Remote sensing-based soil quality analysis method, device and electronic equipment

Abstract

The application provides a soil quality analysis method and device based on remote sensing and electronic equipment, and the method comprises the following steps: acquiring satellite remote sensing data and DEM elevation data; acquiring a land structure image based on the satellite remote sensing data; segmenting and identifying the land structure image based on image processing technology to determine the land type of different land structure regions; identifying the soil level of each land type based on the DEM elevation data to obtain the soil level depth; determining the soil influence parameter based on the remote sensing data, and obtaining the soil quality parameter based on the soil influence parameter, the land type and the soil level depth; and determining the soil quality analysis result based on the soil quality parameter. The application can realize the comprehensiveness and accuracy of soil quality evaluation.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of remote sensing image processing, and in particular to a soil quality analysis method and device based on remote sensing and an electronic device. BACKGROUND

[0002] At present, most of the technical solutions for soil analysis using remote sensing data only focus on single soil parameter estimation, especially water content, and do not comprehensively analyze different land structures, land types, and soil layer depths. In addition, the existing technology is still limited by spectral data, and lacks comprehensive evaluation and grade division of soil quality.

[0003] Therefore, how to realize the comprehensiveness and accuracy of soil quality evaluation is a technical problem to be solved at present. SUMMARY

[0004] The present application provides a remote sensing image processing technical field, in particular to a soil quality analysis method and device based on remote sensing and an electronic device, to solve the above-mentioned defects in the prior art, and realize the comprehensiveness and accuracy of soil quality evaluation.

[0005] The present application provides a soil quality analysis method based on remote sensing, comprising:

[0006] Obtaining satellite remote sensing data and DEM elevation data;

[0007] Obtaining a land structure image based on the satellite remote sensing data;

[0008] Segmenting and identifying the land structure image based on image processing technology to determine the land type of different land structure regions;

[0009] Identifying the soil level of each land type based on the DEM elevation data to obtain the soil level depth;

[0010] Determining the soil influence parameter based on the remote sensing data, and obtaining the soil quality parameter based on the soil influence parameter, land type, and soil level depth;

[0011] Determining the soil quality analysis result based on the soil quality parameter.

[0012] According to the soil quality analysis method based on remote sensing provided by the present application, the land structure image is segmented and identified based on image processing technology to determine the land type of different land structure regions, which comprises:

[0013] Adjusting the image size, balancing the gray value, and detecting the edge of the land structure image to obtain a preprocessed image;

[0014] The preprocessed image is convolutionally, pooled, and upsampled based on the first convolutional neural network model to obtain the image features of the preprocessed image, and different land structures are determined based on the image features.

[0015] The land types of the different land structure regions are obtained by identifying each land structure based on the second convolutional neural network model.

[0016] According to the remote sensing-based soil quality analysis method provided by the present invention, the soil influence parameters include at least one of NDVI value, elevation, surface temperature, and normalized water index.

[0017] According to a remote sensing-based soil quality analysis method provided by the present invention, the step of identifying soil layers for each land type based on the DEM elevation data to obtain soil layer depth includes:

[0018] The initial thickness and spatial coordinates of the land type are determined based on the DEM elevation data.

[0019] The soil layer depth is obtained using an expansion model based on the initial thickness, spatial coordinates, and calculation parameters, as shown in the following formula:

[0020] H (x,y) =H0+K1exp(-K2y)

[0021] H (x,y) y represents the soil layer depth, H0 represents the initial thickness, K1 and K2 are parameters, and y is the spatial coordinate.

[0022] The calculation parameters are determined based on machine learning to fit and validate the expanded model.

[0023] According to a remote sensing-based soil quality analysis method provided by the present invention, the method for obtaining soil quality parameters based on the soil influence parameters, land type, and soil layer depth includes:

[0024] Based on the regression model, the soil impact parameters, land type, and soil layer depth were analyzed to obtain the soil quality parameters, as shown in the following formula:

[0025] M=β0t0+β1t1+β2t2+β3t3+β4t4+β5t5+ε

[0026] Where M is the soil quality parameter, t i β is any of the following: soil influence parameter, land type, or soil depth. i ε represents the regression coefficients, i = 0, 1, 2, 3, 4, 5, and ε is the error term.

[0027] The regression coefficients and error terms are determined based on machine learning fitting and validation of the regression model.

[0028] According to the soil quality analysis method based on remote sensing provided by the application, the soil influencing parameter comprises a soil average particle diameter.

[0029] The method for determining the soil average particle diameter comprises:

[0030] The pixel area of a target image corresponding to the soil quality parameter analysis is determined.

[0031] The target parameter representing the soil particle diameter is searched in the soil quality parameters.

[0032] The soil average particle diameter is determined based on the target parameter and the pixel area.

[0033] According to the soil quality analysis method based on remote sensing provided by the application, the soil quality analysis result is determined based on the soil quality parameters, which comprises:

[0034] The soil quality parameters are converted into standardized scores, and target weights corresponding to the soil quality parameters are determined.

[0035] The soil quality analysis result is obtained by weighted average of the standardized scores and the target weights.

[0036] The application further provides a soil quality analysis device based on remote sensing, which comprises:

[0037] A data acquisition module is configured to acquire satellite remote sensing data and DEM elevation data.

[0038] An image acquisition module is configured to acquire land structure images based on the satellite remote sensing data.

[0039] A segmentation and identification module is configured to segment and identify the land structure images based on image processing technology, and determine land types of different land structure regions.

[0040] A hierarchical identification module is configured to identify soil hierarchies of the land types based on the DEM elevation data, and obtain soil hierarchy depths.

[0041] A parameter determination module is configured to determine soil influencing parameters based on the remote sensing data, and obtain soil quality parameters based on the soil influencing parameters, the land types and the soil hierarchy depths.

[0042] A quality analysis module is configured to determine a soil quality analysis result based on the soil quality parameters.

[0043] The application further provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the remote sensing based soil quality analysis method according to any one of the above when executing the program.

[0044] The application further provides a non-transitory computer readable storage medium, which stores a computer program, wherein the computer program is executable on a processor to implement the remote sensing based soil quality analysis method according to any one of the above.

[0045] The application further provides a computer program product, comprising a computer program, wherein the computer program is executable on a processor to implement the remote sensing based soil quality analysis method according to any one of the above.

[0046] The application provides a remote sensing based soil quality analysis method, device and electronic device, which obtains satellite remote sensing data and DEM elevation data, obtains a land structure image based on the satellite remote sensing data, performs segmentation and identification on the land structure image based on an image processing technology, determines the land types of different land structure regions, performs soil level identification on each land type based on the DEM elevation data, obtains the soil level depth, determines the soil influence parameters based on the remote sensing data, obtains the soil quality parameters based on the soil influence parameters, the land types and the soil level depth, and finally determines the soil quality analysis result based on the soil quality parameters. The application can realize the comprehensiveness and accuracy of soil quality evaluation. BRIEF DESCRIPTION OF DRAWINGS

[0047] In order to more clearly illustrate the technical solutions of the present application or prior art, the following will briefly introduce the drawings needed to be used in the embodiments or prior art description. Obviously, the drawings in the following description are some embodiments of the present application, and other drawings can also be obtained by those skilled in the art without creative labor.

[0048] Figure 1 is one of the flowcharts of the remote sensing based soil quality analysis method provided by the application;

[0049] Figure 2 is another flowchart of the remote sensing based soil quality analysis method provided by the application;

[0050] Figure 3 is another flowchart of the remote sensing based soil quality analysis method provided by the application;

[0051] Figure 4 is a structural schematic diagram of the remote sensing based soil quality analysis device provided by the application;

[0052] Figure 5Fig. 1 is a structural schematic diagram of an electronic device provided by the present application. DETAILED DESCRIPTION

[0053] In order to make the objects, technical solutions and advantages of the present application clearer, the technical solutions of the present application will be described clearly and completely below with reference to the drawings in the present application. Obviously, the described embodiments are only some of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the protection scope of the present application.

[0054] Referring to Figure 1 The soil quality analysis method based on remote sensing provided by the present application comprises but is not limited to the following steps:

[0055] Step 110: acquiring satellite remote sensing data and DEM elevation data;

[0056] Step 120: acquiring a land structure image based on the satellite remote sensing data;

[0057] Step 130: segmenting and identifying the land structure image based on image processing technology to determine the land type of different land structure regions;

[0058] Step 140: identifying the soil level of each land type based on the DEM elevation data to obtain the soil level depth;

[0059] Step 150: determining a soil influence parameter based on the remote sensing data, and obtaining a soil quality parameter based on the soil influence parameter, the land type and the soil level depth;

[0060] Step 160: determining a soil quality analysis result based on the soil quality parameter.

[0061] The above steps will be described in detail below.

[0062] In the above steps 110 and 120, satellite remote sensing data and DEM elevation data are acquired, and then a land structure image is acquired from the satellite remote sensing data. The satellite remote sensing data is high-resolution remote sensing data. High-resolution satellite remote sensing data generally refers to data with finer spatial details and more accurate ground object information. The resolution is defined as the size of a unit ground area corresponding to a sensor pixel, and the resolution of high-resolution satellite remote sensing data can be several meters to tens of centimeters.

[0063] DEM (Digital Elevation Model) elevation data is a digital model used in geographic information systems (GIS) to describe the height of the Earth's surface, representing the elevation or height information of the surface in raster or vector form. DEM elevation data is also obtained through measurement or remote sensing technology, which can provide detailed description and quantitative analysis of the terrain surface. DEM elevation data records the elevation value of each position on the surface, usually in meters. DEM data can be used in many application fields, including terrain analysis, hydrological modeling, land use planning, three-dimensional visualization, etc.

[0064] Then, through the above steps 130 and 140, the acquired land structure image is identified by type and hierarchical depth.

[0065] The process of identifying the type of land structure image can be implemented using a convolutional neural network model, i.e., feature extraction, feature segmentation, etc. of the land structure image, so as to segment the land structure image into different land structures such as soil, desert, grassland, road, building, etc.

[0066] The process of identifying the hierarchical depth of the land structure image can use an expansion model to analyze the soil hierarchy through the habit thickness difference, and the division principle is to divide according to the habit thickness difference of different soil layers, so as to determine the soil hierarchical depth of different levels of the land structure image.

[0067] Further, through the above step 150, the soil quality parameters are obtained by combining the soil influence parameters and comprehensively analyzing the acquired land type and soil hierarchical depth.

[0068] Optionally, the soil influence parameters include at least one of NDVI value, elevation, surface temperature, and normalized water body index.

[0069] It can be understood that the above soil influence parameters, such as NDVI value, elevation, surface temperature, and normalized water body index, can be directly obtained from remote sensing data, and then combined with the land type and soil hierarchical depth obtained by the above steps 130 and 140 to determine the soil quality parameters, including but not limited to soil organic matter content, PH value, moisture content, average particle size diameter, etc.

[0070] Finally, according to the above acquired soil quality parameters combined with the weight data of each parameter, the final soil quality evaluation result is obtained.

[0071] The application provides a soil quality analysis method based on remote sensing, which comprises the following steps: acquiring satellite remote sensing data and DEM elevation data, obtaining a land structure image based on the satellite remote sensing data, segmenting and identifying the land structure image based on an image processing technology to determine the land type of different land structure regions, identifying the soil level of each land type based on the DEM elevation data to obtain the soil level depth, determining the soil influence parameters based on the remote sensing data, obtaining the soil quality parameters based on the soil influence parameters, the land type and the soil level depth, and finally determining the soil quality analysis result based on the soil quality parameters. The application can realize the comprehensiveness and accuracy of soil quality evaluation.

[0072] Reference Figure 2 In some optional embodiments, the land structure image is segmented and identified based on the image processing technology to determine the land type of different land structure regions, which comprises the following steps:

[0073] Step 210: image size adjustment, gray value balancing and edge detection are performed on the land structure image to obtain a pretreated image.

[0074] Step 220: the pretreated image is convolved, pooled and up-sampled based on a first convolutional neural network model to obtain the image features of the pretreated image, and different land structures are determined based on the image features.

[0075] Step 230: each land structure is identified based on a second convolutional neural network model to obtain the land type of the different land structure regions.

[0076] It can be understood that the embodiment is a specific way of land type identification, and the specific process is as follows:

[0077] Image preprocessing: image preprocessing is required first, including image size adjustment, gray value balancing, edge detection and other operations, to obtain better image quality and facilitate subsequent image segmentation and identification.

[0078] Image segmentation: the deep learning segmentation algorithm uses a first convolutional neural network model, such as a U-Net convolutional neural network model, to automatically extract image features through multiple convolution, pooling and up-sampling operations, realize image segmentation, and segment different land structures such as soil, desert, grassland, road and building.

[0079] Recognition algorithm: after obtaining the land structure region, a deep learning algorithm is used to identify the land type through a second convolutional neural network model, such as an AlexNet convolutional neural network model, taking the land structure in the second step as the input, to identify the land type such as farmland, woodland, grassland, urban area, mountain and desert.

[0080] The soil quality analysis method based on remote sensing provided by the application can accurately divide the land structure image obtained by remote sensing into various types, so as to analyze the soil quality of various types of land structure images, and comprehensively and accurately realize soil quality analysis.

[0081] In some optional embodiments, the soil layer depth is obtained by identifying soil layers of each of the land types based on the DEM elevation data, comprising:

[0082] The starting thickness and spatial coordinates of the land type are determined based on the DEM elevation data;

[0083] The soil layer depth is obtained based on the starting thickness, spatial coordinates and calculation parameters by using an expansion model, as shown in the following formula:

[0084] H (x,y) = H0+ K1exp(-K2y)

[0085] H (x,y) is the soil layer depth, H0 is the starting thickness, K1 and K2 are parameters, and y is the spatial coordinates;

[0086] The calculation parameters are determined based on machine learning fitting verification of the expansion model.

[0087] It can be understood that the embodiment is a process for obtaining the soil layer depth.

[0088] Based on the elevation data, the expansion model is used to analyze the soil layer by using the habit thickness difference, and the division principle is to divide according to the habit thickness difference of different soil layers, and the formula is as follows:

[0089] H (x,y) = H0+ K1exp(-K2y)

[0090] Wherein, H (x,y) represents the soil layer thickness at a position with a horizontal distance of y from the ground, H0 represents the starting thickness, K1 and K2 are parameters, and y is the spatial coordinates.

[0091] By actually sampling and measuring the depth of different soil types, machine learning is used to fit and verify the parameters K1 and K2, and finally the parameters K1 and K2 are used as the calculation parameters of the layer depth of different soil types in the entire area.

[0092] The application provides a remote sensing-based soil quality analysis method, which obtains the starting thickness and spatial coordinates of a land structure image through DEM elevation data, obtains relevant parameters through fitting verification, and then obtains the soil level depth of the land structure image, so as to analyze the soil quality according to the soil level depth, thereby realizing the comprehensiveness and accuracy of soil quality analysis.

[0093] In some optional embodiments, the soil quality parameter is obtained based on the soil influence parameter, the land type and the soil level depth, including:

[0094] The soil influence parameter, the land type and the soil level depth are analyzed based on a regression model to obtain the soil quality parameter, as shown in the following formula:

[0095] M = β 0 t 0 + β 1 t 1 + β 2 t 2 + β 3 t 3 + β 4 t 4 + β 5 t 5 + ε

[0096] Wherein, M is the soil quality parameter, t i is any one of the soil influence parameter, the land type or the soil level depth, β i is a regression coefficient, i = 0, 1, 2, 3, 4, 5, and ε is an error term.

[0097] The regression coefficient and the error term are determined based on machine learning fitting verification of the regression model.

[0098] It can be understood that the embodiment is a determination process of the soil quality parameter.

[0099] In the embodiment, the NDVI value, the elevation, the land surface temperature, the normalized water body index, the land type and the soil level depth are denoted as t0, t1, t2, t3, t4 and t5 respectively, and the corresponding regression coefficients are denoted as β0, β1, β2, β3, β4 and β5 respectively.

[0100] According to the above formula, the soil quality parameter obtained by using the soil influence parameter, the land type or the soil level depth is calculated.

[0101] Further, the soil influence parameter includes a soil average particle diameter.

[0102] The determination method of the soil average particle diameter includes:

[0103] The pixel area of a target image corresponding to the soil quality parameter analysis is determined.

[0104] A target parameter representing the soil particle diameter is traversed in the soil quality parameter.

[0105] determine the average soil particle diameter based on the target parameter and the pixel area.

[0106] The soil quality parameters include soil organic matter content, pH value, water content, average particle size, and the like.

[0107] In this embodiment, the average soil particle diameter is mainly determined.

[0108] First, the pixel area of the target image in the land type identification process of the land structure image is determined. Then all soil particle diameters in the target image are traversed, and finally the average soil particle diameter is determined by summation and averaging. The specific formula can be embodied as:

[0109]

[0110] where Di represents the diameter of the i-th particle, and Ai is the pixel area thereof. By traversing the entire land classification cutting image, the average particle diameter under different cutting images can be obtained.

[0111] In some optional embodiments, the soil quality analysis result is determined based on the soil quality parameters, including:

[0112] The soil quality parameters are converted into standardized scores, and target weights corresponding to each soil quality parameter are determined;

[0113] The standardized scores and the target weights are weighted and averaged to obtain the soil quality analysis result.

[0114] It can be understood that the present embodiment is a soil quality analysis process.

[0115] First, the soil quality parameters are converted into standardized scores, and the numerical values are scaled to 0-1 in proportion. Then, according to the weighted average method, for n soil parameters, assuming that the standardized score of the i-th parameter is S i , and the corresponding weight is W i , then the comprehensive score S tat is calculated as follows:

[0116]

[0117] In the actual implementation process, the soil quality can be divided into overall grades according to the score interval, i.e. excellent: [0.8, 1.0); good: [0.6, 0.8); medium: [0.4, 0.6); poor: [0.2, 0.4); and very poor: [0.0, 0.2).

[0118] The soil quality analysis method based on remote sensing provided by the application realizes comprehensiveness and accuracy of soil quality analysis by converting soil quality parameters into standardized scores and performing weighted summation by using weight coefficients to obtain a final comprehensive score of soil quality.

[0119] The soil quality analysis device based on remote sensing provided by the application is described below, and the soil quality analysis device based on remote sensing described below can be correspondingly referred to the soil quality analysis method based on remote sensing described above.

[0120] Referring to Figure 4 The soil quality analysis device based on remote sensing provided by the application comprises the following modules:

[0121] The data acquisition module 410 is configured to acquire satellite remote sensing data and DEM elevation data.

[0122] The image acquisition module 420 is configured to acquire a land structure image based on the satellite remote sensing data.

[0123] The segmentation and identification module 430 is configured to segment and identify the land structure image based on image processing technology to determine the land type of different land structure regions.

[0124] The hierarchical identification module 440 is configured to identify the soil hierarchy of each land type based on the DEM elevation data to obtain the soil hierarchy depth.

[0125] The parameter determination module 450 is configured to determine soil influence parameters based on the remote sensing data, and to obtain soil quality parameters based on the soil influence parameters, the land type, and the soil hierarchy depth.

[0126] The quality analysis module 460 is configured to determine a soil quality analysis result based on the soil quality parameters.

[0127] The soil quality analysis device based on remote sensing provided by the application acquires satellite remote sensing data and DEM elevation data, acquires a land structure image based on the satellite remote sensing data, segments and identifies the land structure image based on image processing technology to determine the land type of different land structure regions, identifies the soil hierarchy of each land type based on the DEM elevation data to obtain the soil hierarchy depth, determines soil influence parameters based on the remote sensing data, obtains soil quality parameters based on the soil influence parameters, the land type, and the soil hierarchy depth, and finally determines a soil quality analysis result based on the soil quality parameters. The application can realize comprehensiveness and accuracy of soil quality evaluation.

[0128] In some optional embodiments, the segmentation and identification of the land structure image based on image processing technology to determine the land type of different land structure regions comprises:

[0129] perform image size adjustment, gray value balancing and edge detection on the land structure image to obtain a preprocessed image;

[0130] perform convolution, pooling and up-sampling on the preprocessed image based on a first convolutional neural network model to obtain image features of the preprocessed image, and determine different land structures based on the image features;

[0131] identify each of the land structures based on a second convolutional neural network model to obtain land types of the different land structure regions.

[0132] In some optional embodiments, the soil influence parameters include at least one of an NDVI value, an elevation, a land surface temperature, and a normalized water body index.

[0133] In some optional embodiments, the soil level identification based on the DEM elevation data for each of the land types to obtain a soil level depth includes:

[0134] determining a starting thickness and a spatial coordinate of the land type based on the DEM elevation data;

[0135] obtaining the soil level depth based on the starting thickness, the spatial coordinate and a calculation parameter through an expansion model, as shown in the following formula:

[0136] H (x,y) = H0+ K1exp(-K2y)

[0137] H (x,y) is the soil level depth, H0 is the starting thickness, K1 and K2 are parameters, and y is the spatial coordinate.

[0138] wherein the calculation parameter is determined based on machine learning fitting and verification of the expansion model

[0139] In some optional embodiments, the soil quality parameter is obtained based on the soil influence parameters, the land types and the soil level depth, including:

[0140] analyzing the soil influence parameters, the land types and the soil level depth based on a regression model to obtain the soil quality parameter, as shown in the following formula:

[0141] M = β0t0+ β1t1+ β2t2+ β3t3+ β4t4+ β5t5+ ε

[0142] wherein M is the soil quality parameter, t i is any one of the soil influence parameters, the land types or the soil level depth, β iε represents the regression coefficients, i = 0, 1, 2, 3, 4, 5, and ε is the error term.

[0143] The regression coefficients and error terms are determined based on machine learning to fit and validate the regression model.

[0144] In some optional embodiments, the soil quality parameter includes the average soil particle diameter;

[0145] The method for determining the average particle diameter of the soil includes:

[0146] Determine the pixel area of ​​the target image corresponding to the soil quality parameter analysis;

[0147] The target parameters representing soil particle diameter are iterated through in the soil quality parameters;

[0148] The average particle diameter of the soil is determined based on the target parameters and pixel area.

[0149] In some optional embodiments, determining the soil quality analysis results based on the soil quality parameters includes:

[0150] The soil quality parameters are converted into standardized scores, and the target weights corresponding to each soil quality parameter are determined.

[0151] The soil quality analysis results are obtained by taking a weighted average of the standardized score and the target weight.

[0152] Figure 5 An example is a schematic diagram of the physical structure of an electronic device, such as... Figure 5 As shown, the electronic device may include: a processor 510, a communication interface 520, a memory 530, and a communication bus 540, wherein the processor 510, the communication interface 520, and the memory 530 communicate with each other via the communication bus 540. The processor 510 can call logical instructions in the memory 530 to execute a remote sensing-based soil quality analysis method, which includes:

[0153] Acquire satellite remote sensing data and DEM elevation data;

[0154] Land structure images are obtained based on the satellite remote sensing data;

[0155] Based on image processing technology, the land structure image is segmented and identified to determine the land type of different land structure areas;

[0156] Soil layer identification is performed on each of the land types based on the DEM elevation data to obtain soil layer depth;

[0157] determining a soil impact parameter based on the remote sensing data, and determining a soil quality parameter based on the soil impact parameter, the land type, and the soil layer depth;

[0158] determining a soil quality analysis result based on the soil quality parameter.

[0159] Further, the logic instructions in the memory 530 described above can be implemented in the form of software functional units and sold or used as independent products, and can be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the present application or parts of the present application that essentially contribute to the prior art or parts of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk or an optical disk, and various media that can store program codes.

[0160] In another aspect, the present application also provides a computer program product, which includes a computer program, the computer program can be stored on a non-transitory computer readable storage medium, and the computer program can be executed by a processor to enable a computer to execute the remote sensing based soil quality analysis method provided by the above-mentioned methods, the method includes:

[0161] acquiring satellite remote sensing data and DEM elevation data;

[0162] acquiring a land structure image based on the satellite remote sensing data;

[0163] segmenting and identifying the land structure image based on image processing technology to determine the land type of different land structure regions;

[0164] identifying the soil layer of each land type based on the DEM elevation data to obtain the soil layer depth;

[0165] determining a soil impact parameter based on the remote sensing data, and determining a soil quality parameter based on the soil impact parameter, the land type, and the soil layer depth;

[0166] determining a soil quality analysis result based on the soil quality parameter.

[0167] In yet another aspect, the present application also provides a non-transitory computer-readable storage medium having stored thereon a computer program, which, when executed by a processor, implements a remote sensing-based soil quality analysis method provided by each of the above methods, the method comprising:

[0168] obtaining satellite remote sensing data and DEM elevation data;

[0169] obtaining a land structure image based on the satellite remote sensing data;

[0170] segmenting and identifying the land structure image based on image processing technology to determine the land type of different land structure regions;

[0171] identifying the soil level of each of the land types based on the DEM elevation data to obtain the soil level depth;

[0172] determining a soil influence parameter based on the remote sensing data, and obtaining a soil quality parameter based on the soil influence parameter, the land type, and the soil level depth;

[0173] determining a soil quality analysis result based on the soil quality parameter.

[0174] The apparatus embodiments described above are merely illustrative, wherein the units described as separate components can or can not be physically separate, and the components displayed as units can or can not be physical units, i.e., can be located in one place, or can be distributed on multiple network units. Part or all of the modules can be selected to achieve the purpose of the present embodiment scheme according to actual needs. Those skilled in the art can understand and implement without creative labor.

[0175] From the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be realized by means of software plus a necessary general hardware platform, and of course can also be realized by hardware. Based on such understanding, the above technical solutions, essentially or in terms of the contribution to the prior art, can be embodied in the form of a software product, which can be stored in a computer-readable storage medium, such as a ROM / RAM, a magnetic disk, an optical disk, etc., and includes a plurality of instructions to make a computer device (which can be a personal computer, a server, or a network device, etc.) execute the methods described in each embodiment or some parts of the embodiments.

[0176] It should be pointed out finally that the above embodiments are only used to illustrate the technical solutions of the present application, but not to limit the same; and although the present application has been described in detail with reference to the foregoing embodiments, it should be appreciated by those skilled in the art that the technical solutions recorded in the foregoing embodiments can be modified, or some technical features thereof can be replaced equivalently; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A remote sensing-based soil quality analysis method, characterized by, The method comprises the following steps: acquiring satellite remote sensing data and DEM elevation data; acquiring land structure images based on the satellite remote sensing data; segmenting and identifying the land structure images based on image processing technology to determine the land types of different land structure regions; identifying soil levels of each land type based on the DEM elevation data to obtain soil level depths; determining soil influence parameters based on the remote sensing data, and obtaining soil quality parameters based on the soil influence parameters, land types, and soil level depths; determining soil quality analysis results based on the soil quality parameters; The method of identifying soil levels of each land type based on the DEM elevation data to obtain soil level depths comprises the following steps: determining the starting thickness and spatial coordinates of the land type based on the DEM elevation data; obtaining the soil level depth based on the starting thickness, spatial coordinates, and calculation parameters through an expansion model, as shown in the following formula: ； for the depth of the soil layer, for the initial thickness, and is a parameter, y is the spatial coordinate; wherein the calculation parameters are determined based on machine learning fitting and verification of the expansion model; The method of obtaining soil quality parameters based on the soil influence parameters, land types, and soil level depths comprises the following steps: analyzing the soil influence parameters, land types, and soil level depths based on a regression model to obtain the soil quality parameters, as shown in the following formula: ； wherein M is the soil quality parameter, is any one of a soil impact parameter, a land type, and a soil layer depth, is a regression coefficient, i = 0, 1, 2, 3, 4, 5, and ε is an error term. The regression coefficients and error terms are determined based on machine learning fitting and verification of the regression model.

2. The remote-sensing-based soil quality analysis method according to claim 1, characterized in that, The soil influence parameters comprise a soil average particle diameter; The method of determining the soil average particle diameter comprises the following steps: determining the pixel area of a target image corresponding to soil quality parameter analysis; traversing a target parameter representing soil particle diameter in the soil quality parameters; determining the soil average particle diameter based on the target parameter and pixel area.

3. The remote-sensing-based soil quality analysis method according to claim 1, characterized in that, The method of determining soil quality analysis results based on the soil quality parameters comprises the following steps: converting the soil quality parameters into standardized scores, and determining target weights corresponding to each soil quality parameter; performing weighted averaging on the standardized scores and target weights to obtain the soil quality analysis results.

4. The remote-sensing-based soil quality analysis method of claim 1, wherein, The method of segmenting and identifying the land structure images based on image processing technology to determine the land types of different land structure regions comprises the following steps: performing image size adjustment, gray value balancing, and edge detection on the land structure images to obtain preprocessed images; performing convolution, pooling, and upsampling on the preprocessed images based on a first convolutional neural network model to obtain image features of the preprocessed images, and determining different land structures based on the image features; identifying each land structure based on a second convolutional neural network model to obtain the land types of different land structure regions.

5. The remote-sensing-based soil quality analysis method according to any one of claims 1 to 4, characterized in that, The soil influence parameters comprise at least one of an NDVI value, an elevation, a land surface temperature, and a normalized water body index.

6. A remote sensing-based soil quality analysis device, characterized by, The method comprises the following steps: a data acquisition module for acquiring satellite remote sensing data and DEM elevation data; an image acquisition module for acquiring land structure images based on the satellite remote sensing data; The segmentation and recognition module is configured to segment and recognize the land structure image based on an image processing technique, and determine a land type of different land structure regions; The hierarchical recognition module is configured to perform soil hierarchical recognition on each land type based on the DEM elevation data, and obtain a soil hierarchical depth; The parameter determination module is configured to determine a soil influence parameter based on the remote sensing data, and obtain a soil quality parameter based on the soil influence parameter, the land type, and the soil hierarchical depth; The quality analysis module is configured to determine a soil quality analysis result based on the soil quality parameter; The hierarchical recognition on each land type based on the DEM elevation data, and obtaining the soil hierarchical depth, includes: determining a starting thickness and a spatial coordinate of the land type based on the DEM elevation data; obtaining the soil hierarchical depth based on the starting thickness, the spatial coordinate, and a calculation parameter through an expansion model, as shown in the following formula: ； is the depth of the soil layer, is the initial thickness, and is a parameter, y is the spatial coordinate; wherein the calculation parameter is determined based on machine learning fitting and verification on the expansion model; The obtaining of the soil quality parameter based on the soil influence parameter, the land type, and the soil hierarchical depth includes: analyzing the soil influence parameter, the land type, and the soil hierarchical depth based on a regression model, and obtaining the soil quality parameter, as shown in the following formula: ； wherein M is the soil quality parameter, is any one of a soil impact parameter, a land type, and a soil layer depth, is a regression coefficient, i = 0, 1, 2, 3, 4, 5, and ε is an error term. The regression coefficient and the error term are determined based on machine learning fitting and verification on the regression model.

7. An electronic device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, The processor executes the program to implement the remote sensing-based soil quality analysis method according to any one of claims 1 to 5.

8. A non-transitory computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to implement the remote sensing-based soil quality analysis method according to any one of claims 1 to 5.

Images