A method, device, equipment, medium and product for in-situ measurement of soil texture

Abstract

The present application discloses a method, device, equipment, medium and product for in-situ measurement of soil texture, relating to the field of soil texture measurement. The method includes: obtaining dielectric spectrum data of a soil sample to be measured; preprocessing the dielectric spectrum data to obtain processed data; inputting the processed data into a soil texture prediction model to obtain a prediction result; the prediction result includes the category of soil texture; the soil texture prediction model is a model constructed based on a machine learning method to represent the mapping relationship between dielectric spectrum data and soil texture. The present application aims to quickly and accurately realize the prediction of soil texture.

Description

Technical Field

[0001] The present application relates to the field of soil texture measurement, and particularly to a method, device, equipment, medium and product for in-situ measurement of soil texture. Background Art

[0002] Soil texture is an important agricultural property of soil, which is the percentage content of soil particles of each level. Soil texture classification divides soil into several categories according to the similarity of soil particle composition.

[0003] Soil texture refers to the proportion of particles with different diameters (such as sand particles, silt particles and clay particles) in the soil, also known as soil particle composition or mechanical composition. It affects the air permeability, water permeability, water retention and fertility of the soil, and is crucial for plant growth and the health of the ecosystem. Soils with different textures have different tillage properties. Understanding soil texture helps to scientifically and reasonably manage the soil and plant crops. Soil texture classification divides soil into several categories according to the similarity of soil particle composition. Usually, a soil texture triangle is used to describe soil texture, which represents the three main particle components of sand, silt and clay in the soil at the vertices of an equilateral triangle. Any point inside this triangle represents a specific soil texture combination, which includes the proportions of the three particles.

[0004] Currently, the methods for measuring soil texture include field observation method, suspension analysis method, laser particle size analyzer, sedimentation tube method and density gradient tube method. The field observation method is a traditional method, which preliminarily judges the soil texture by observing the characteristics such as particle size, color and texture of soil samples, but the accuracy is affected by the operator's experience and subjective factors. Modern technologies such as laser particle size analyzers can quickly and accurately measure the size distribution of soil particles, but the equipment is expensive. The sedimentation tube method and the density gradient tube method infer the soil texture by measuring the sedimentation positions of different particles in the soil sample in the liquid, which has a certain accuracy but requires a long time or complex laboratory equipment. Summary of the Invention

[0005] The purpose of the present application is to provide a method, device, equipment, medium and product for in-situ measurement of soil texture, which can quickly and accurately predict soil texture.

[0006] To achieve the above purpose, the present application provides the following solutions:

[0007] In the first aspect, the present application provides a method for in-situ measurement of soil texture, including:

[0008] Obtaining dielectric spectrum data of a soil sample to be measured;

[0009] Preprocessing the dielectric spectrum data to obtain processed data;

[0010] Input the processed data into the soil texture prediction model to obtain a prediction result; the prediction result includes the category of the soil texture; the soil texture prediction model is a model constructed based on a machine learning method to represent the mapping relationship between dielectric spectrum data and soil texture.

[0011] Optionally, the dielectric spectrum data is measured by using a vector network analyzer and a dielectric probe.

[0012] Optionally, preprocess the dielectric spectrum data to obtain processed data, specifically including:

[0013] Perform normalization processing on the dielectric spectrum data to eliminate the dimension difference and obtain normalized dielectric spectrum data;

[0014] Extract the data features of the normalized dielectric spectrum data to capture the spectral correlation and dielectric relaxation behavior characteristics, and obtain the processed data.

[0015] Optionally, the method for determining the soil texture prediction model specifically includes:

[0016] Obtain an information data set; the information data set includes historical processed data and label data corresponding to the historical processed data; the label data is the category of the soil texture;

[0017] Divide the information data set into a training set and a validation set;

[0018] Construct a machine learning model;

[0019] Input the training set into the machine learning model, train the machine learning model, and optimize and adjust the hyperparameters of the trained machine learning model by using a grid search method or a Bayesian optimization method to obtain an optimized machine learning model;

[0020] Use the validation set to validate the optimized machine learning model based on validation metrics to obtain a validated machine learning model;

[0021] Determine the validated machine learning model as the soil texture prediction model.

[0022] Optionally, the machine learning model uses a support vector machine, a random forest or a deep neural network.

[0023] Optionally, the validation metrics include: accuracy, confusion matrix and F1 score.

[0024] In a second aspect, the present application provides a soil texture in-situ measurement device, including:

[0025] A data acquisition module for acquiring dielectric spectrum data of a soil sample to be measured;

[0026] A preprocessing module for preprocessing the dielectric spectrum data to obtain processed data;

[0027] A prediction module for inputting the processed data into a soil texture prediction model to obtain a prediction result; the prediction result includes the category of the soil texture; the soil texture prediction model is a model constructed based on a machine learning method to characterize the mapping relationship between dielectric spectrum data and soil texture.

[0028] In a third aspect, the present application provides a computer device, including: a memory, a processor, and a computer program stored on the memory and executable on the processor, where the processor executes the computer program to implement the soil texture in-situ measurement method described above.

[0029] In a fourth aspect, the present application provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the soil texture in-situ measurement method described above is implemented.

[0030] In a fifth aspect, the present application provides a computer program product, including a computer program, and when the computer program is executed by a processor, the soil texture in-situ measurement method described above is implemented.

[0031] According to the specific embodiments provided by the present application, the present application has the following technical effects:

[0032] The present application provides a soil texture in-situ measurement method, which acquires dielectric spectrum data of a soil sample to be measured; preprocesses the dielectric spectrum data to obtain processed data; and inputs the processed data into a soil texture prediction model to obtain a prediction result. Since dielectric spectrum data describes the response of dielectric materials to electromagnetic fields of different frequencies and can reflect the absorption and propagation characteristics of soil to electromagnetic waves, based on the characteristic that the dielectric relaxation characteristics of different types of soil textures are also different, the present application determines the category of soil texture by constructing a soil texture prediction model. Also, since the soil texture prediction model is a model constructed based on a machine learning method to characterize the mapping relationship between dielectric spectrum data and soil texture and is not affected by the operator's experience and subjective factors, the accuracy is improved and the speed is also increased. Thus, the present application can quickly and accurately predict the soil texture. Description of the Drawings

[0033] To more clearly illustrate the technical solutions in the embodiments of the present application or the related art, the following will briefly introduce the accompanying drawings required for use in the embodiments. Obviously, the accompanying drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other accompanying drawings can be obtained based on these drawings.

[0034] Figure 1 It is a flowchart of the in-situ measurement method for soil texture;

[0035] Figure 2 It is a schematic diagram of the measurement results for dielectric spectrum measurement. Specific embodiments

[0036] The following will clearly and completely describe the technical solutions in the embodiments of the present application in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only some embodiments of the present application, rather than all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present application.

[0037] To make the above objects, features, and advantages of the present application more obvious and understandable, the present application will be further described in detail below in conjunction with the accompanying drawings and specific embodiments.

[0038] Dielectric spectrum describes the response of dielectric materials to electromagnetic fields of different frequencies. The dielectric spectrum of soil reflects the absorption and propagation characteristics of soil to electromagnetic waves, and its characteristics are related to factors such as soil texture, water content, salt content, and temperature. Among them, different types of soil textures have different dielectric relaxation characteristics due to containing different types and proportions of particles. For example, soils containing a large amount of clay usually show higher dielectric relaxation, while sandy soils show lower dielectric relaxation.

[0039] In an exemplary embodiment, as Figure 1 shown, a method for in-situ measurement of soil texture is provided. This method is executed by a computer device, specifically, it can be executed alone by a computer device such as a terminal or a server, or jointly executed by a terminal and a server. In the embodiments of the present application, taking this method applied to a server as an example for illustration, it includes the following steps.

[0040] As Figure 1 shown, the present application provides a method for in-situ measurement of soil texture, and this method includes:

[0041] Step 100: Obtain the dielectric spectrum data of the soil sample to be measured. The dielectric spectrum data is measured using a vector network analyzer and a dielectric probe.

[0042] Step 200: Preprocess the dielectric spectrum data to obtain processed data.

[0043] In one embodiment, preprocessing the dielectric spectrum data to obtain processed data specifically includes:

[0044] Perform standardization processing on the dielectric spectrum data to eliminate the dimension difference and obtain standardized dielectric spectrum data.

[0045] Extract the data features of the standardized dielectric spectrum data to capture the spectrum correlation and dielectric relaxation behavior characteristics, and obtain processed data.

[0046] Step 300: Input the processed data into the soil texture prediction model to obtain a prediction result. The prediction result includes the category of the soil texture. The category is sandy soil, sandy loam, loam, silt loam, or clay, etc., and this category is not limited to these several. The soil texture prediction model is a model constructed based on machine learning methods to characterize the mapping relationship between dielectric spectrum data and soil texture.

[0047] The method for determining the soil texture prediction model specifically includes:

[0048] Obtain an information data set; the information data set includes historical processed data and label data corresponding to the historical processed data; the label data is the category of the soil texture. Divide the information data set into a training set and a validation set.

[0049] Construct a machine learning model; input the training set into the machine learning model, train the machine learning model, and use the grid search method or Bayesian optimization method to optimize and adjust the hyperparameters of the trained machine learning model to obtain an optimized machine learning model.

[0050] Use the validation set to validate the optimized machine learning model based on validation metrics to obtain a validated machine learning model; determine the validated machine learning model as the soil texture prediction model.

[0051] In one embodiment, the machine learning model uses a support vector machine, random forest, or deep neural network. The validation metrics include: accuracy, confusion matrix, and F1 score.

[0052] In practical applications, the technical concept of the method mentioned in this application has the following corresponding process.

[0053] First, a soil sample library covering different soil texture ranges needs to be established. The selection of soil samples should cover different types such as sandy soil, loam soil, and clay soil, and at the same time cover various combinations of water content, bulk density, electrical conductivity, and porosity. The water content should cover the full range from dry state to saturated state, the bulk density should be controlled within the range of 0.9 g / cm³ to 1.6 g / cm³, the electrical conductivity should be considered from low salt to high salt conditions, and the porosity should be controlled within the range of 30% to 60%. The establishment of the soil sample library should be as comprehensive as possible to ensure the wide applicability of subsequent models.

[0054] Secondly, use a vector network analyzer and a dielectric probe to measure the dielectric spectrum of soil samples, and the frequency range is set from dozens of kHz to several GHz. The equipment for measurement (i.e., the vector network analyzer and the dielectric probe) needs to be strictly calibrated, including short-circuit calibration, open-circuit calibration, and load calibration. When measuring, insert the dielectric probe into the soil sample and gradually obtain the real part and imaginary part of the dielectric constant at different frequencies. The measurement results include the real part of the dielectric constant (reflecting capacitive behavior) and the imaginary part (reflecting conductivity and loss characteristics). To ensure the reliability of the data, it is recommended to perform multiple repeated measurements on each sample. The schematic diagram corresponding to the measurement results is shown in Figure 2 . Figure 2 in is the volumetric water content; quartz sand (0.38) refers to quartz sand (volumetric water content = 0.38 cm 3 cm -3 ); sodium bentonite (0.3) refers to sodium bentonite (volumetric water content = 0.3 cm 3 cm -3 ); sodium bentonite (0.05) refers to sodium bentonite (volumetric water content = 0.05 cm 3 cm -3 ).

[0055] Next, based on the measured data, that is, the dielectric spectrum data, use a machine learning model to construct the relationship between the dielectric spectrum and soil texture. Preprocessing the dielectric spectrum data is the first step in model construction. It is necessary to standardize the dielectric spectrum data to eliminate the dimension difference, and at the same time analyze the data characteristics to ensure that the model can capture the frequency correlation and dielectric relaxation behavior. In the process of machine learning model selection and training, support vector machine (SVM), random forest (RF), or deep neural networks (DNN) can be considered.

[0056] SVM is suitable for dealing with small-sample high-dimensional data, while RF can effectively handle non-linear relationships and has strong anti-noise ability. DNN has strong learning ability for complex non-linear relationships and is suitable for large-scale sample data. The input of the machine learning model includes the real part and imaginary part of the dielectric constant and their trends with frequency change, that is, the input processed data, and its output target is the category of soil texture, such as sandy soil, loam, clay, etc.

[0057] After training, it is necessary to optimize the hyperparameters of the trained machine learning model to improve the prediction accuracy. Methods such as grid search or Bayesian optimization can be used to adjust the hyperparameters. Subsequently, the optimized machine learning model is verified through an independent validation set, that is, performance evaluation. The verification metrics include accuracy, confusion matrix, and F1 score. The F1 score includes Precision and Recall. In addition, to verify the actual applicability, it can be compared with traditional laboratory texture analysis methods to evaluate the consistency of the results of both.

[0058] Finally, the obtained model, that is, the soil texture prediction model, is deployed to the actual in-situ measurement device to achieve real-time measurement of soil texture. The deployment of the soil texture prediction model needs to be combined with an embedded processor, which can receive dielectric spectrum data in real time and process it, and finally output the measurement result, that is, the category of soil texture. The in-situ measurement device can be designed with a friendly user interface to directly display the texture classification result (the category of soil texture) or output soil-related parameters in digital form.

[0059] Currently, the acquisition of soil texture information relies on laboratory analysis or measurement after field sampling. The method for in-situ acquisition of soil texture based on dielectric spectrum information in this application has multiple advantages. First, it is a non-destructive measurement method that does not require collecting soil samples or treating the soil, avoiding interference with the soil environment. Second, this method can obtain data in real time, which helps to understand the soil texture situation in a timely manner and provides immediate information for agricultural production and soil management. At the same time, compared with traditional field surveys or laboratory analyses, the method based on dielectric spectrum information is more efficient and can obtain a large amount of soil texture data in a short time.

[0060] Based on the same inventive concept, the embodiments of this application also provide a soil texture in-situ measurement device for implementing the above-mentioned soil texture in-situ measurement method. The solution provided by this device to solve the problem is similar to the solution described in the above method. Therefore, the specific limitations in one or more embodiments of the soil texture in-situ measurement device provided below can refer to the limitations on the soil texture in-situ measurement method in the above text and will not be elaborated here.

[0061] In an exemplary embodiment, a device for in-situ measurement of soil texture is provided, including:

[0062] A data acquisition module for acquiring dielectric spectrum data of a soil sample to be measured.

[0063] A preprocessing module for preprocessing the dielectric spectrum data to obtain processed data.

[0064] A prediction module for inputting the processed data into a soil texture prediction model to obtain a prediction result; the prediction result includes the category of soil texture. The category is sandy soil, sandy loam, loam, silt loam or clay, etc., and this category is not limited to these several. The soil texture prediction model is a model constructed based on machine learning methods to characterize the mapping relationship between dielectric spectrum data and soil texture.

[0065] In an exemplary embodiment, a computer device is provided. The computer device can be a server or a terminal. The computer device includes a processor, a memory, an input / output interface (Input / Output, abbreviated as I / O) and a communication interface. Among them, the processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store in-situ measurement data of soil texture. The input / output interface of the computer device is used to exchange information between the processor and external devices. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a method for in-situ measurement of soil texture is implemented.

[0066] Those skilled in the art can understand that the computer device structure shown in this application is only a part of the structure related to the solution of this application, and does not constitute a limitation on the computer device to which the solution of this application is applied. A specific computer device may include more or fewer components than those described above, or combine certain components, or have different component arrangements. In an exemplary embodiment, a computer device is provided, including a memory and a processor. A computer program is stored in the memory, and when the processor executes the computer program, the steps in the above method embodiments are implemented.

[0067] In an exemplary embodiment, a computer-readable storage medium is provided, storing a computer program, and when the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0068] In an exemplary embodiment, a computer program product is provided, including a computer program which, when executed by a processor, implements the steps in the above method embodiments.

[0069] In this application, all actions of obtaining signals, information, or data are carried out on the premise of complying with the corresponding data protection regulations and policies of the country where it is located and obtaining authorization from the owner of the corresponding device. It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use, and processing of relevant data need to comply with relevant regulations.

[0070] Those of ordinary skill in the art can understand that all or part of the processes in the above method embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the above method embodiments. Among them, any reference to a memory, database, or other medium used in the embodiments provided in this application can include at least one of non-volatile and volatile memories. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc.

[0071] In each of the embodiments provided in the present application, the database involved may include at least one of a relational database and a non-relational database. The non-relational database may include a distributed database based on blockchain, etc., and is not limited thereto. In each of the embodiments provided in the present application, the processor may be a general-purpose processor, a central processing unit, a graphics processing unit, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., and is not limited thereto.

[0072] The technical features of the above embodiments can be combined arbitrarily. For the sake of concise description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope described in this specification.

[0073] Specific examples are used in this article to elaborate on the principles and implementation manners of the present application. The description of the above embodiments is only used to help understand the method and its core idea of the present application; at the same time, for those of ordinary skill in the art, according to the idea of the present application, there will be changes in the specific implementation manners and application scopes. In summary, the content of this specification should not be construed as a limitation to the present application.

Claims

1. A method for in-situ measurement of soil texture, characterized in that, The in-situ soil texture measurement method includes: Obtaining dielectric spectrum data of a soil sample to be measured; Preprocessing the dielectric spectrum data to obtain processed data; Inputting the processed data into a soil texture prediction model to obtain a prediction result; the prediction result includes the category of soil texture; the soil texture prediction model is a model constructed based on machine learning methods to represent the mapping relationship between dielectric spectrum data and soil texture; The method for determining the soil texture prediction model specifically includes: Obtaining an information data set; the information data set includes historical processed data and label data corresponding to the historical processed data; the label data is the category of soil texture; Dividing the information data set into a training set and a validation set; Constructing a machine learning model; Inputting the training set into the machine learning model, training the machine learning model, and using a grid search method or a Bayesian optimization method to optimize and adjust the hyperparameters of the trained machine learning model to obtain an optimized machine learning model; Using the validation set to validate the optimized machine learning model based on validation metrics to obtain a validated machine learning model; Determining the validated machine learning model as the soil texture prediction model.

2. The in-situ soil texture measurement method according to claim 1, characterized in that, The dielectric spectrum data is measured using a vector network analyzer and a dielectric probe.

3. The in-situ soil texture measurement method according to claim 1, characterized in that, Preprocessing the dielectric spectrum data to obtain processed data specifically includes: Performing standardization processing on the dielectric spectrum data to eliminate dimension differences and obtain standardized dielectric spectrum data; Extracting the data features of the standardized dielectric spectrum data to capture spectral correlation and dielectric relaxation behavior features to obtain processed data.

4. The in-situ soil texture measurement method according to claim 1, characterized in that The machine learning model uses a support vector machine, a random forest, or a deep neural network.

5. The in-situ soil texture measurement method according to claim 1, characterized in that, The validation metrics include: accuracy, confusion matrix, and F1 score.

6. An in-situ soil texture measuring device, characterized in that, The in-situ soil texture measurement device includes: A data acquisition module for obtaining dielectric spectrum data of a soil sample to be measured; A preprocessing module for preprocessing the dielectric spectrum data to obtain processed data; A prediction module for inputting the processed data into a soil texture prediction model to obtain a prediction result; the prediction result includes the category of soil texture; the soil texture prediction model is a model constructed based on machine learning methods to represent the mapping relationship between dielectric spectrum data and soil texture; The method for determining the soil texture prediction model specifically includes: Obtaining an information data set; the information data set includes historical processed data and label data corresponding to the historical processed data; the label data is the category of soil texture; Dividing the information data set into a training set and a validation set; Constructing a machine learning model; Inputting the training set into the machine learning model, training the machine learning model, and using a grid search method or a Bayesian optimization method to optimize and adjust the hyperparameters of the trained machine learning model to obtain an optimized machine learning model; Using the validation set to validate the optimized machine learning model based on validation metrics to obtain a validated machine learning model; The verified machine learning model is determined as the soil texture prediction model.

7. A computer device, comprising: A memory, a processor, and a computer program stored on the memory and executable on the processor, wherein the processor executes the computer program to implement the in-situ soil texture measurement method according to any one of claims 1-5.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the in-situ soil texture measurement method according to any one of claims 1-5.

9. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the in-situ soil texture measurement method according to any one of claims 1-5.

Images