Intelligent evaluation method and system for green space supplement cost based on vacancy probability

Abstract

The invention discloses an intelligent evaluation method and system for green space supplement cost based on vacancy probability, and relates to the technical field of urban and rural planning, landscape gardens and artificial intelligence. The method comprises the following steps: receiving a remote sensing image of a target area unit land, performing identification based on the remote sensing image to obtain an urban vacant land and a low-efficiency construction land, and summarizing the urban vacant land and the low-efficiency construction land to obtain an overall vacant land identification result. According to the method, the remote sensing image and the multi-source data are comprehensively used, the vacant land types including the urban outdoor vacant land and the low-efficiency construction land can be recognized, the recognition range of the vacant land is more comprehensive, the random forest regression algorithm is adopted to predict the vacant probability, dynamic prediction of the land vacant state can be achieved, and the prediction efficiency is improved. Meanwhile, based on the vacancy probability and the supplement cost, the green space supplement priority is determined, and the feasibility of the green space supplement scheme is improved based on green supplement cost evaluation.

Description

Technical Field

[0001] The present invention relates to the fields of urban and rural planning, landscape gardening and artificial intelligence technology, and in particular to an intelligent evaluation method and system for green space supplementary cost based on vacancy probability. Background Art

[0002] In recent years, the phenomenon of urban shrinkage has spread across China and will become the norm in China's urban development in the future. In this process, the number of vacant urban land has increased, causing multiple problems in the ecological environment, social economy and other aspects. Green space has been proven to be one of the effective means to solve the problems caused by vacant urban land, and it has the functions of improving the level of ecosystem services, alleviating social problems and driving economic development.

[0003] The existing patent (Announcement No.: CN116310786A) discloses a method for identifying and / or implementing the potential for supplementing streetside micro-green spaces based on deep learning. For the supplementation of streetside micro-green spaces, the spatial conditions, demand conditions, facility conditions, and perception conditions of potential supplementation points are evaluated based on street view images, and the evaluation results of the supplementation potential of streetside micro-green spaces are obtained by integrating the evaluation results from multiple aspects. However, the evaluation method based on spatial conditions, demand conditions, facility conditions, perception conditions, etc. using street view images, although it reflects the construction feasibility of the site to a certain extent, is limited by the coverage of street view images and cannot evaluate the green space supplementation conditions of land areas lacking street view coverage. In addition, it only focuses on the pattern optimization before and after the green space supplementation, and cannot reflect the cost feasibility of the green space supplementation. For this reason, we propose an intelligent evaluation method and system for the cost of green space supplementation based on vacancy probability. Summary of the invention

[0004] The purpose of the present invention is to provide an intelligent evaluation method and system for green space supplement cost based on vacancy probability, which can predict the future vacancy probability of land based on historical and current vacant land conditions, evaluate the green space supplement potential in combination with the green space supplement cost, and provide optimization direction for subsequent green space supplement. The proposed optimization scheme fully considers the development background of urban shrinkage, and the optimization direction is highly targeted.

[0005] According to a first aspect of the present invention, in order to achieve the above-mentioned purpose, the present invention provides the following technical solution: a method for intelligently evaluating the green space supplementary cost based on vacancy probability, comprising the following steps:

[0006] Receive remote sensing images of unit land in the target area, obtain urban vacant land and inefficient construction land based on remote sensing image recognition, and summarize urban vacant land and inefficient construction land to obtain the overall vacant land identification result;

[0007] Based on the overall vacant land identification results, the driving factors were calculated using the random forest regression algorithm. The relationship model between (x) and the expansion of vacant land is divided into the driving factor value Input into the finally constructed relationship model to obtain the probability of non-vacant land being converted into vacant land;

[0008] Calculate the green space supplement cost of the unit land in the target area, where the green space supplement cost includes land acquisition cost, land remediation cost and green space construction optimization cost;

[0009] The unit supplement potential is calculated based on the probability of conversion to vacant land and the green space supplement cost. The green space supplement priority of the unit land is determined based on the calculation result. The unit land is converted into green space in order from high to low priority. The form of converted green space is determined according to the unit land type. The supplement cost of the green space supplement plan is calculated. If the supplement cost is lower than the supplement budget, the plan will be included in the alternative supplement plan. Otherwise, the supplement plan will be modified until the supplement plan meets the supplement budget requirements.

[0010] Furthermore, the remote sensing images are from Google Maps, Tiandi Map or Gaofen series optical remote sensing satellites, and the screening criteria for remote sensing images are that the cloud cover of the image is less than 10% and the spatial resolution is not less than 2.5m.

[0011] Furthermore, based on remote sensing image recognition, urban vacant land and inefficient construction land are obtained, and the urban vacant land and inefficient construction land are summarized to obtain the overall vacant land identification results, as follows:

[0012] (31) Identification of urban vacant land based on remote sensing images, including:

[0013] (31.1) Based on the definition of urban vacant land and the actual situation of the research site, a land use type system is constructed:

[0014] The urban vacant land is divided into n types. According to the definition of land use type, the ArcGIS Pro classification tool is used to select training samples, and automatic classification is performed based on the samples. The accuracy of classification is evaluated using the accuracy assessment tool.

[0015] The land use type system includes bare land, natural grassland, artificial grassland, natural shrubs, woodland, area under construction, residential land, commercial and office land, industrial land, hard square, and water body;

[0016] (31.2) Use ArcGIS pro image classification tools to delineate the number of samples to 50 for each type, and evenly distribute the samples in the target area. Select support vector machine as the classifier, select the delineated samples as training samples, identify the land use types in remote sensing images, and use the accuracy assessment tool to evaluate the recognition results;

[0017] (31.3) Input the study unit, which is the smallest grid unit for green space supplementation. The grid size is 10m-20m. The land use data is resampled to the grid size using the majority resampling method.

[0018] (32) Identification of inefficient construction land based on multi-source data:

[0019] (32.1) Based on the land use data of the research unit identified in (31), the construction land type units are extracted and an evaluation system for the inefficiency of each construction land type is constructed, including:

[0020] (32.11) Extract construction land type units, including residential land units, commercial and office land units, and industrial land units;

[0021] (32.12) Establish an evaluation system for the inefficiency of each type of construction land:

[0022] For each type of construction land, set n evaluation indicators (X 1 , X 2 , …X n ), for this type of land unit, the unit's low efficiency is the unit's evaluation index value (x 1 , x 2 , …x n ), the specific expression is as follows:

[0023]

[0024] In the formula, C represents the inefficiency of the construction land type unit, and n represents the number of evaluation indicators;

[0025] (32.2) After completing the inefficiency assessment of each type of construction land unit, the inefficiency is normalized and a certain proportion of land is selected as inefficient construction land;

[0026] (33) Summary of vacant land identification results:

[0027] The urban vacant land identified in (31) and the inefficient construction land identified in (32) are aggregated, and the distribution locations of urban vacant land and inefficient construction land are obtained respectively.

[0028] Furthermore, in step (32), the evaluation indicators of the residential land unit include road intersection density, service facility density, and night lighting index; the evaluation indicators of the commercial land unit include popularity index and commercial facility density; and the evaluation indicators of the industrial land unit include building density and GDP.

[0029] Furthermore, based on the overall vacant land identification results, the random forest regression algorithm was used to calculate the driving factors The relationship model between the expansion of vacant land and the driving factor value Input into the finally constructed relationship model, and obtain the probability of non-vacant land being converted into vacant land, as follows:

[0030] (51) The distribution locations of urban vacant land and inefficient construction land obtained in (33) are superimposed to obtain the area of ​​vacant land expansion. The relationship model between driving factors and vacant land expansion is constructed using the random forest regression algorithm, as follows:

[0031] (51.1) For study unit i, the study unit with expanded vacant land is assigned a value of 1, and the rest of the study units are assigned a value of 0. Whether it is expanded is taken as the dependent variable, all driving factors are normalized, and the normalized driving factors are taken as the independent variables;

[0032] (51.2) Aggregate the independent variables and dependent variables of all research units in the target area to form a training sample set, divide the training sample set into a training set and a test set, use the training set to train the random forest regression algorithm, and use the test set to evaluate the training results;

[0033] (51.3) During the training process, the performance of the random forest regression algorithm was adjusted by adjusting the number of decision trees, the maximum depth of the decision tree, the minimum number of samples for node splitting, the minimum number of samples for leaf nodes, and the number of features for node splitting;

[0034] The basic formula of the random forest regression algorithm is:

[0035]

[0036] Where X = {X 1 ,X 2 ,…X n} is a feature vector containing n driving factors, M is the number of decision trees, T m (X) is the predicted value of the mth decision tree for the input feature vector X. Each decision tree T m (X)

[0037] Process the input features according to its own tree structure and splitting rules, and finally give a predicted value;

[0038] (51.4) After completing each round of training, use the test set to evaluate, and use the mean square error, mean absolute error, and R square indicators to evaluate the explanatory power of the constructed regression model. Output all models with R square greater than the threshold, and select the model with the smallest mean square error and mean absolute error as the final constructed relationship model;

[0039] The mean square error is calculated as:

[0040]

[0041] Where y i is the actual value of the test set, is the value calculated based on the random forest regression model, and n is the number of samples in the test set;

[0042] The mean absolute error (RMSE) is calculated as:

[0043]

[0044] In the formula, y i is the actual value of the test set, is the value calculated based on the random forest regression model, and n is the number of samples in the test set;

[0045] The calculation method of R squared is:

[0046]

[0047] In the formula, y i is the actual value of the test set, is the value calculated based on the random forest regression model, and n is the number of samples in the test set;

[0048] (52) Vacancy probability calculation:

[0049] For each non-vacant land research unit, all driving factor values ​​X are input into the final constructed relationship model, which is the probability Pi of the research unit being transferred to vacant land:

[0050] The formula for the relational model is:

[0051] Where X = {X 1 ,X 2 ,…X n} is a feature vector containing n driving factors, M is the number of decision trees in the final relationship model, T m (X) is the predicted value of the mth decision tree for the input feature vector X;

[0052] For the research unit that is currently vacant land, the probability that it will be converted into vacant land in the future is 1.

[0053] Furthermore, the green space supplement cost of the target area unit land is calculated as follows:

[0054] C=C1+C2+C3

[0055] In the formula, C represents the green space supplement cost, C1 represents the land acquisition cost, C2 represents the land remediation cost, and C3 represents the green space construction optimization cost.

[0056] Furthermore, the unit supplement potential is calculated based on the probability of conversion to vacant land and the cost of green space supplement. The priority of green space supplement for unit land is determined based on the calculation result. A green space supplement plan is formulated based on the priority, and the supplement cost of the green space supplement plan is calculated as follows:

[0057] (71) Unit supplement potential evaluation:

[0058] Normalize the cost to get the final comprehensive cost C i ', the normalized formula is:

[0059]

[0060] In the formula, C i is the additional cost of the unit green space, C min is the minimum value of green space supplement cost in all units, C max is the maximum value of green space supplement cost in all units, C i ' is the normalized comprehensive cost;

[0061] The unit supplementation potential calculation formula is:

[0062] T i =P i *(1-C i ')

[0063] (72) Unit Supplement Priority Setting

[0064] Priorities are set based on the calculation result T of the unit supplementation potential. The higher the unit supplementation potential, the higher the priority of green space supplementation. The priorities are divided into 5 levels, from low to high:

[0065] Very low [0, 0.2], low [0.2, 0.4], medium [0.4, 0.6], high [0.6, 0.8], very high [0.8, 1.0];

[0066] (73) Verification of green space supplement plan

[0067] Formulate a green space supplement plan based on the unit green space supplement priority, convert the units into green spaces in order from high to low priority, and determine the form of the converted green space based on the land use type around the unit. After each green space supplement plan is completed, calculate the green supplement cost required for the plan and compare it with the supplement budget. If it is lower than the supplement budget requirement, include the plan in the alternative supplement plan. Otherwise, modify the supplement plan until the plan meets the supplement budget requirement.

[0068] According to a second aspect of the present invention, the present invention provides a system for intelligently evaluating green space supplementary cost based on vacancy probability, which is used to implement the above-mentioned intelligent evaluation method for green space supplementary cost based on vacancy probability, comprising:

[0069] The vacant land identification module is used to receive remote sensing images of unit land in the target area, obtain urban vacant land and inefficient construction land based on remote sensing image recognition, and summarize the urban vacant land and inefficient construction land to obtain the overall vacant land identification result;

[0070] The vacancy probability prediction module is used to calculate the driving factors based on the overall vacant land identification results using the random forest regression algorithm The relationship model between the expansion of vacant land and the driving factor value Input into the finally constructed relationship model to obtain the probability of non-vacant land being converted into vacant land;

[0071] The additional cost calculation module is used to calculate the additional cost of green space in the target area unit land, where the additional cost of green space includes land acquisition cost, land remediation cost and green space construction optimization cost;

[0072] The supplementary scheme derivation and verification module is used to calculate the unit supplementary potential according to the probability of conversion to vacant land and the green space supplementary cost, determine the unit land green space supplementary priority according to the calculation result, convert the unit land into green space in order from high to low priority, determine the converted green space form according to the unit land type, calculate the supplementary cost of the green space supplementary scheme, if the supplementary cost is lower than the supplementary budget, the scheme will be included in the alternative supplementary scheme, otherwise the supplementary scheme will be modified until the supplementary scheme meets the supplementary budget requirements.

[0073] According to a third aspect of the present invention, the present invention provides a terminal device, including a memory, a processor, and a computer program stored in the memory and capable of running on the processor, wherein the memory stores a computer program capable of running on the processor, and when the processor loads and executes the computer program, the above-mentioned intelligent evaluation method for green space supplementary cost based on vacancy probability is adopted.

[0074] According to a fourth aspect of the present invention, the present invention provides a storage medium comprising computer executable instructions, which, when executed by a computer processor, are used to execute the above-mentioned intelligent evaluation method for green space supplementary cost based on vacancy probability.

[0075] The present invention has at least the following beneficial effects:

[0076] 1. The present invention uses remote sensing images and multi-source data in combination to identify two types of vacant land, namely, urban outdoor space and inefficient construction land. The identified vacant land includes completely vacant land with no development purpose, abandoned and vacant land from construction and development purposes, and land that has been built but not fully utilized, so the identification scope of vacant land is more comprehensive.

[0077] 2. Compared with previous solutions that only consider the current land use situation, the present invention adopts a random forest regression algorithm to predict the vacancy probability, which can analyze the land vacancy mechanism and further calculate the vacancy probability of the land, thereby realizing a dynamic prediction of the vacancy status of the land.

[0078] 3. The present invention determines the priority of green space supplementation based on vacancy probability and supplementation cost, and improves the feasibility of the green space supplementation plan based on green supplementation cost evaluation. Compared with the existing technology that only considers the benefits after green space construction, this invention fully considers the construction feasibility.

[0079] Of course, any product implementing the present invention does not necessarily need to achieve all of the advantages described above at the same time. BRIEF DESCRIPTION OF THE DRAWINGS

[0080] Figure 1 Schematic diagram of the process of the method described in the embodiment of the present invention;

[0081] Figure 2 Schematic diagram of the framework principle of the method structure described in the embodiment of the present invention. DETAILED DESCRIPTION

[0082] The following will be combined with the drawings in the embodiments of the present disclosure to clearly and completely describe the technical solutions in the embodiments of the present disclosure. Obviously, the described embodiments are only part of the embodiments of the present disclosure, rather than all the embodiments. Based on the embodiments in the present disclosure, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present disclosure.

[0083] Embodiment 1:

[0084] See also Figure 1 and Figure 2 The present invention provides a technical solution: an intelligent evaluation method for green space supplementation cost based on vacancy probability, comprising the following steps:

[0085] S1: Vacant Land Identification

[0086] Vacant land identification includes S11 urban vacant land identification based on remote sensing images, S12 inefficient construction land identification based on multi-source data, and S13 vacant land distribution summary;

[0087] S11: Identification of urban vacant land based on remote sensing images, including:

[0088] (S11.1) Collect remote sensing images of the study area that meet the study period and accuracy requirements:

[0089] Obtain high-resolution remote sensing images within the study area (target area). Remote sensing images can come from free high-resolution data such as Google Maps, Esri World Image, and Tiandi Map, or paid high-resolution data such as Gaofen series, GeoEye, Quickbird, and Ikonos;

[0090] Data screening requirements include image cloud cover less than 10% and spatial resolution not less than 2.5m. For the technical solution of this example, 17-level Google map remote sensing images with a spatial resolution of 2.15m were selected.

[0091] Remote sensing images of the study area in 2023 and 2017 were downloaded;

[0092] (S11.2) Based on the definition of urban vacant land and the actual situation of the research site, a land use type system is constructed:

[0093] According to the definition of urban vacant land and the actual situation of the research site, urban land is divided into n types. According to the definition of land type, ArcGIS pro classification tool is used to select training samples, automatic classification is performed based on the samples, and the classification accuracy is evaluated using accuracy assessment tools;

[0094] It needs to be further explained that in this example, the land use type system is subdivided into 14 types, including bare land, natural grassland, artificial grassland, natural shrubs, woodland, under-construction area, residential land, commercial land, industrial land, hard square, water body, etc.; according to the definition of urban vacant land, that is, completely vacant land without development purposes within the city, and abandoned, vacant or underutilized land from construction purposes; among them, bare land, natural grassland, artificial grassland, natural shrubs, woodland, and under-construction area belong to urban vacant land type, and residential land, commercial land, and industrial land belong to construction land type;

[0095] (S11.3) Using ArcGIS Pro image classification tools, the number of samples for each type is delineated to 50, and each sample is evenly distributed in the target area. Support Vector Machine is selected as the classifier, and the delineated samples are selected as training samples to identify the land use type of the remote sensing image to be identified. The accuracy evaluation tool is used to evaluate the identification results. In this example, the output model accuracy reaches more than 80% of the land use type identification results, and the identification accuracy is consistent with the spatial resolution of the input image;

[0096] It should be noted that the support vector machine is a built-in tab in the ArcGIS pro image classification tool and can be called directly;

[0097] (S11.4) Input the research unit, which is the smallest grid unit for subsequent green space supplementation. The grid size is determined according to the research needs, preferably 10m-20m. The land use data is resampled to the grid size using the majority resampling method. In this embodiment, the land use data is resampled to 10m using the majority resampling method;

[0098] S12: Identification of inefficient construction land based on multi-source data, including:

[0099] (S12.1) Based on the land use data of the research unit identified in S11, the construction land type units are extracted and the inefficiency evaluation system of each construction land type is constructed:

[0100] Extract construction land type units, which in this example include residential land, commercial land, and industrial land units;

[0101] Construct an evaluation system for the inefficiency of each type of construction land. For each type of construction land, set n evaluation indicators (X 1 , X 2 , …X n ), for this type of land, the unit's inefficiency is the unit's evaluation index value (x 1 , x 2 , …x n ) is superimposed with equal weights, and the expression is as follows:

[0102]

[0103] In the formula, C represents the inefficiency of the construction land type unit, and n represents the number of evaluation indicators;

[0104] It should be further explained that in this example, the evaluation indicators of residential land include road intersection density, service facility density, and night light index. The road intersection data comes from the OSM road network, and the intersections of the road network are extracted through ArcGIS. The service facility density comes from Amap, and the POIs of "catering services, shopping services, life services, sports and leisure services, and medical care services" are screened out. The night light index comes from NPP / VIIRS data; the evaluation indicators of commercial land include popularity index and commercial facility density. The popularity index comes from Baidu heat map, and the commercial facility density data comes from Amap; the evaluation indicators of industrial land include building density and GDP. The building density comes from remote sensing satellite images, and the GDP data comes from the resource and environmental science data platform;

[0105] (S12.2) After completing the inefficiency evaluation of the construction land type units, the inefficiency is normalized and a certain proportion of the land is selected as the inefficient construction land. In this example, according to the actual situation, the units with the lowest inefficiency evaluation of each type are selected as the inefficient construction land;

[0106] S13: Summary of vacant land identification results

[0107] The urban vacant land identified by S11 and the inefficient construction land in S12 are included in the urban vacant land unit to obtain the distribution locations of the vacant land in the two phases respectively;

[0108] S2: Vacancy probability prediction, based on the overall vacant land identification results, the driving factors are calculated using the random forest regression algorithm The relationship model between the expansion of vacant land and the driving factor value Input into the final constructed relationship model to obtain the probability of non-vacant land being converted into vacant land, including S21 exploration of vacant land change mechanism and S22 vacancy probability calculation, as follows:

[0109] S21: Exploration of change mechanisms

[0110] The distribution positions of vacant land in the two phases obtained by S13 are superimposed to obtain the area of ​​vacant land expansion. The relationship between driving factors and vacant land expansion is constructed through regression algorithm, as follows:

[0111] For study unit i, the study unit with expanded vacant land is assigned a value of 1, and the rest of the study units are assigned a value of 0. Whether it is expanded is taken as the dependent variable, all driving factors are normalized, and the normalized driving factors are taken as the independent variables;

[0112] The independent variables and dependent variables of all research units in the research area are aggregated to form a training sample set, and the training sample set is randomly divided into a training set and a test set according to a certain ratio. The training set is used to train the random forest regression algorithm, and the test set is used to evaluate the training results.

[0113] During the training process, the performance of the random forest regression algorithm was adjusted by adjusting the number of decision trees (n_estimators), the maximum depth of the decision tree (max_depth), the minimum number of samples for node splitting (min_samples_split), the minimum number of samples for leaf nodes (min_samples_leaf), and the number of features for node splitting (max_features);

[0114] The basic formula of the random forest regression algorithm is:

[0115]

[0116] Where X = {X 1 ,X 2 ,…X n} is a feature vector containing n driving factors, M is the number of decision trees, T m (X) is the predicted value of the mth decision tree for the input feature vector X. Each decision tree T m (X) Processes the input features according to its own tree structure and splitting rules, and finally gives a prediction value. In the actual training process, the RandomForestRegressor function of the Python scikitlearn library is directly called for calculation;

[0117] After completing each round of training, the test set is used for evaluation. The explanatory power of the constructed regression model is evaluated through indicators such as mean squared error (MSE), mean absolute error (MAE), and R-squared (R2). All models with R-squared greater than a certain threshold are output, and the model with the smallest mean squared error and mean absolute error is selected as the final constructed relationship model; the calculation method of mean squared error is:

[0118]

[0119] Where y i is the actual value of the test set, is the value calculated based on the random forest regression model, and n is the number of samples in the test set;

[0120] In the actual training process, the mean square error is calculated by directly calling the mean_squared_error function of the Python scikitlearn library;

[0121] The mean absolute error (RMSE) is calculated as:

[0122]

[0123] Among them, y i is the actual value of the test set, is the value calculated based on the random forest regression model, and n is the number of samples in the test set;

[0124] The calculation method of R squared is:

[0125]

[0126] Among them, y i is the actual value of the test set, is the value calculated based on the random forest regression model, and n is the number of samples in the test set;

[0127] In this example, ArcGIS was used to apply the Erase tool to output the location of vacant land expansion;

[0128] Furthermore, in this example, referring to the relevant research on the evolution mechanism of construction land in shrinking cities, the slope (x 1 ), water body sensitivity (x 2 ), distance to the city center (x 3 ), commercial facility density (x 4 ), transportation facility density (x 5 )、House price(x 6 ), population density (x 7 ), hard surface ratio (x 8 ) as the driving factor;

[0129] In this example, the R-squared is greater than the threshold value and is taken as 0.7;

[0130] It should be further explained that the regression algorithm used can be a random forest regression algorithm, a Markov transition matrix algorithm or an artificial neural network algorithm. However, the artificial neural network algorithm takes a long time to train and is prone to overfitting. It is not suitable for conditions with limited data and it is difficult to explain the impact of different factors on the vacancy probability. The Markov transition matrix algorithm predicts the conversion of different land use types to vacant land in the next period based on the transfer probability of the previous period. This prediction method ignores the impact of other factors on land vacancy except land use factors. Therefore, the random forest regression algorithm is used in this example to estimate the future land vacancy status.

[0131] S22: Vacancy probability calculation

[0132] For each non-vacant land research unit, all driving factor values ​​X are input into the final constructed relationship model, which is the probability Pi of the research unit being transferred to vacant land:

[0133] The formula for the relational model is:

[0134] In the formula, X={X 1 ,X 2 ,…X n} is a feature vector containing n driving factors, M is the number of decision trees in the final relationship model, T m (X) is the predicted value of the mth decision tree for the input feature vector X;

[0135] The driving factor value The probability of converting non-vacant land into vacant land is obtained by inputting the final constructed relationship model. In actual operation, the trained rf model is used to calculate by calling Python's predict method. Pi = rf.predict(X);

[0136] For the research unit that is currently vacant land, the probability that it will be converted into vacant land in the future is 1;

[0137] S3: Supplemental cost assessment

[0138] According to the green space construction life cycle theory, the green space supplement cost is calculated. The green space supplement cost (C) includes land acquisition cost (C1), land remediation cost (C2), and green space construction optimization cost (C3), as follows:

[0139] C=C1+C2+C3

[0140] Furthermore, according to the current land use situation obtained in S1, the land acquisition cost (C1) and land remediation cost (C2) are determined according to the regional situation and the budget quota standards of relevant land development and consolidation projects, and the green space construction optimization cost (C3) is determined according to the type of green space to be added;

[0141] S4: Supplementation scheme derivation, including S41 unit supplementation potential evaluation, S42 unit supplementation priority setting, S43 supplementation scheme comparison and selection;

[0142] Evaluation of the potential for supplementation of S41 unit

[0143] The unit supplement potential (T) comprehensively considers the vacancy probability (P) of S2 and the comprehensive supplement cost (C) calculated by S3. Previous studies have shown that the higher the vacancy probability and the lower the supplement cost, the higher the feasibility of green space supplement. In order to unify the statistical dimensions, each type of cost is normalized to obtain the final comprehensive cost C'. The normalization formula is:

[0144]

[0145] In the formula, C i is the additional cost of the unit green space, C min is the minimum value of green space supplement cost in all units, C max is the maximum value of green space supplement cost in all units, C i ' is the normalized comprehensive cost;

[0146] The unit supplementation potential calculation formula is:

[0147] T i =P i *(1-C i ')

[0148] S42 Unit Supplement Priority Setting

[0149] According to the evaluation results of unit supplementation potential (T), the unit with higher unit supplementation potential has higher priority for green space supplementation;

[0150] In this example, the green space replenishment priority is divided into 5 levels according to the unit replenishment potential (T) with an interval of 0.2, from low to high: very low [0, 0.2), low [0.2, 0.4), medium [0.4, 0.6), high [0.6, 0.8), very high [0.8, 1.0];

[0151] S43 Green Space Supplement Scheme Verification

[0152] A green space supplementation plan is formulated according to the unit green space supplementation priority. Units are converted into green spaces in order from high to low priority. The converted green space form is determined according to the land use type around the unit. After each green space supplementation plan is formulated, the supplementation cost required for the plan is calculated according to S3 and compared with the supplementation budget. If it is lower than the supplementation budget requirement, the plan is included in the alternative supplementation plan. Otherwise, the supplementation plan is modified until the plan meets the supplementation budget requirement.

[0153] It should be noted that the technical solution of this embodiment can also realize the cost evaluation of converting vacant land to other land types by modifying S3 to add cost evaluation items. For example, by calculating the cost of new commercial facilities, the supplementary cost evaluation of commercial facilities can be realized.

[0154] In summary, the present invention can identify two types of vacant land, including urban outdoor spaces and inefficient construction land, by comprehensively using remote sensing images and multi-source data. The identified vacant land includes completely vacant land with no development purpose, abandoned and vacant land from construction and development purposes, and land that has been built but not fully utilized. The identification scope of vacant land is more comprehensive, and compared with the previous solution that only considers the current land situation, the random forest regression algorithm is used to predict the vacancy probability, which can analyze the land vacancy mechanism and then calculate the vacancy probability of the land, thereby realizing the dynamic prediction of the vacancy status of the land. At the same time, based on the vacancy probability and the supplementary cost, the priority of green space supplementation is determined, and the feasibility of the green space supplementation plan is improved based on the green supplementary cost evaluation. Compared with the existing technology that only considers the benefits after green space construction, the construction feasibility is fully considered.

[0155] Embodiment 2:

[0156] This embodiment provides a system for intelligently evaluating the supplementary cost of green space based on vacancy probability, which is used to implement the intelligent evaluation method for supplementary cost of green space based on vacancy probability described in the first embodiment, including:

[0157] The vacant land identification module is used to receive remote sensing images of unit land in the target area, obtain urban vacant land and inefficient construction land based on remote sensing image recognition, and summarize the urban vacant land and inefficient construction land to obtain the overall vacant land identification result;

[0158] The vacancy probability prediction module is used to calculate the driving factors based on the overall vacant land identification results using the random forest regression algorithm The relationship model between the expansion of vacant land and the driving factor value Input into the finally constructed relationship model to obtain the probability of non-vacant land being converted into vacant land;

[0159] The additional cost calculation module is used to calculate the additional cost of green space in the target area unit land, where the additional cost of green space includes land acquisition cost, land remediation cost and green space construction optimization cost;

[0160] The supplementary scheme derivation and verification module is used to calculate the unit supplementary potential according to the probability of conversion to vacant land and the green space supplementary cost, determine the unit land green space supplementary priority according to the calculation result, convert the unit land into green space in order from high to low priority, determine the converted green space form according to the unit land type, calculate the supplementary cost of the green space supplementary scheme, if the supplementary cost is lower than the supplementary budget, the scheme will be included in the alternative supplementary scheme, otherwise the supplementary scheme will be modified until the supplementary scheme meets the supplementary budget requirements.

[0161] Specifically, the above-mentioned vacant land identification module, vacancy probability prediction module, supplementary cost calculation module and supplementary plan derivation and verification module can be embedded in a computer processing system. The computer calls the above-mentioned modules to complete the task of evaluating the green space supplementary plan according to the above-mentioned intelligent evaluation method for green space supplementary cost based on vacancy probability; the above-mentioned vacant land identification module, vacancy probability prediction module, supplementary cost calculation module and supplementary plan derivation and verification module can perform operations according to the specific steps given in the above-mentioned intelligent evaluation method for green space supplementary cost based on vacancy probability.

[0162] It should be noted that it should be understood that the division of the various modules of the above system is only a division of logical functions. In actual implementation, they can be fully or partially integrated into one physical entity, or they can be physically separated. These modules can all be implemented in the form of software calling through processing elements; they can also be all implemented in the form of hardware; some modules can also be implemented in the form of software calling through processing elements, and some modules can be implemented in the form of hardware. For example, the vacant land identification module can be a separately established processing element, or it can be integrated in a chip of the above device. In addition, it can also be stored in the memory of the above device in the form of program code, and called and executed by a processing element of the above device. The implementation of other modules is similar. In addition, these modules can be fully or partially integrated together, or they can be implemented independently. The processing element described here can be an integrated circuit with signal processing capabilities. In the implementation process, each step of the above method or each of the above modules can be completed by the hardware integrated logic circuit in the processor element or software instructions.

[0163] For example, the above modules may be one or more integrated circuits configured to implement the above methods, such as one or more application specific integrated circuits (ASIC), or one or more digital singnal processors (DSP), or one or more field programmable gate arrays (FPGA). For another example, when a module is implemented in the form of a processing element scheduling program code, the processing element may be a general-purpose processor, such as a central processing unit (CPU) or other processors that can call program code. For another example, these modules may be integrated together and implemented in the form of a system-on-a-chip (SOC).

[0164] Embodiment three:

[0165] The present invention provides a terminal device, including a memory, a processor, and a computer program stored in the memory and capable of running on the processor. The memory stores a computer program capable of running on the processor. When the processor loads and executes the computer program, the above-mentioned intelligent evaluation method for green space supplementary cost based on vacancy probability is adopted.

[0166] It should be noted that the terminal device can be a computer device such as a desktop computer, a laptop computer or a cloud server, and the terminal device includes but is not limited to a processor and a memory. For example, the terminal device can also include input and output devices, network access devices and buses.

[0167] Furthermore, the processor may adopt a central processing unit (CPU). Of course, according to actual usage, other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGAs) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. may also be adopted. The general-purpose processor may adopt a microprocessor or any conventional processor, etc., and the present application does not impose any restrictions on this.

[0168] Embodiment 4:

[0169] The present invention provides a storage medium containing computer executable instructions, which are used to execute the above-mentioned intelligent evaluation method of green space supplementary cost based on vacancy probability when executed by a computer processor.

[0170] Among them, the computer program can be stored in a computer-readable medium, the computer program includes computer program code, the computer program code can be in the form of source code, object code, executable file or certain middleware, etc. The computer-readable medium includes any entity or device that can carry the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signal, telecommunication signal and software distribution medium, etc. It should be noted that the computer-readable medium includes but is not limited to the above-mentioned components.

[0171] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0172] For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to the specific circumstances. When an element is referred to as being "assembled on", "installed on", "fixed on" or "set on" another element, it can be directly on the other element or there can also be a centered element. When an element is considered to be "connected" to another element, it can be directly connected to the other element or there may be a centered element at the same time. The terms "vertical", "horizontal", "up", "down", "left", "right" and similar expressions used herein are only for illustrative purposes and are not intended to be the only implementation method.

[0173] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

[0174] In the description of this specification, the description with reference to the terms "one embodiment", "example", "specific example", etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present disclosure. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.

Claims

1. An intelligent evaluation method for green space supplementary cost based on vacancy probability, characterized in that: The following steps are involved: Receive remote sensing images of unit land in the target area, obtain urban vacant land and inefficient construction land based on remote sensing image recognition, and summarize urban vacant land and inefficient construction land to obtain the overall vacant land identification result; Based on the overall vacant land identification results, the driving factors were calculated using the random forest regression algorithm. The relationship model between (x) and the expansion of vacant land is divided into the driving factor value Input into the finally constructed relationship model to obtain the probability of non-vacant land being converted into vacant land; Calculate the green space supplement cost of the unit land in the target area, where the green space supplement cost includes land acquisition cost, land remediation cost and green space construction optimization cost; The unit supplement potential is calculated based on the probability of conversion to vacant land and the green space supplement cost. The green space supplement priority of the unit land is determined based on the calculation result. The unit land is converted into green space in order from high to low priority. The form of converted green space is determined according to the unit land type. The supplement cost of the green space supplement plan is calculated. If the supplement cost is lower than the supplement budget, the plan will be included in the alternative supplement plan. Otherwise, the supplement plan will be modified until the supplement plan meets the supplement budget requirements.

2. The intelligent evaluation method for green space supplementary cost based on vacancy probability according to claim 1 is characterized by: The remote sensing images are from Google Maps, Tiandi Map or Gaofen series optical remote sensing satellites. The screening criteria for remote sensing images are that the cloud cover of the image is less than 10% and the spatial resolution is not less than 2.5m.

3. The intelligent evaluation method for green space supplementary cost based on vacancy probability according to claim 2 is characterized in that: Based on remote sensing image recognition, urban vacant land and inefficient construction land are obtained, and the overall vacant land identification results are obtained by summarizing urban vacant land and inefficient construction land, as follows: (31) Identification of urban vacant land based on remote sensing images, including: (31.1) Based on the definition of urban vacant land and the actual situation of the research site, a land use type system is constructed: The urban vacant land is divided into n types. According to the definition of land use type, the ArcGIS Pro classification tool is used to select training samples, and automatic classification is performed based on the samples. The accuracy of classification is evaluated using the accuracy assessment tool. The land use type system includes bare land, natural grassland, artificial grassland, natural shrubs, woodland, area under construction, residential land, commercial and office land, industrial land, hard square, and water body; (31.2) Use ArcGIS pro image classification tools to delineate the number of samples to 50 for each type, and evenly distribute the samples in the target area. Select support vector machine as the classifier, select the delineated samples as training samples, identify the land use types in remote sensing images, and use the accuracy assessment tool to evaluate the recognition results; (31.3) Input the study unit, which is the smallest grid unit for green space supplementation. The grid size is 10m-20m. The land use data is resampled to the grid size using the majority resampling method. (32) Identification of inefficient construction land based on multi-source data: (32.1) Based on the land use data of the research unit identified in (31), the construction land type units are extracted and an evaluation system for the inefficiency of each construction land type is constructed, including: (32.11) Extract construction land type units, including residential land units, commercial and office land units, and industrial land units; (32.12) Establish an evaluation system for the inefficiency of each type of construction land: For each type of construction land, set n evaluation indicators (X1, X2, ...X n ), for this type of land unit, the unit's inefficiency is the unit's evaluation index value (x1, x2, ...x n ), the specific expression is as follows: In the formula, C represents the inefficiency of the construction land type unit, and n represents the number of evaluation indicators; (32.2) After completing the inefficiency assessment of each type of construction land unit, the inefficiency is normalized and a certain proportion of land is selected as inefficient construction land; (33) Summary of vacant land identification results: The urban vacant land identified in (31) and the inefficient construction land identified in (32) are aggregated, and the distribution locations of urban vacant land and inefficient construction land are obtained respectively.

4. The intelligent evaluation method for green space supplementary cost based on vacancy probability according to claim 3 is characterized in that: In step (32), the evaluation indicators of residential land units include road intersection density, service facility density, and night lighting index; the evaluation indicators of commercial land units include popularity index and commercial facility density; and the evaluation indicators of industrial land units include building density and GDP.

5. The intelligent evaluation method for green space supplementary cost based on vacancy probability according to claim 4 is characterized in that: Based on the overall vacant land identification results, the driving factors were calculated using the random forest regression algorithm. The relationship model between the expansion of vacant land and the driving factor value Input into the finally constructed relationship model, and obtain the probability of non-vacant land being converted into vacant land, as follows: (51) The distribution locations of urban vacant land and inefficient construction land obtained in (33) are superimposed to obtain the area of ​​vacant land expansion. The relationship model between driving factors and vacant land expansion is constructed using the random forest regression algorithm, as follows: (51.1) For study unit i, the study unit with expanded vacant land is assigned a value of 1, and the rest of the study units are assigned a value of 0. Whether it is expanded is taken as the dependent variable, all driving factors are normalized, and the normalized driving factors are taken as the independent variables; (51.2) Aggregate the independent variables and dependent variables of all research units in the target area to form a training sample set, divide the training sample set into a training set and a test set, use the training set to train the random forest regression algorithm, and use the test set to evaluate the training results; (51.3) During the training process, the performance of the random forest regression algorithm was adjusted by adjusting the number of decision trees, the maximum depth of the decision tree, the minimum number of samples for node splitting, the minimum number of samples for leaf nodes, and the number of features for node splitting; The basic formula of the random forest regression algorithm is: Where X={X1,X2,…X n } is a feature vector containing n driving factors, M is the number of decision trees, T m (X) is the predicted value of the mth decision tree for the input feature vector X. Each decision tree T m (X) Processes the input features according to its own tree structure and splitting rules, and finally gives a prediction value; (51.4) After completing each round of training, the test set is used for evaluation. The explanatory power of the constructed regression model is evaluated by the mean square error, mean absolute error, and R square indicators. All models with R square greater than the threshold are output, and the model with the smallest mean square error and mean absolute error is selected as the final constructed relationship model; the mean square error is calculated as: Where y i is the actual value of the test set, is the value calculated based on the random forest regression model, and n is the number of samples in the test set; The mean absolute error (RMSE) is calculated as: In the formula, y i is the actual value of the test set, is the value calculated based on the random forest regression model, and n is the number of samples in the test set; The calculation method of R squared is: In the formula, y i is the actual value of the test set, is the value calculated based on the random forest regression model, and n is the number of samples in the test set; (52) Vacancy probability calculation: For each non-vacant land research unit, all driving factor values ​​X are input into the final constructed relationship model, which is the probability Pi of the research unit being transferred to vacant land: The formula for the relational model is: Where X={X1,X2,…X n } is a feature vector containing n driving factors, M is the number of decision trees in the final relationship model, T m (X) is the predicted value of the mth decision tree for the input feature vector X; For the research unit that is currently vacant land, the probability that it will be converted into vacant land in the future is 1.

6. The intelligent evaluation method for green space supplementary cost based on vacancy probability according to claim 5 is characterized in that: The additional cost of green space for unit land in the target area is calculated as follows: C=C1+C2+C3 In the formula, C represents the green space supplement cost, C1 represents the land acquisition cost, C2 represents the land remediation cost, and C3 represents the green space construction optimization cost.

7. The intelligent evaluation method for green space supplementary cost based on vacancy probability according to claim 6 is characterized in that: The unit supplement potential is calculated based on the probability of conversion to vacant land and the cost of green space supplement. The priority of green space supplement for unit land is determined based on the calculation results. A green space supplement plan is formulated based on the priority, and the supplement cost of the green space supplement plan is calculated as follows: (71) Unit supplement potential evaluation: Normalize the cost to get the final comprehensive cost C i ', the normalized formula is: In the formula, C i is the additional cost of the unit green space, C min is the minimum value of green space supplement cost in all units, C max is the maximum value of green space supplement cost in all units, C i ' is the normalized comprehensive cost; The unit supplementation potential calculation formula is: T i =P i *(1-C i ') (72) Unit Supplement Priority Setting Priorities are set based on the calculation result T of the unit supplementation potential. The higher the unit supplementation potential, the higher the priority of green space supplementation. The priorities are divided into 5 levels, from low to high: Very low [0, 0.2], low [0.2, 0.4], medium [0.4, 0.6], high [0.6, 0.8], very high [0.8, 1.0]; (73) Verification of green space supplement plan Formulate a green space supplement plan based on the unit green space supplement priority, convert the units into green spaces in order from high to low priority, and determine the form of the converted green space based on the land use type around the unit. After each green space supplement plan is completed, calculate the green supplement cost required for the plan and compare it with the supplement budget. If it is lower than the supplement budget requirement, include the plan in the alternative supplement plan. Otherwise, modify the supplement plan until the plan meets the supplement budget requirement.

8. A system for intelligently evaluating the supplementary cost of green space based on vacancy probability, used to implement the intelligent evaluation method for supplementary cost of green space based on vacancy probability as claimed in any one of claims 1 to 7, characterized in that: include: The vacant land identification module is used to receive remote sensing images of unit land in the target area, obtain urban vacant land and inefficient construction land based on remote sensing image recognition, and summarize the urban vacant land and inefficient construction land to obtain the overall vacant land identification result; The vacancy probability prediction module is used to calculate the driving factors based on the overall vacant land identification results using the random forest regression algorithm The relationship model between the expansion of vacant land and the driving factor value Input into the finally constructed relationship model to obtain the probability of non-vacant land being converted into vacant land; The additional cost calculation module is used to calculate the additional cost of green space in the target area unit land, where the additional cost of green space includes land acquisition cost, land remediation cost and green space construction optimization cost; The supplementary scheme derivation and verification module is used to calculate the unit supplementary potential according to the probability of conversion to vacant land and the green space supplementary cost, determine the unit land green space supplementary priority according to the calculation result, convert the unit land into green space in order from high to low priority, determine the converted green space form according to the unit land type, calculate the supplementary cost of the green space supplementary scheme, if the supplementary cost is lower than the supplementary budget, the scheme will be included in the alternative supplementary scheme, otherwise the supplementary scheme will be modified until the supplementary scheme meets the supplementary budget requirements.

9. A terminal device, comprising a memory, a processor, and a computer program stored in the memory and capable of running on the processor, characterized in that: The memory stores a computer program that can be run on the processor. When the processor loads and executes the computer program, the intelligent evaluation method for green space supplementary cost based on vacancy probability described in any one of claims 1 to 7 is adopted.

10. A storage medium containing computer executable instructions, characterized in that: When the computer executable instructions are executed by a computer processor, they are used to perform the intelligent evaluation method for green space supplementary cost based on vacancy probability according to any one of claims 1 to 7.

Images