Method and system for estimating carbon sink increase in artificial grass based on artificial intelligence

Abstract

The invention discloses an artificial intelligence-based estimation method and system for carbon sink increase in artificial grass, and the method comprises the steps: collecting remote sensing data and survey data of a preset region, and carrying out the preprocessing of the remote sensing data and the survey data; performing regional division on the remote sensing data according to biomass to obtain regional data, and obtaining proportion data through the regional data and the biomass; performing time-varying influence analysis on the survey data to obtain a time-varying factor, and constructing an artificial grass planting carbon sink increase estimation model according to the proportion data and the time-varying factor; and according to the estimation error, optimizing the estimation model for increasing the carbon sink in the artificial grass, inputting data to be estimated into the estimation model for increasing the carbon sink in the artificial grass, and outputting an estimation result. The method not only can improve the estimation precision of the carbon sink increase in artificial grass based on artificial intelligence, but also has good interpretability, and can be directly applied to an estimation system of the carbon sink increase in artificial grass.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence, and in particular to an estimation method and system for increasing carbon sinks by artificial grass planting based on artificial intelligence. Background Art

[0002] As global climate change becomes increasingly severe, carbon sink management has become one of the important means to mitigate greenhouse gas emissions and respond to climate change. Artificial grass planting, as an effective way to increase carbon sinks, can significantly increase soil carbon storage capacity and reduce atmospheric carbon dioxide concentrations by planting herbaceous plants with high carbon fixation capacity. However, traditional artificial grass planting carbon sink estimation methods often rely on empirical formulas or simple statistical models, which are difficult to accurately reflect the complex and changing ecological environment and grass planting effects.

[0003] In recent years, with the rapid development of remote sensing technology and artificial intelligence technology, new technical means have been provided for carbon sink estimation. Remote sensing technology can efficiently obtain a wide range of surface information, including vegetation cover, soil type, etc., providing a rich data source for carbon sink estimation. Artificial intelligence technology, especially machine learning and deep learning algorithms, can process and analyze large-scale data, explore complex relationships between data, and improve the accuracy and reliability of carbon sink estimation.

[0004] However, there is no mature artificial intelligence-based estimation method for increasing carbon sinks by artificial grass planting. Therefore, the present invention proposes a new estimation method, which aims to combine remote sensing data and survey data and use artificial intelligence technology to achieve accurate estimation of the amount of carbon sinks increased by artificial grass planting. Summary of the invention

[0005] The purpose of the present invention is to provide an estimation method for increasing carbon sinks by artificial grass planting based on artificial intelligence.

[0006] To achieve the above object, the present invention is implemented according to the following technical solutions:

[0007] The present invention comprises the following steps:

[0008] Collecting remote sensing data and survey data of a preset area, and preprocessing the remote sensing data and the survey data;

[0009] Divide the remote sensing data into regions according to the biomass to obtain regional data, and obtain proportional data through the regional data and the biomass;

[0010] Conducting a time-varying impact analysis on the survey data to obtain time-varying factors, and constructing an estimation model for increasing carbon sinks by planting artificial grass based on the proportion data and the time-varying factors;

[0011] The artificial grass planting carbon sink estimation model is optimized according to the estimation error, the data to be estimated is input into the artificial grass planting carbon sink estimation model, and the estimation result is output.

[0012] Furthermore, the method of obtaining regional data by dividing the remote sensing data into regions according to biomass includes:

[0013] The remote sensing data is input into the artificial neural network model, the remote sensing data is divided into multiple pixels, and the biomass is calculated:

[0014]

[0015] The pixel to be tested is z, the ιth known pixel closest to the pixel to be tested in the spectral space is zι, and the spectral distance between the pixel to be tested and the known pixel is S zι,z , the number of pixels is n, the error term is υ, and the regression coefficients to be determined are d o , d1, …, d ι ,d n , the information parameters are k1, k n , k ι , the biomass of the pixel to be measured is h z ;

[0016] The biomass of the pixels to be measured is input into the convolutional neural network classification model according to the region, and the convolution layer output is calculated:

[0017]

[0018] The convolutional layer output of the qth region is The activation function is The output bias is Q, the convolution kernel size is H, the classification boundary vector is ρ, and the penalty coefficient is The numerical error of the qth region feature is σ q , the diagonal matrix is ​​W, the biomass correlation matrix in the qth region is K, transposed to T, The biomass correlation matrix of different regions is: forward The biomass correlation matrix of different regions is as follows: forward The column diagonal matrix is The inverse matrix of the diagonal matrix is ​​W -1 , the number of pixels to be tested in the qth area is

[0019] Calculate the maximum pooling function value:

[0020]

[0021] The maximum pooling function value of the qth region is The moving length is ω and the pooling size is f;

[0022] Get a set of neighboring regions, classify them consistently based on biomass, and calculate the consistency between regions:

[0023]

[0024] The number of regions is The output of the fully connected layer of the qth region is g q , the adjustment coefficient is δ, and the category of the ath area is ε q , the pixel to be tested in the qth area is q z , the number of pixels to be tested in the qth area is No. The pixels to be tested in the region are No. The number of pixels to be tested in the region is The qth region and the The consistency of the region is Pixel to be measured q z The biomass is Pixel to be tested The biomass is The average biomass of the qth region is No. The average biomass of the area is

[0025] The target area and the neighboring areas with consistency greater than 0.759 are merged, and the merged results are output as regional data.

[0026] Furthermore, the method for obtaining proportion data through the area data and the biomass includes:

[0027] Calculate proportion data based on area data and biomass:

[0028]

[0029] Among them, the sth moment The proportion of regional data is The sth moment The biomass of the regional data is The upper limit of observation time for regional data is M s , the number of pixels in the regional data is The weight coefficient of regional data is The number of regional data is The offset coefficient at time s is θ s .

[0030] Furthermore, the method of performing time-varying impact analysis on the survey data to obtain time-varying factors includes:

[0031] The survey data is input into the time-varying impact analysis model to extract the time pattern information of the survey data. The expression is:

[0032]

[0033] The number of sliding windows y is N, and the temporal pattern information of the survey data in the ath dimension on the jth convolution kernel C is C a,j , time is s, time limit is T s , the time pattern information of the survey data of the ath dimension on the sb-1+yth convolution kernel is K a,(s-b-1+y) , the jth convolution kernel in the T-b+y dimension is R j,T-b+y ;

[0034] The temporal pattern information is fed into the attention mechanism to calculate the correlation between the survey data and the normalized vegetation index:

[0035]

[0036] The normalized vegetation index at the sth moment is The weight matrix is ​​D, the survey data x and the normalized vegetation index The correlation is The survey data at the sth moment is x s The average value of the survey data is The average value of the Normalized Difference Vegetation Index is

[0037] Calculate the contribution of survey data:

[0038]

[0039] The contribution of survey data x is The normalized vegetation index after the survey data x changes is The starting value of the normalized index is The cth survey data x c The probability of occurrence is p(x c ), the number of survey data is N c , the number of survey data that changed is N ce ;

[0040] The survey data with contribution greater than 0.391 are used as research data, and the time-varying factor is calculated based on the contribution and research data:

[0041]

[0042] The adjustment constant is δ, the control coefficient is τ, the genetic factor is μ, the L2 norm is ‖·‖2, the number of research data is N1, the error constant is σ, and the time-varying factor at the sth moment is

[0043] Furthermore, a method for constructing an estimation model for increasing carbon sinks by planting artificial grass according to the proportion data and the time-varying factors includes:

[0044] The objective function is constructed based on the proportional data and time-varying factors, and the expression is:

[0045]

[0046] The objective function at the sth moment is The number of regional data is The number of regional data is The biomass of the target area at time s is h s , at the sth moment The proportion of regional data is The sth moment The time-varying factor of the regional data is The actual carbon sink of artificial grass planting is The loss function is

[0047] The estimation model of artificial grass planting to increase carbon sink includes support vector machine algorithm, decision tree algorithm, and graph attention network algorithm;

[0048] The support vector machine algorithm searches for the best segmentation hyperplane in the feature space of the input data, and then performs vegetation recognition based on the segmentation hyperplane to obtain artificial grass planting data.

[0049] The decision tree algorithm calculates the comprehensive weighted score of the relevance importance of artificial grass planting data, selects artificial grass planting data according to the comprehensive weighted score of the relevance importance, and obtains key data;

[0050] The graph attention network algorithm constructs graph structure data based on key data to represent the grass planting area and its related factors, and uses the graph neural network to learn the characteristics of nodes and edges to estimate the increase in carbon sinks caused by artificial grass planting.

[0051] Furthermore, the method for optimizing the artificial grass planting carbon sink estimation model according to the estimation error includes:

[0052] Introduce a particle population, calculate the fitness of the particles, and take the position of the particle with the smallest fitness value as the optimal position;

[0053] Calculate the position of the particle:

[0054] Z w (t+1)=Z best(t)-β·(2r1-1)·|2r2·Z best (t)-Z w (t)|

[0055] The convergence coefficient is β, the random numbers from 0 to 1 are r1 and r2, and the best position of the t+1th iteration is Z best (t), the position of the wth particle in the t+1th iteration is Z w (t+1), the position of the wth particle in the tth iteration is Z w (t);

[0056] Update the particle position to obtain the search position. The expression is:

[0057]

[0058] The search position of the wth particle in the t+1th iteration is The spiral constant is v, and the random number between -1 and 1 is

[0059] Update the convergence coefficient, the expression is:

[0060]

[0061] The updated convergence coefficient is The maximum number of iterations is t max , the initial value of the convergence coefficient is β o , the final value of the convergence coefficient is β fn , the adjustment coefficient is α;

[0062] Introduce random step size, use random step size to update particle position to obtain step size position, the expression is:

[0063]

[0064] The gamma function is Γ(·), the constant is ω, the random step length is η, the random variables that obey the normal distribution are ε and λ, and the step length position of the wth particle in the t+1th iteration is The search position of the wth particle in the tth iteration is

[0065] Update the particle position, the expression is:

[0066]

[0067] The random numbers from 0 to 1 are r3 and r4 respectively. The updated position of the wth particle in the tth iteration is The worst position of the wth particle is Z wt ;

[0068] Continue to iterate until the maximum number of iterations is reached, otherwise update the convergence coefficient.

[0069] Second, an artificial intelligence-based estimation system for increasing carbon sinks through artificial grass planting includes:

[0070] Data acquisition module: used to collect remote sensing data and survey data of a preset area, and pre-process the remote sensing data and the survey data;

[0071] A division ratio module: used for dividing the remote sensing data into regions according to the biomass to obtain regional data, and obtaining ratio data through the regional data and the biomass;

[0072] Impact construction module: used to perform time-varying impact analysis on the survey data to obtain time-varying factors, and to construct an estimation model for increasing carbon sinks by planting artificial grass according to the proportion data and the time-varying factors;

[0073] Optimization output module: optimizes the artificial grass planting carbon sink estimation model according to the estimation error, inputs the data to be estimated into the artificial grass planting carbon sink estimation model, and outputs the estimation result.

[0074] In a third aspect, an embodiment of the present application further provides an electronic device, including:

[0075] A processor; and a memory arranged to store computer executable instructions, which when executed cause the processor to perform the method steps described in the first aspect.

[0076] In a fourth aspect, an embodiment of the present application further provides a computer-readable storage medium, which stores one or more programs. When the one or more programs are executed by an electronic device including multiple applications, the electronic device executes the method steps described in the first aspect.

[0077] The beneficial effects of the present invention are:

[0078] The present invention is an artificial grass planting carbon sink estimation method and system based on artificial intelligence. Compared with the prior art, the present invention has the following technical effects:

[0079] The present invention can improve the accuracy of the estimation of the increase in carbon sinks by artificial grass planting based on artificial intelligence through preprocessing, regional division, carbon sink prediction, deviation analysis, time-varying impact analysis, model construction and model optimization steps, thereby improving the accuracy of the estimation of the increase in carbon sinks by artificial grass planting based on artificial intelligence, and optimizing the estimation of the increase in carbon sinks by artificial grass planting based on artificial intelligence can greatly save resources and improve work efficiency, and can realize intelligent estimation of the increase in carbon sinks by artificial grass planting based on artificial intelligence, and perform carbon sink correction on the estimation of the increase in carbon sinks by artificial grass planting based on artificial intelligence in real time, which is of great significance to the estimation of the increase in carbon sinks by artificial grass planting based on artificial intelligence, can adapt to the estimation of the increase in carbon sinks by artificial grass planting of different standards and the estimation needs of different artificial grass planting, and has a certain universality. BRIEF DESCRIPTION OF THE DRAWINGS

[0080] Figure 1 A flowchart of the steps of an artificial grass planting carbon sink estimation method based on artificial intelligence of the present invention;

[0081] Figure 2 It is a schematic diagram of the structure of an electronic device in an embodiment of this specification. DETAILED DESCRIPTION

[0082] The present invention is further described below by means of specific embodiments. The illustrative embodiments and descriptions of the present invention are used to explain the present invention but are not intended to limit the present invention.

[0083] The present invention provides an artificial intelligence-based method and system for estimating carbon sinks by artificial grass planting, which comprises the following steps:

[0084] like Figure 1 As shown, in this embodiment, the following steps are included:

[0085] Collecting remote sensing data and survey data of a preset area, and preprocessing the remote sensing data and the survey data;

[0086] In the actual evaluation, the grassland area in the XXX region was selected as the research area; the remote sensing data were satellite remote sensing images, including multispectral and hyperspectral images, with a time span of one year and a resolution of 10 meters; the survey data included recording vegetation type, height, coverage, soil, biomass data, environmental data, and meteorological data; the preprocessing included radiation correction, geometric correction, and atmospheric correction;

[0087] Divide the remote sensing data into regions according to the biomass to obtain regional data, and obtain proportional data through the regional data and the biomass;

[0088] In the actual evaluation, the remote sensing data were divided into high biomass area, medium biomass area and low biomass area. The biomass in the high biomass area, medium biomass area and low biomass area were greater than 830 g / m 2 , greater than 285g / m 2 Less than 830g / m 2 Less than 285g / m 2 ; The proportions of high biomass area, medium biomass area and low biomass area are 44%, 32% and 24% respectively;

[0089] Conducting a time-varying impact analysis on the survey data to obtain time-varying factors, and constructing an estimation model for increasing carbon sinks by planting artificial grass based on the proportion data and the time-varying factors;

[0090] In the actual evaluation, the time-varying factor is 0.316;

[0091] Optimizing the artificial grass planting carbon sink estimation model according to the estimation error, inputting the data to be estimated into the artificial grass planting carbon sink estimation model, and outputting the estimation result;

[0092] In the actual assessment, artificial grass planting increased carbon sequestration by 49.37 tons per hectare.

[0093] In this embodiment, the method of obtaining regional data by dividing the remote sensing data into regions according to biomass includes:

[0094] The remote sensing data is input into the artificial neural network model, the remote sensing data is divided into multiple pixels, and the biomass is calculated:

[0095]

[0096] The pixel to be tested is z, the ιth known pixel closest to the pixel to be tested in the spectral space is zι, and the spectral distance between the pixel to be tested and the known pixel is S zι,z , the number of pixels is n, the error term is υ, and the regression coefficients to be determined are d o , d1, …, d ι d n , the information parameters are k1, k n , k ι , the biomass of the pixel to be measured is h z ;

[0097] The biomass of the pixels to be measured is input into the convolutional neural network classification model according to the region, and the convolution layer output is calculated:

[0098]

[0099] The convolutional layer output of the qth region is The activation function is The output bias is Q, the convolution kernel size is H, the classification boundary vector is ρ, and the penalty coefficient is The numerical error of the qth region feature is σ q , the diagonal matrix is ​​W, the biomass correlation matrix in the qth region is K, the transpose is T, the front The biomass correlation matrix of different regions is: forward The biomass correlation matrix of different regions is as follows: forward The column diagonal matrix is The inverse matrix of the diagonal matrix is ​​W -1 , the number of pixels to be tested in the qth area is

[0100] Calculate the maximum pooling function value:

[0101]

[0102] The maximum pooling function value of the qth region is The moving length is ω and the pooling size is f;

[0103] Get a set of neighboring regions, classify them consistently based on biomass, and calculate the consistency between regions:

[0104]

[0105] The number of regions is The output of the fully connected layer of the qth region is g q , the adjustment coefficient is δ, and the category of the ath area is ε q , the pixel to be tested in the qth area is q z , the number of pixels to be tested in the qth area is No. The pixels to be tested in the region are No. The number of pixels to be tested in the region is The qth region and the The consistency of the region is Pixel to be measured q z The biomass is Pixel to be tested The biomass is The average biomass of the qth region is No. The average biomass of the area is

[0106] The target area and the neighboring areas with consistency greater than 0.759 are merged, and the merged results are output as regional data.

[0107] In this embodiment, the method of obtaining proportion data through the area data and the biomass includes:

[0108] Calculate proportion data based on area data and biomass:

[0109]

[0110] Among them, the sth moment The proportion of regional data is The sth moment The biomass of the regional data is The upper limit of observation time for regional data is M s , the number of pixels in the regional data is The weight coefficient of regional data is The number of regional data is The offset coefficient at time s is θ s .

[0111] In this embodiment, the method of performing time-varying impact analysis on the survey data to obtain the time-varying factor includes:

[0112] The survey data is input into the time-varying impact analysis model to extract the time pattern information of the survey data. The expression is:

[0113]

[0114] The number of sliding windows y is N, and the temporal pattern information of the survey data in the ath dimension on the jth convolution kernel C is C a,j , time is s, time limit is T s , the time pattern information of the survey data in the ath dimension on the sb-1+yth convolution kernel is K a,(s-b-1+y) , the jth convolution kernel in the T-b+y dimension is R j,T-b+y ;

[0115] The temporal pattern information is fed into the attention mechanism to calculate the correlation between the survey data and the normalized vegetation index:

[0116]

[0117] The normalized vegetation index at the sth moment is The weight matrix is ​​D, the survey data x and the normalized vegetation index The correlation is The survey data at the sth moment is x s The average value of the survey data is The average value of the Normalized Difference Vegetation Index is

[0118] Calculate the contribution of survey data:

[0119]

[0120] The contribution of survey data x is The normalized vegetation index after the survey data x changes is The starting value of the normalized index is The cth survey data x c The probability of occurrence is p(x c ), the number of survey data is N c , the number of survey data that changed is N ce ;

[0121] The survey data with contribution greater than 0.391 are used as research data, and the time-varying factor is calculated based on the contribution and research data:

[0122]

[0123] The adjustment constant is δ, the control coefficient is τ, the genetic factor is μ, the L2 norm is ‖·‖2, the number of research data is N1, the error constant is σ, and the time-varying factor at the sth moment is

[0124] In this embodiment, the method for constructing an estimation model for increasing carbon sinks by planting artificial grass according to the proportion data and the time-varying factors includes:

[0125] The objective function is constructed based on the proportional data and time-varying factors, and the expression is:

[0126]

[0127] The objective function at the sth moment is The number of regional data is The number of regional data is The biomass of the target area at time s is h s , at the sth moment The proportion of regional data is The sth moment The time-varying factor of the regional data is The actual carbon sink of artificial grass planting is The loss function is

[0128] The estimation model of artificial grass planting to increase carbon sink includes support vector machine algorithm, decision tree algorithm, and graph attention network algorithm;

[0129] The support vector machine algorithm searches for the best segmentation hyperplane in the feature space of the input data, and then performs vegetation recognition based on the segmentation hyperplane to obtain artificial grass planting data.

[0130] The decision tree algorithm calculates the comprehensive weighted score of the relevance importance of artificial grass planting data, selects artificial grass planting data according to the comprehensive weighted score of the relevance importance, and obtains key data;

[0131] The graph attention network algorithm constructs graph structure data based on key data to represent the grass planting area and its related factors, and uses the graph neural network to learn the characteristics of nodes and edges to estimate the increase in carbon sinks caused by artificial grass planting.

[0132] In this embodiment, the method for optimizing the artificial grass planting carbon sink estimation model according to the estimation error includes:

[0133] Introduce a particle population, calculate the fitness of the particles, and take the position of the particle with the smallest fitness value as the optimal position;

[0134] Calculate the position of the particle:

[0135] Z w (t+1)=Z best (t)-β·(2r1-1)·|2r2·Z best (t)-Z w (t)|

[0136] The convergence coefficient is β, the random numbers from 0 to 1 are r1 and r2, and the best position of the t+1th iteration is Z best (t), the position of the wth particle in the t+1th iteration is Z w (t+1), the position of the wth particle in the tth iteration is Z w (t);

[0137] Update the particle position to obtain the search position. The expression is:

[0138]

[0139] The search position of the wth particle in the t+1th iteration is The spiral constant is v, and the random number between -1 and 1 is

[0140] Update the convergence coefficient, the expression is:

[0141]

[0142] The updated convergence coefficient is The maximum number of iterations is t max , the initial value of the convergence coefficient is β o , the final value of the convergence coefficient is βfn , the adjustment coefficient is α;

[0143] Introduce random step size, use random step size to update particle position to obtain step size position, the expression is:

[0144]

[0145] The gamma function is Γ(·), the constant is ω, the random step length is η, the random variables that obey the normal distribution are ε and λ, and the step length position of the wth particle in the t+1th iteration is The search position of the wth particle in the tth iteration is

[0146] Update the particle position, the expression is:

[0147]

[0148] The random numbers from 0 to 1 are r3 and r4 respectively. The updated position of the wth particle in the tth iteration is The worst position of the wth particle is Z wt ;

[0149] Continue to iterate until the maximum number of iterations is reached, otherwise update the convergence coefficient.

[0150] Second, an artificial intelligence-based estimation system for increasing carbon sinks through artificial grass planting includes:

[0151] Data acquisition module: used to collect remote sensing data and survey data of a preset area, and pre-process the remote sensing data and the survey data;

[0152] A division ratio module: used for dividing the remote sensing data into regions according to the biomass to obtain regional data, and obtaining ratio data through the regional data and the biomass;

[0153] Impact construction module: used to perform time-varying impact analysis on the survey data to obtain time-varying factors, and to construct an estimation model for increasing carbon sinks by planting artificial grass according to the proportion data and the time-varying factors;

[0154] Optimization output module: optimizes the artificial grass planting carbon sink estimation model according to the estimation error, inputs the data to be estimated into the artificial grass planting carbon sink estimation model, and outputs the estimation result.

[0155] Figure 2 This is a schematic diagram of the structure of an electronic device according to an embodiment of the present application. Figure 2At the hardware level, the electronic device includes a processor, and optionally also includes an internal bus, a network interface, and a memory. The memory may include a memory, such as a high-speed random access memory (RAM), and may also include a non-volatile memory (non-volatile memory), such as at least one disk storage. Of course, the electronic device may also include hardware required for other services.

[0156] The processor, network interface and memory can be interconnected through an internal bus, which can be an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus. The bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 2 Only one bidirectional arrow is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0157] The memory is used to store the program. Specifically, the program may include a program code, and the program code includes a computer operation instruction. The memory may include a memory and a non-volatile memory, and provides instructions and data to the processor.

[0158] The processor reads the corresponding computer program from the non-volatile memory into the memory and then runs it, forming an estimation device for increasing carbon sinks by artificial grass planting based on artificial intelligence at the logical level. The processor executes the program stored in the memory and is specifically used to execute any of the aforementioned estimation methods for increasing carbon sinks by artificial grass planting based on artificial intelligence.

[0159] The above application Figure 1The embodiment shown discloses an artificial intelligence-based artificial grass planting carbon sink estimation method that can be applied to a processor or implemented by a processor. The processor may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method can be completed by an integrated logic circuit of hardware in the processor or an instruction in software form. The above processor may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it may also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The methods, steps and logic block diagrams disclosed in the embodiments of the present application can be implemented or executed. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc. The steps of the method disclosed in conjunction with the embodiments of the present application can be directly embodied as being executed by a hardware decoding processor, or executed by a combination of hardware and software modules in a decoding processor. The software module can be located in a storage medium mature in the art such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory, or an electrically erasable programmable memory, a register, etc. The storage medium is located in the memory, and the processor reads the information in the memory and completes the steps of the above method in combination with its hardware.

[0160] The electronic device may also perform Figure 1 An artificial intelligence-based estimation method for increasing carbon sinks by planting artificial grass Figure 1 The functions of the illustrated embodiment will not be described in detail in the embodiments of the present application.

[0161] An embodiment of the present application also proposes a computer-readable storage medium, which stores one or more programs, and the one or more programs include instructions. When the instructions are executed by an electronic device including multiple application programs, any one of the aforementioned artificial intelligence-based artificial grass planting to increase carbon sink estimation methods is executed.

[0162] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.

[0163] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0164] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0165] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0166] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0167] The memory may include non-permanent storage in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0168] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0169] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity 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, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0170] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0171] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. An artificial intelligence-based method for estimating carbon sinks increased by artificial grass planting, characterized in that: The following steps are involved: Collecting remote sensing data and survey data of a preset area, and preprocessing the remote sensing data and the survey data; Divide the remote sensing data into regions according to the biomass to obtain regional data, and obtain proportional data through the regional data and the biomass; Conducting a time-varying impact analysis on the survey data to obtain time-varying factors, and constructing an estimation model for increasing carbon sinks by planting artificial grass based on the proportion data and the time-varying factors; The artificial grass planting carbon sink estimation model is optimized according to the estimation error, the data to be estimated is input into the artificial grass planting carbon sink estimation model, and the estimation result is output.

2. The artificial intelligence-based method for estimating carbon sinks increased by artificial grass planting according to claim 1 is characterized in that: The method for obtaining regional data by dividing the remote sensing data into regions according to biomass comprises: The remote sensing data is input into the artificial neural network model, the remote sensing data is divided into multiple pixels, and the biomass is calculated: The pixel to be tested is z, the ιth known pixel closest to the pixel to be tested in the spectral space is zι, and the spectral distance between the pixel to be tested and the known pixel is S zι,z , the number of pixels is n, the error term is υ, and the regression coefficients to be determined are d o , d1, …, d ι ,d n , the information parameters are k1, k n , k ι , the biomass of the pixel to be measured is h z ; The biomass of the pixels to be measured is input into the convolutional neural network classification model according to the region, and the convolution layer output is calculated: The convolutional layer output of the qth region is The activation function is The output bias is Q, the convolution kernel size is H, the classification boundary vector is ρ, and the penalty coefficient is The numerical error of the qth region feature is σ q , the diagonal matrix is ​​W, the biomass correlation matrix in the qth region is K, transposed to T, The biomass correlation matrix of different regions is: forward The biomass correlation matrix of different regions is as follows: forward The column diagonal matrix is The inverse matrix of the diagonal matrix is ​​W -1 , the number of pixels to be tested in the qth area is Calculate the maximum pooling function value: The maximum pooling function value of the qth region is The moving length is ω and the pooling size is f; Get a set of neighboring regions, classify them consistently based on biomass, and calculate the consistency between regions: The number of regions is The output of the fully connected layer of the qth region is g q , the adjustment coefficient is δ, and the category of the ath area is ε q , the pixel to be tested in the qth area is q z , the number of pixels to be tested in the qth area is No. The pixels to be tested in the region are No. The number of pixels to be tested in the region is The qth region and the The consistency of the region is Pixel to be measured q z The biomass is Pixel to be tested The biomass is The average biomass of the qth region is No. The average biomass of the area is The target area and the neighboring areas with consistency greater than 0.759 are merged, and the merged results are output as regional data.

3. The method for estimating carbon sinks increased by artificial grass planting based on artificial intelligence according to claim 1, characterized in that: The method for obtaining proportion data through the area data and the biomass comprises: Calculate proportion data based on area data and biomass: Among them, the sth moment The proportion of regional data is The sth moment The biomass of the regional data is The upper limit of observation time of regional data is M s , the number of pixels in the regional data is The weight coefficient of regional data is The number of regional data is The offset coefficient at time s is θ s .

4. The method for estimating carbon sinks increased by artificial grass planting based on artificial intelligence according to claim 1, characterized in that: The method of performing time-varying impact analysis on the survey data to obtain time-varying factors includes: The survey data is input into the time-varying impact analysis model to extract the time pattern information of the survey data. The expression is: The number of sliding windows y is N, and the temporal pattern information of the survey data in the ath dimension on the jth convolution kernel C is C a,j , time is s, time limit is T s , the time pattern information of the survey data of the ath dimension on the sb-1+yth convolution kernel is K a,(s-b-1+y) , the jth convolution kernel in the T-b+y dimension is R j,T-b+y ; The temporal pattern information is fed into the attention mechanism to calculate the correlation between the survey data and the normalized vegetation index: The normalized vegetation index at the sth moment is The weight matrix is ​​D, the survey data x and the normalized vegetation index The correlation is The survey data at the sth moment is x s The average value of the survey data is The average value of the Normalized Difference Vegetation Index is Calculate the contribution of survey data: The contribution of survey data x is l x , the normalized vegetation index after the survey data x changes is The starting value of the normalized index is The cth survey data x c The probability of occurrence is p(x c ), the number of survey data is N c , the number of survey data that changed is N ce ; The survey data with contribution greater than 0.391 are used as research data, and the time-varying factor is calculated based on the contribution and research data: The adjustment constant is δ, the control coefficient is τ, the genetic factor is μ, the L2 norm is ‖·‖2, the number of research data is N1, the error constant is σ, and the time-varying factor at the sth moment is 5. The method for estimating carbon sinks increased by artificial grass planting based on artificial intelligence according to claim 1, characterized in that: The method for constructing an estimation model for increasing carbon sinks by planting artificial grass according to the proportion data and the time-varying factors includes: The objective function is constructed based on the proportional data and time-varying factors, and the expression is: The objective function at the sth moment is The number of regional data is The number of regional data is The biomass of the target area at time s is h s , at the sth moment The proportion of regional data is The sth moment The time-varying factor of the regional data is The actual carbon sink of artificial grass planting is The loss function is The estimation model of artificial grass planting to increase carbon sink includes support vector machine algorithm, decision tree algorithm, and graph attention network algorithm; The support vector machine algorithm searches for the best segmentation hyperplane in the feature space of the input data, and then performs vegetation recognition based on the segmentation hyperplane to obtain artificial grass planting data. The decision tree algorithm calculates the comprehensive weighted score of the relevance importance of artificial grass planting data, selects artificial grass planting data according to the comprehensive weighted score of the relevance importance, and obtains key data; The graph attention network algorithm constructs graph structure data based on key data to represent the grass planting area and its related factors, and uses the graph neural network to learn the characteristics of nodes and edges to estimate the increase in carbon sinks caused by artificial grass planting.

6. The artificial grass planting carbon sink estimation method based on artificial intelligence according to claim 1 is characterized in that: The method for optimizing the artificial grass planting carbon sink estimation model according to the estimation error includes: Introduce a particle population, calculate the fitness of the particles, and take the position of the particle with the smallest fitness value as the optimal position; Calculate the position of the particle: Z w (t+1)=Z best (t)-β·(2r1-1)·|2r2·Z best (t)-Z w (t)| The convergence coefficient is β, the random numbers from 0 to 1 are r1 and r2, and the best position of the t+1th iteration is Z best (t), the position of the wth particle in the t+1th iteration is Z w (t+1), the position of the wth particle in the tth iteration is Z w (t); Update the particle position to obtain the search position. The expression is: The search position of the wth particle in the t+1th iteration is The spiral constant is v, and the random number between -1 and 1 is Update the convergence coefficient, the expression is: The updated convergence coefficient is The maximum number of iterations is t max , the initial value of the convergence coefficient is β o , the final value of the convergence coefficient is β fn , the adjustment coefficient is α; Introduce random step size, use random step size to update particle position to obtain step size position, the expression is: The gamma function is Γ(·), the constant is ω, the random step length is η, the random variables that obey the normal distribution are ε and λ, and the step length position of the wth particle in the t+1th iteration is The search position of the wth particle in the tth iteration is Update the particle position, the expression is: The random numbers from 0 to 1 are r3 and r4 respectively. The updated position of the wth particle in the tth iteration is The worst position of the wth particle is Z wt ; Continue to iterate until the maximum number of iterations is reached, otherwise update the convergence coefficient.

7. An artificial intelligence-based estimation system for increasing carbon sinks by planting artificial grass, used to execute the method described in any one of claims 1 to 6, characterized in that: include: Data acquisition module: used to collect remote sensing data and survey data of a preset area, and pre-process the remote sensing data and the survey data; A division ratio module: used for dividing the remote sensing data into regions according to the biomass to obtain regional data, and obtaining ratio data through the regional data and the biomass; Impact construction module: used to perform time-varying impact analysis on the survey data to obtain time-varying factors, and to construct an estimation model for increasing carbon sinks by planting artificial grass according to the proportion data and the time-varying factors; Optimization output module: optimizes the artificial grass planting carbon sink estimation model according to the estimation error, inputs the data to be estimated into the artificial grass planting carbon sink estimation model, and outputs the estimation result.

8. An electronic device comprising: processor; as well as A memory arranged to store computer executable instructions, which when executed cause the processor to perform the method according to any one of claims 1 to 6.

9. A computer-readable storage medium storing one or more programs, which, when executed by an electronic device including a plurality of application programs, causes the electronic device to execute the method according to any one of claims 1 to 6.

Images