the above embodiment.

[0099] In addition, the logic instructions in the memory 101 can be implemented in the form of software functional units and can be stored in a computer-readable storage medium when sold or used as an independent product.

[0100] Memory 101, as a computer-readable storage medium, can be used to store software programs and computer-executable programs, such as program instructions / modules corresponding to the methods in the embodiments of the present disclosure. Processor 100 executes the program instructions / modules stored in memory 101 to perform functional applications and data processing, thereby implementing the method for assessing the quality of a biological living environment in the above-mentioned embodiments.

[0101] The memory 101 may include a program storage area and a data storage area. The program storage area may store an operating system and at least one application required for a function; the data storage area may store data generated based on the use of the terminal device. Furthermore, the memory 101 may include high-speed random access memory and non-volatile memory.

[0102] An embodiment of the present disclosure provides an electronic device, comprising: an electronic device body, and the above-mentioned device for evaluating the quality of a living environment for organisms. The device for evaluating the quality of a living environment for organisms is installed on the electronic device body. The installation relationship described here is not limited to placement inside the electronic device, but also includes installation connections with other components of the electronic device, including but not limited to physical connections, electrical connections, or signal transmission connections. It can be understood by those skilled in the art that the device for evaluating the quality of a living environment for organisms can be adapted to a feasible electronic device body, thereby realizing other feasible embodiments.

[0103] An embodiment of the present disclosure provides a computer-readable storage medium storing computer-executable instructions, wherein the computer-executable instructions are configured to execute the above-mentioned method for evaluating the quality of a biological living environment.

[0104] The technical solutions of the embodiments of the present disclosure may be embodied in the form of a software product, which is stored in a storage medium and includes one or more instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in the embodiments of the present disclosure. The aforementioned storage medium may be a non-transitory storage medium, including: a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, an optical disk, or other media that can store program code.

[0105] The above description and the accompanying drawings fully illustrate the embodiments of the present disclosure so that those skilled in the art can practice them. Other embodiments may include structural, logical, electrical, process and other changes. The embodiments represent only possible variations. Unless explicitly required, individual components and functions are optional, and the order of operations may vary. Parts and features of some embodiments may be included in or replace parts and features of other embodiments. Moreover, the words used in this application are only used to describe the embodiments and are not used to limit the claims. As used in the description of the embodiments and claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to also include plural forms. Similarly, the term "and / or" as used in this application refers to any and all possible combinations of one or more associated listings. In addition, when used in this application, the term "comprise" and its variations "comprises" and / or comprising refer to the presence of stated features, wholes, steps, operations, elements, and / or components, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or groups of these. In the absence of further restrictions, an element defined by the sentence "comprising a..." does not exclude the presence of other identical elements in the process, method or device that includes the element. In this article, each embodiment may focus on the differences from other embodiments, and the same and similar parts between the various embodiments can be referenced to each other. For the methods, products, etc. disclosed in the embodiments, if they correspond to the method part disclosed in the embodiments, then the relevant parts can be found in the description of the method part.

[0106] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software may depend on the specific application and design constraints of the technical solution. The technicians may use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the embodiments of the present disclosure. The technicians will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0107] In the embodiments disclosed herein, the disclosed methods and products (including but not limited to devices, equipment, etc.) can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the units can be merely a logical functional division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between each other shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, and can be electrical, mechanical or other forms. The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the units may be selected to implement this embodiment according to actual needs. In addition, the functional units in the embodiments of the present disclosure may be integrated into a processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0108] The flowcharts and block diagrams in the accompanying drawings show the possible implementation architectures, functions and operations of the systems, methods and computer program products according to the embodiments of the present disclosure. In this regard, each box in the flowchart or block diagram can represent a module, program segment or part of the code, and the module, program segment or part of the code contains one or more executable instructions for implementing the specified logical functions. In some alternative implementations, the functions marked in the box can also occur in an order different from that marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, or they can sometimes be executed in the opposite order, which can depend on the functions involved. In the descriptions corresponding to the flowcharts and block diagrams in the accompanying drawings, the operations or steps corresponding to different boxes can also occur in an order different from that disclosed in the description, and sometimes there is no specific order between different operations or steps. For example, two consecutive operations or steps can actually be executed substantially in parallel, or they can sometimes be executed in the opposite order, which can depend on the functions involved. Each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, may be implemented by a dedicated hardware-based system that performs the specified function or action, or may be implemented by a combination of dedicated hardware and computer instructions.

Claims

1. A method for evaluating the quality of a living environment, characterized in that: include: Acquiring spatialized environmental data of the target organism's living environment to be evaluated; wherein the spatialized environmental data includes spatialized raster data of the environmental data of the living environment to be evaluated; dividing the living environment to be evaluated into a plurality of spatial grids, and determining raster data for each spatial grid; The spatialized environmental data is input into the living environment quality assessment model corresponding to the target organism to obtain the nest probability of the target organism in each spatial grid in the living environment to be assessed. The living environment quality assessment model is trained based on a machine learning algorithm based on the gradient boosting framework. Determine the habitat quality assessment results of the target organisms in the living environment to be assessed based on the nest probability; The living environment quality assessment model corresponding to the target organism is obtained in the following manner: a training data set is obtained based on the environmental data of the living environment of the target organism and the distribution data of the target organism's nest; the environmental data is environmental data that affects the nesting of the organism, including water source information, soil information, elevation and land use type; the training data set includes a training set and a test set; a machine learning algorithm based on the gradient boosting framework is used to train the training set, determine the model parameters, and obtain an initial living environment quality assessment model; the initial living environment quality assessment model is optimized and trained based on the test set to obtain a trained living environment quality assessment model; A training data set is obtained based on environmental data of the target organism's living environment and the target organism's nest distribution data, including: spatially gridding the environmental data to obtain environmental grid data, and spatially gridding the target organism's nest distribution data to obtain labeled grid data; using the environmental grid data and the labeled grid data as a grid data set; and extracting the grid data set into a corresponding spatial grid to obtain a training data set; Performing spatial gridding processing on environmental data to obtain environmental grid data, including: performing spatial processing on water source information and soil information to obtain raster data; uniformly projecting the raster data with elevation and land use type to obtain projection data; dividing the target biological habitat into spatial grids of a preset size, and assigning the projection data to the spatial grid corresponding to the projection data to obtain environmental grid data; The target organism nest distribution data is spatially gridded, including: after dividing the space of the target organism's living environment into spatial grids of preset sizes, for each spatial grid, if there is a target organism nest, the label of the spatial grid is recorded as 1, otherwise the label of the spatial grid is recorded as 0.

2. The method according to claim 1, characterized in that Also includes: Normalize elevation, soil information, and water source information; and, Land use types are treated as categorical variables.

3. The method according to claim 1, characterized in that Based on the machine learning algorithm of the gradient boosting framework, the model is trained on the training set and the model parameters are determined, including: The machine learning algorithm based on the gradient boosting framework is used to train the model on the training set and determine multiple initial parameter combinations; According to the preset evaluation indicators, determine the target parameter combination among multiple initial parameter combinations; Optimize and adjust the target parameter combination to obtain the model parameters.

4. The method according to claim 1, wherein The initial living environment quality assessment model is optimized and trained based on the test set to obtain a trained living environment quality assessment model, including: Input the test set into the initial living environment quality assessment model to obtain the test results; Conduct error assessment on the initial living environment quality assessment model based on the test results; According to the error evaluation results, the initial living environment quality evaluation model is optimized and adjusted to obtain a trained living environment quality evaluation model.

5. The method according to any one of claims 1 to 4, characterized in that Based on the nest probability, determine the habitat quality assessment results of the living environment to be assessed for the target organisms, including: The nest probability is divided according to the preset habitat quality assessment level; Based on the divided nest probabilities, a habitat quality assessment map is generated.

6. A device for evaluating the quality of a living environment, comprising a processor and a memory storing program instructions, characterized in that: The processor is configured to execute the method for evaluating the quality of a biological living environment according to any one of claims 1 to 5 when running the program instructions.

7. An electronic device, characterized in that: include: Electronic device body; The device for evaluating the quality of a living environment of a living being as claimed in claim 6 is installed in the electronic device body.

Images