Soil fertility quality analysis method and analysis system based on corn-growing farmland

Abstract

The present invention provides a soil fertility quality analysis method and analysis system based on corn-planted cultivated land, which relates to the technical field of soil quality analysis and evaluation, and includes: collecting visible-near infrared spectral images of soil samples, extracting reflectance data of key bands from the spectral images, and calculating a soil health index and a nutrient balance index; calculating a soil fertility quality comprehensive score for comprehensively evaluating soil fertility according to the soil health index, the nutrient balance index and the soil moisture content; comparing the soil fertility quality comprehensive score with a preset threshold to evaluate soil fertility; introducing a machine learning model, training and predicting historical soil data, building a quality analysis model based on a random forest algorithm, putting the trained quality analysis model into use, and continuously learning new data to dynamically adjust and update, thereby simplifying the soil detection process, being able to monitor soil fertility changes in real time, and helping agricultural managers to make scientific decisions.

Description

Technical Field

[0001] The present invention relates to the technical field of soil quality analysis and evaluation, and in particular to a soil fertility quality analysis method and analysis system based on corn-planting farmland. Background Art

[0002] In modern agriculture, the assessment and monitoring of soil fertility is an important part of ensuring the healthy growth of crops and improving agricultural production efficiency. However, traditional soil testing methods often rely on laboratory analysis, which is not only time-consuming but also costly. Especially in large-scale farmland management, real-time monitoring cannot be achieved. Traditional methods are usually limited to the determination of a single indicator of the physical and chemical properties of soil samples, such as pH value, organic matter content, nutrient concentration, etc., and fail to effectively integrate these data, resulting in a lack of comprehensiveness and systematicness in the assessment of soil health status. In addition, there are complex interrelationships between many soil-related indicators. Using a single indicator alone often cannot accurately reflect the actual fertility status of the soil, which can easily lead to misjudgment and affect the scientific nature of agricultural production decisions.

[0003] In addition, as climate change and human activities have increasingly significant impacts on the soil environment, problems such as soil nutrient loss, acidification and pollution are becoming increasingly serious. Traditional assessment methods are not adaptable enough to the rapidly changing soil environment. Many existing technologies are still unable to capture changes in soil properties in a timely manner and lack the ability to dynamically monitor soil fertility conditions. More importantly, traditional methods usually do not make full use of modern information technology and data analysis methods, and fail to achieve deep mining and intelligent analysis of soil data. Therefore, there is an urgent need for a new, comprehensive soil fertility assessment system that can obtain multiple key indicators of soil in real time and use advanced analysis tools to improve the accuracy and efficiency of soil fertility quality evaluation to support scientific agricultural management and decision-making.

[0004] The above information disclosed in this Background section is only for enhancement of understanding of the background of the present disclosure and therefore it may contain information that does not constitute the prior art that is already known to one of ordinary skill in the art. Summary of the invention

[0005] The purpose of the present invention is to provide a soil fertility quality analysis method and analysis system based on corn-growing farmland to solve the problems raised in the above-mentioned background technology.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] A soil fertility quality analysis method based on corn-growing farmland, the specific steps include:

[0008] Step 1: Soil samples are collected from the corn-planting farmland to be evaluated, with a sampling depth of 15-20 cm, visible-near infrared spectral images of the soil samples are collected, and reflectance data of key bands are extracted from the spectral images. The relative contents of nitrogen, phosphorus, and potassium, microbial biomass, and organic matter in the soil are calculated based on the reflected wave data, wherein the nitrogen, phosphorus, and potassium are available nitrogen, available phosphorus, and available potassium;

[0009] Step 2: Process the soil samples to obtain the soil conductivity, pH value and soil moisture content data of the soil samples, and calculate the soil health index based on the dimensionless conductivity, microbial biomass, pH value and relative content of organic matter, and calculate the nutrient balance index based on the dimensionless relative contents of nitrogen, phosphorus and potassium;

[0010] Step 3: Calculate the soil fertility quality comprehensive score for comprehensive evaluation of soil fertility quality based on the soil health index, nutrient balance index and soil moisture content;

[0011] Step 4: Compare the comprehensive score of soil fertility quality with the preset threshold, and evaluate the soil of the corn-planting arable land to be evaluated based on the comparison results;

[0012] Step 5: Introduce a machine learning model to train and predict historical soil data, build a quality analysis model based on the random forest algorithm, put the trained quality analysis model into use, and continuously learn new data to dynamically adjust and update it.

[0013] Furthermore, the key band refers to the band that can reflect the relative content of nitrogen, phosphorus, potassium, microbial biomass and organic matter in the soil in the spectral image;

[0014] The specific logic for obtaining soil conductivity is as follows:

[0015] The soil sample was dried naturally in a cool place, avoiding direct sunlight. The dried soil sample was sieved through a 2 mm sieve to remove large particles and impurities. 10 g of the soil sample was placed in a measuring cylinder, 100 ml of distilled water was added, and the mixture was stirred to make a soil suspension. The conductivity meter probe was inserted into the soil suspension, and the conductivity value was read after the instrument stabilized. It was recorded as ;

[0016] The relative content of nitrogen is calculated based on the reflected wave data according to the following formula:

[0017] ;

[0018] In the formula, is the relative content of nitrogen, is the reflectivity in the 970nm band, is the reflectivity in the 550nm band, is a constant used to reflect the effect of nitrogen content on reflectance ratio in a specific soil type;

[0019] The relative content of phosphorus is calculated based on the reflected wave data according to the following formula:

[0020] ;

[0021] In the formula, is the relative content of phosphorus, is the reflectivity in the 520nm band, is the reflectivity in the 550nm band, is a constant used to reflect the effect of phosphorus content on reflectance ratio in a specific soil type;

[0022] The relative potassium content is calculated based on the reflected wave data according to the following formula:

[0023] ;

[0024] In the formula, is the relative content of potassium, is the reflectivity in the 850nm band, is the reflectivity in the 550nm band, is a constant used to reflect the effect of potassium content on reflectance ratio in a specific soil type;

[0025] The relative content of organic matter is calculated based on the reflected wave data according to the following formula:

[0026] ;

[0027] In the formula, is the relative content of organic matter, reflecting the relative level of organic matter in the soil. is the reflectivity in the 750nm band, is the reference band, here is the reflectivity of the 550nm band, is a constant used to reflect the effect of organic matter content on reflectance ratio in a specific soil type;

[0028] The microbial biomass is calculated based on the reflected wave data according to the following formula:

[0029] ;

[0030] In the formula, Microbial biomass reflects the relative amount of organisms in the soil and is used to characterize the relative changes in microbial activity and abundance in different soil samples. is the reflectivity in the 1450nm band, is the reference band, here is the reflectivity of the 550nm band, is a constant used to reflect the effect of microbial biomass on reflectance ratio in a specific soil type;

[0031] Weigh the soil sample after removing impurities and record the wet weight, which is recorded as Then put the sample into the oven and set the temperature to 105℃ to dry it until the weight of the sample no longer changes. Record the dry weight and record it as , calculate the soil moisture content according to the following formula:

[0032] ;

[0033] In the formula, is the soil moisture content, is the wet weight of the sample, is the sample dry weight;

[0034] use Sensors to obtain soil Value, denoted as .

[0035] Furthermore, the organic matter content in the soil The organic matter content is compared with the preset threshold value. If it exceeds the threshold range, the difference between the organic matter content and the preset threshold value is calculated and recorded as ; If it does not exceed the threshold range, it will not be included in the subsequent calculation of the soil health index;

[0036] The soil conductivity is compared with the preset conductivity threshold. If it exceeds the threshold range, the difference between the conductivity and the preset threshold is calculated and recorded as ; If it does not exceed the threshold range, it will not be included in the subsequent calculation of the soil health index;

[0037] The soil health index is calculated based on the following formula:

[0038] ;

[0039] In the formula, is the soil health index, is the absolute difference between the relative content of organic matter and the threshold value of organic matter. That is the relative content value of the best organic matter for corn growth. is the absolute difference between conductivity and conductivity threshold, conductivity threshold This is the optimal conductivity value for corn growth. and Soil The maximum and minimum values ​​of , , Soil for planting value, Optimal for corn growth value; , , is the preset scaling factor, and ;

[0040] The nutrient balance index is calculated based on the following formula:

[0041] ;

[0042] In the formula, is the nutrient balance index, is the relative content of nitrogen, The optimum nitrogen content for corn growth is is the relative content of phosphorus, The optimal phosphorus content for corn growth is is the relative content of potassium, The optimal potassium content for corn growth. is the proportionality coefficient, and .

[0043] Further, a soil fertility quality comprehensive score for comprehensively evaluating soil fertility quality is calculated based on the soil health index, the nutrient balance index and the soil moisture content;

[0044] ;

[0045] In the formula, A comprehensive score for soil fertility quality. is a natural constant, is the soil health index, is the nutrient balance index, is the soil moisture content, The most suitable soil moisture content for corn planting. , , is the proportionality coefficient, obtained according to the analytic hierarchy process;

[0046] The specific logic for determining the proportionality coefficient by analytic hierarchy process is as follows:

[0047] Soil Health Index , nutrient balance index and The three indicators are marked, and the relative importance between the two indicators is determined by the nine-scale method to construct a judgment matrix. The index of is marked as 1, The index of is 2, The index mark is 3, and the constructed judgment matrix is:

[0048] ;

[0049] in, , denotes the index of the index, and , , Indicates that the index is The index relative to the index is The index importance of , the importance is scaled from 1 to 9, and The larger the value, the higher the index. The index is compared to the index The greater the importance of the index, , ;

[0050] Divide each element value in the judgment matrix by the sum of its columns to obtain a normalized judgment matrix. Calculate the mean of each row of element values ​​in the normalized judgment matrix, and use the mean of the first row of element values ​​as The weight of the second row is taken as The weight of the third row is taken as the mean of the elements in the third row. The weights are scaled in equal proportions with the constraint that the sum of the scaled values ​​is 1, and the scaled weights are used as the proportional coefficients of the corresponding exponents.

[0051] Furthermore, the soil fertility quality comprehensive score is compared with the preset threshold, based on the following specific logic:

[0052] like , it means that the soil fertility quality of the cultivated land is good, suitable for corn growth, and has high production potential;

[0053] like , it means that the soil fertility quality of the cultivated land is at a medium level and is within an acceptable range, and needs to be monitored regularly;

[0054] like , it means that the soil fertility quality of the cultivated land is poor and improvement measures are needed to improve the soil fertility quality;

[0055] In the formula, A comprehensive score for soil fertility quality. is the preset quality assessment threshold.

[0056] Furthermore, a machine learning model is introduced to train and predict historical soil data, and the performance indicators of the model are calculated, specifically:

[0057] Collect a batch of historical soil key parameters, including soil nitrogen, phosphorus, potassium concentration, microbial biomass, organic matter content, soil conductivity, pH value and soil moisture content data, calculate the soil fertility quality comprehensive score corresponding to each set of historical soil key parameters, and take each set of historical soil key parameters and the corresponding soil fertility quality comprehensive score as a set of data to form a data set;

[0058] The data set is divided into training set and test set, 70% as training set and 30% as test set;

[0059] Use Z-score normalization to transform each feature into a standard normal distribution:

[0060] ;

[0061] In the formula, is the standardized value, is the original value, is the mean of the feature, is the standard deviation of the feature, calculated as:

[0062] ;

[0063] ;

[0064] In the formula, is the mean of the feature, is the standard deviation of the feature, is the total number of observations of the feature, It is The observed value of a feature;

[0065] Build a random forest model and determine the feature vector , expressed as:

[0066] ;

[0067] in, is the soil moisture content, For soil value, is the conductivity, is the relative content of nitrogen, is the relative content of phosphorus, is the relative content of potassium, is the microbial biomass, is the relative content of organic matter;

[0068] Randomly select samples from the training set to generate There are decision trees, each of which randomly selects features for decision making when a node splits, and the prediction results of each tree are averaged to obtain the final prediction result:

[0069] ;

[0070] in, is the final prediction result, that is, the model prediction value, is the total number of decision trees constructed, is the index, It is The predicted output of each tree;

[0071] The mean square error and Coefficient of determination

[0072] ;

[0073] in, is the mean square error, is the number of samples in the test set, is the model prediction value, which is the above , is the true value, is the index of the sample in the test set;

[0074] ;

[0075] in, is the coefficient of determination, is the model prediction value, which is the above , is the true value, is the mean of the true values ​​in the test set, is the index of the sample in the test set.

[0076] The present invention further provides a soil fertility quality analysis system based on corn-planted farmland, wherein the soil fertility quality analysis system based on corn-planted farmland is used to execute the above-mentioned soil fertility quality analysis method based on corn-planted farmland, comprising:

[0077] The data acquisition module is used to collect soil samples from the corn-planting farmland to be evaluated, with a sampling depth of 15-20 cm, collect visible-near infrared spectral images of the soil samples, and extract reflectance data of key bands from the spectral images, and calculate the relative contents of nitrogen, phosphorus, and potassium, microbial biomass, and organic matter in the soil according to the reflected wave data, wherein the nitrogen, phosphorus, and potassium are fast-acting nitrogen, fast-acting phosphorus, and fast-acting potassium;

[0078] An index calculation module is used to process soil samples to obtain soil conductivity, pH value and soil moisture content data of the soil samples, and calculate the soil health index based on the conductivity, microbial biomass, pH value and relative content of organic matter after dimensionless processing, and calculate the nutrient balance index based on the relative contents of nitrogen, phosphorus and potassium after dimensionless processing;

[0079] A comprehensive scoring module, used to calculate a comprehensive score of soil fertility quality for comprehensive evaluation of soil fertility quality based on a soil health index, a nutrient balance index and soil moisture content;

[0080] An evaluation module is used to compare the comprehensive score of soil fertility quality with a preset threshold value, and evaluate the soil of the corn-planting arable land to be evaluated based on the comparison result;

[0081] The model building module is used to introduce machine learning models, train and predict historical soil data, build a quality analysis model based on the random forest algorithm, put the trained quality analysis model into use, and continuously learn new data to dynamically adjust and update it.

[0082] Compared with the prior art, the present invention has the following beneficial effects:

[0083] Through the systematic data collection and analysis process, the present invention not only simplifies the soil testing process, but also monitors the changes in soil fertility quality in real time, providing timely and reliable information support for agricultural managers to help them make scientific decisions. In addition, the use of machine learning models for dynamic adjustment ensures the continuous optimization of the evaluation model, adapts to the changes in different soil conditions and crop requirements, and improves the efficiency and sustainability of agricultural production. BRIEF DESCRIPTION OF THE DRAWINGS

[0084] Figure 1 It is a schematic diagram of the overall method flow of the present invention;

[0085] Figure 2 It is a schematic diagram of the system module of the present invention. DETAILED DESCRIPTION

[0086] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with specific embodiments.

[0087] It should be noted that, unless otherwise defined, the technical terms or scientific terms used in the present invention should be understood by people with ordinary skills in the field to which the present invention belongs. The words "first", "second" and similar words used in the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. "Include" or "comprise" and similar words mean that the elements or objects appearing before the word include the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connect" or "connected" and similar words are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0088] Example:

[0089] See also Figure 1 , the present invention provides a technical solution:

[0090] A soil fertility quality analysis method based on corn-growing farmland, the specific steps comprising:

[0091] Step 1: Soil samples are collected from the corn-planting farmland to be evaluated, with a sampling depth of 15-20 cm, visible-near infrared spectral images of the soil samples are collected, and reflectance data of key bands are extracted from the spectral images. The relative contents of nitrogen, phosphorus, and potassium, microbial biomass, and organic matter in the soil are calculated based on the reflected wave data, wherein the nitrogen, phosphorus, and potassium are available nitrogen, available phosphorus, and available potassium;

[0092] In this embodiment, the key band refers to the band that can reflect the relative content of nitrogen, phosphorus, potassium, microbial biomass and organic matter in the soil in the spectral image;

[0093] The specific logic for obtaining soil conductivity is as follows:

[0094] The soil sample was dried naturally in a cool place, avoiding direct sunlight. The dried soil sample was sieved through a 2 mm sieve to remove large particles and impurities. 10 g of the soil sample was placed in a measuring cylinder, 100 ml of distilled water was added, and the mixture was stirred to make a soil suspension. The conductivity meter probe was inserted into the soil suspension, and the conductivity value was read after the instrument stabilized. It was recorded as ;

[0095] The relative content of nitrogen is calculated based on the reflected wave data according to the following formula:

[0096] ;

[0097] In the formula, is the relative content of nitrogen, which is a relative value used to compare the nitrogen levels of different soil samples. is the reflectivity in the 970nm band, is the reflectivity in the 550nm band, is a constant that reflects the effect of nitrogen content on reflectance ratio in a specific soil type; the 970nm band is located in the near-infrared region and is closely related to the content of water and organic matter in the soil. Organic matter is an important carrier of nitrogen, so the reflectance of this band can be used to indirectly estimate the nitrogen content; the 550nm band is located in the visible light region and is related to the color and physical structure of the soil. It can be used as a reference band to eliminate the interference of light conditions, soil background characteristics, etc. on spectral data.

[0098] The relative content of phosphorus is calculated based on the reflected wave data according to the following formula:

[0099] ;

[0100] In the formula, is the relative content of phosphorus, which is a relative value used to compare the phosphorus levels of different soil samples. is the reflectivity in the 520nm band, is the reflectivity in the 550nm band, is a constant, which is used to reflect the influence of phosphorus content on reflectance ratio in a specific soil type; the 520nm band is located in the green light region of visible light, which is affected by phosphorus content and other related parameters in the soil, and can be used to indirectly estimate phosphorus content; the 550nm band is located in the green light region, which reflects soil color and background physical properties, and is used as a reference band to eliminate the influence of ambient light intensity and soil matrix differences;

[0101] The relative potassium content is calculated based on the reflected wave data according to the following formula:

[0102] ;

[0103] In the formula, is the relative content of potassium, which is a relative value used to compare the potassium levels of different soil samples. is the reflectivity in the 850nm band, is the reflectivity in the 550nm band, is a constant, which is used to reflect the influence of potassium content on reflectance ratio in a specific soil type; the 850nm band is located in the near-infrared region, which is related to soil mineral composition, water content and organic matter content, and potassium has an indirect correlation with these parameters, so it can be used to indirectly estimate potassium content; the 550nm band is located in the green light region, which mainly reflects soil color and background physical properties, and is used as a reference band to eliminate external influences;

[0104] The relative content of organic matter is calculated based on the reflected wave data according to the following formula:

[0105] ;

[0106] In the formula, is the relative content of organic matter, which is a relative value that reflects the relative level of organic matter in the soil. is the reflectivity in the 750nm band, is the reference band, here is the reflectivity of the 550nm band, is a constant used to reflect the effect of organic matter content on reflectance ratio in a specific soil type;

[0107] The microbial biomass is calculated based on the reflected wave data according to the following formula:

[0108] ;

[0109] In the formula, Microbial biomass reflects the relative amount of organisms in the soil. This relative amount is used to characterize the relative changes in microbial activity and abundance in different soil samples. is the reflectivity in the 1450nm band, is the reference band, here is the reflectivity of the 550nm band, is a constant, which is used to reflect the effect of microbial biomass on reflectance ratio in a specific soil type; the 1450nm band is a typical water absorption band in the near-infrared region, which is usually related to soil water content and organic matter content, while soil organic matter is often positively correlated with microbial biomass, so this band is used as an indirect indicator of microbial activity; the 550nm band is located in the visible light region and is related to soil color and mineral composition. It is a reference band used to eliminate interference with spectral signals caused by ambient light, differences in soil physical properties, etc.

[0110] Weigh the soil sample after removing impurities and record the wet weight, which is recorded as Then put the sample into the oven and set the temperature to 105℃ to dry it until the weight of the sample no longer changes. Record the dry weight and record it as , calculate the soil moisture content according to the following formula:

[0111] ;

[0112] In the formula, is the soil moisture content, is the wet weight of the sample, is the sample dry weight;

[0113] use Sensors to obtain soil Value, denoted as .

[0114] The advantage of step 1 is that through spectral imaging technology and the extraction of reflectance data, efficient and rapid determination of nitrogen, phosphorus, potassium concentrations, microbial biomass and organic matter content in the soil is achieved. Compared with traditional soil testing methods, this method can not only significantly shorten the analysis time, but also improve the accuracy and consistency of data acquisition. At the same time, spectral analysis can provide more detailed information, help identify subtle changes in soil properties, and provide more reliable basic data for subsequent soil health assessments, overcoming the limitations of traditional technologies in the determination of multiple indicators.

[0115] In this plan, the adoption of this step can effectively promote the implementation of the overall plan. By systematically acquiring key data, step 1 lays a solid foundation for the subsequent calculation of soil health index, nutrient balance index and comprehensive score. Its efficient data collection and processing capabilities make the entire soil fertility quality analysis and evaluation process more scientific and reasonable, and can timely reflect changes in soil status, providing strong support for agricultural management and decision-making. In addition, based on accurate soil composition data, the machine learning model introduced later will be able to better train and predict, further improving the intelligence level of the plan.

[0116] Step 2: Process the soil samples to obtain the soil conductivity, pH value and soil moisture content data of the soil samples, and calculate the soil health index based on the dimensionless conductivity, microbial biomass, pH value and relative content of organic matter, and calculate the nutrient balance index based on the dimensionless relative contents of nitrogen, phosphorus and potassium;

[0117] In this embodiment, the soil health index is calculated based on the following formula:

[0118] ;

[0119] In the formula, is the soil health index, is the absolute difference between the organic matter content and the organic matter threshold. This is the optimal organic matter content for corn growth. is the absolute difference between conductivity and conductivity threshold, conductivity threshold That is the optimal conductivity value for corn growth; and Soil The maximum and minimum values ​​of , , Soil for planting value, Optimal for corn growth value; , , is the preset scaling factor, and This is because organic matter is one of the key factors in soil health, and it has an important impact on soil structure, fertility and microbial activity, so its corresponding proportional coefficient It is set to the maximum, emphasizing the dominant role of organic matter in soil health; electrical conductivity reflects the concentration of salt in the soil and also has an important impact on plant growth, but its impact is relatively minor compared to organic matter; soil pH affects nutrient availability and microbial activity, but its impact is generally considered to be minimal.

[0120] when When it increases, it means that the deviation between the organic matter content and the threshold value increases. Too much or too little organic matter may lead to reduced soil nutrient supply capacity and poor plant growth. will decrease; when When it increases, it means that the deviation of conductivity from the preset threshold increases, which will affect the growth of plant roots, causing dehydration and nutrient absorption disorders in plants. will decrease; when When the value deviates from the optimal range, The value will decrease, reflecting the decline in soil health. Soil acidity or alkalinity may inhibit plant growth and affect nutrient absorption, so when When increasing, will decrease; that is, , , and There is a negative correlation.

[0121] The nutrient balance index is calculated based on the following formula:

[0122] ;

[0123] In the formula, is the nutrient balance index, is the relative content of nitrogen, The optimum nitrogen content for corn growth is is the relative content of phosphorus, The optimal phosphorus content for corn growth is is the relative content of potassium, The optimal potassium content for corn growth. is the proportionality coefficient, and .

[0124] when When it increases, it means that the nitrogen supply in the soil is far from the optimal demand for corn cultivation, which may lead to limited corn growth, affecting crop health and yield, thus causing decrease; similarly, when or When it increases, it will lead to Decrease; that is, , , There is a negative correlation.

[0125] The advantage of step 2 is that by simultaneously acquiring soil conductivity, pH value and soil moisture content data, combined with dimensionless processing, it can comprehensively evaluate soil health and nutrient balance. This multi-parameter comprehensive evaluation method is more effective than the traditional single indicator determination, can more accurately reflect the actual state of the soil, and avoids the misjudgment that may be caused by relying on a single indicator. By combining multiple key indicators for analysis, more reliable and comprehensive soil fertility quality information is provided, providing a solid foundation for subsequent decision-making.

[0126] Step 2 provides rich data support for building a soil fertility quality assessment model and generating a comprehensive soil fertility quality score. This process not only enhances the calculation accuracy of the soil health index and nutrient balance index, but also provides sufficient training data for subsequent machine learning models, thereby improving the intelligence level and predictive ability of the entire system. In addition, the comprehensive soil fertility quality analysis provides a data basis for corn planting management, which helps to achieve more refined agricultural production decisions and promote the development of sustainable agriculture.

[0127] Step 3: Calculate the soil fertility quality comprehensive score for comprehensive evaluation of soil fertility quality based on the soil health index, nutrient balance index and soil moisture content;

[0128] In this embodiment, the soil fertility quality comprehensive score is calculated according to the following formula:

[0129] ;

[0130] In the formula, A comprehensive score for soil fertility quality. is a natural constant, is the soil health index, is the nutrient balance index, is the soil moisture content, The most suitable soil moisture content for corn planting. , , is the proportionality coefficient, obtained according to the analytic hierarchy process;

[0131] It takes into account the soil's health, nutrient balance and moisture content, and represents the overall quality level of the soil. The higher the value, the better the soil fertility quality and the more suitable it is for crop growth; When it increases, it means that the soil structure, fertility and biological activity have improved. will increase accordingly; Low values ​​indicate a large imbalance between the soil's nutrient supply and crop demand, usually meaning that the soil is either deficient or excessive in nutrients. will also decrease; when When it increases, it means that the deviation between the soil moisture content and the ideal value increases. Too much or too little moisture will have a negative impact on the quality of soil fertility. will decrease; that is, , and There is a positive correlation, and There is a negative correlation.

[0132] The specific logic for determining the proportionality coefficient by analytic hierarchy process is as follows:

[0133] Soil Health Index , nutrient balance index and The three indicators are marked, and the relative importance between the two indicators is determined by the nine-scale method to construct a judgment matrix. The index of is marked as 1, The index of is 2, The index mark is 3, and the constructed judgment matrix is:

[0134] ;

[0135] in, , denotes the index of the index, and , , Indicates that the index is The index relative to the index is The index importance of , the importance is scaled from 1 to 9, and The larger the value, the higher the index. The index is compared to the index The greater the importance of the index, , ;

[0136] Divide each element value in the judgment matrix by the sum of its columns to obtain a normalized judgment matrix. Calculate the mean of each row of element values ​​in the normalized judgment matrix, and use the mean of the first row of element values ​​as The weight of the second row is taken as The weight of the third row is taken as the mean of the elements in the third row. The weights are scaled in equal proportions with the constraint that the sum of the scaled values ​​is 1, and the scaled weights are used as the proportional coefficients of the corresponding exponents.

[0137] The advantage of step 3 is that it constructs a comprehensive soil fertility quality assessment model by combining the soil health index and the nutrient balance index. This comprehensive assessment not only takes into account the interrelationships between multiple key indicators, but also introduces the correction of soil moisture content, which improves the accuracy and adaptability of the model. Compared with existing technologies, this multi-dimensional assessment method can more comprehensively reflect the actual health status of the soil, avoid the one-sided analysis that may be caused by a single indicator, and thus provide a more scientific decision-making basis for agricultural management.

[0138] In this solution, the adoption of step 3 can effectively promote the implementation of the overall solution. By establishing a systematic soil fertility quality assessment model, step 3 provides an accurate framework for the subsequent comprehensive scoring of soil fertility quality, thereby ensuring the scientificity and consistency of the entire assessment process. In addition, this step provides a clear and reliable data basis for the introduction of machine learning models, so that subsequent dynamic adjustments and updates can be based on accurate soil fertility quality assessment results, which helps to improve the level of intelligent agricultural management and promote the development of precision agriculture.

[0139] Step 4: Compare the comprehensive score of soil fertility quality with the preset threshold, and evaluate the soil of the corn-planting arable land to be evaluated based on the comparison results;

[0140] In this embodiment, the soil fertility quality comprehensive score is compared with a preset threshold, and the specific logic is as follows:

[0141] like , it means that the soil fertility quality of the cultivated land is good, suitable for corn growth, and has high production potential;

[0142] like , it means that the soil fertility quality of the cultivated land is at a medium level and is within an acceptable range, and needs to be monitored regularly;

[0143] like , it means that the soil fertility quality of the cultivated land is poor and improvement measures are needed to improve the soil fertility quality;

[0144] In the formula, A comprehensive score for soil fertility quality. It is a preset quality assessment threshold, and its specific value is determined based on historical data and expert evaluation.

[0145] The advantage of step 4 is that by comparing the comprehensive score of soil fertility quality with the preset threshold, the actual health status of the soil can be effectively evaluated. This process not only provides a clear classification standard, but also provides a basis for corresponding management decisions. Compared with existing technologies, this evaluation method based on comprehensive scoring is more intuitive and easy to implement, which can help farmers quickly understand the soil status and take corresponding measures in time.

[0146] In this plan, the adoption of step 4 can effectively promote the implementation of the overall plan. Through the dynamic evaluation of the comprehensive score of soil fertility quality, step 4 provides a reliable basis for subsequent agricultural management and decision-making. This process ensures that farmers can adjust management measures such as fertilization and irrigation in a timely manner according to actual conditions, thereby improving the growth potential and yield of corn. In addition, this evaluation mechanism also helps to achieve refined agricultural management, promote the development of sustainable agriculture, and make more reasonable use and protection of soil resources.

[0147] Step 5: Introduce a machine learning model to train and predict historical soil data, build a quality analysis model based on the random forest algorithm, put the trained quality analysis model into use, and continuously learn new data to dynamically adjust and update it;

[0148] In this embodiment, a machine learning model is introduced to train and predict historical soil data, and the performance indicators of the model are calculated, specifically:

[0149] Collect a batch of historical soil key parameters, including soil nitrogen, phosphorus, potassium concentration, microbial biomass, organic matter content, soil conductivity, pH value and soil moisture content data, calculate the soil fertility quality comprehensive score corresponding to each set of historical soil key parameters, and take each set of historical soil key parameters and the corresponding soil fertility quality comprehensive score as a set of data to form a data set;

[0150] The data set is divided into training set and test set, 70% as training set and 30% as test set;

[0151] Use Z-score normalization to transform each feature into a standard normal distribution:

[0152] ;

[0153] In the formula, is the standardized value, is the original value, is the mean of the feature, is the standard deviation of the feature, calculated as:

[0154] ;

[0155] ;

[0156] In the formula, is the mean of the feature, is the standard deviation of the feature, is the total number of observations of the feature, It is The observed value of a feature;

[0157] Build a random forest model and determine the feature vector , expressed as:

[0158] ;

[0159] in, is the soil moisture content, For soil value, is the conductivity, is the relative content of nitrogen, is the relative content of phosphorus, is the relative content of potassium, is the microbial biomass, is the relative content of organic matter;

[0160] Randomly select samples from the training set to generate There are decision trees, each of which randomly selects features for decision making when a node splits, and the prediction results of each tree are averaged to obtain the final prediction result:

[0161] ;

[0162] in, is the final prediction result, that is, the model prediction value, is the total number of decision trees constructed, is the index, It is The predicted output of each tree;

[0163] The mean square error and Coefficient of determination

[0164] ;

[0165] in, is the mean square error, is the number of samples in the test set, is the model prediction value, which is the above , is the true value, is the index of the sample in the test set;

[0166] ;

[0167] in, is the coefficient of determination, is the model prediction value, which is the above , is the true value, is the mean of the true values ​​in the test set, is the index of the sample in the test set.

[0168] The advantage of step 5 is that it introduces a machine learning model that can perform efficient training and prediction based on historical soil data, improving the accuracy and intelligence of soil fertility quality analysis. Through the application of the random forest algorithm, this step can use multi-dimensional data for in-depth analysis, capture and optimize complex soil characteristics, overcome the limitations of traditional methods in data processing and model building, and provide more reliable prediction results.

[0169] Compared with the existing technology, the beneficial effect of step 5 is that it can dynamically update the analysis model, adapt to changes in soil fertility quality by continuously learning new data, and improve the scientificity and timeliness of management decisions. This process enables precise agricultural production management, ensuring that growers make more scientific fertilization, irrigation and other field management measures based on real-time data, thereby effectively improving the growth potential and crop yield of corn. In this solution, the use of step 5 can effectively promote the implementation of the overall solution. By introducing machine learning, step 5 not only enhances the real-time and reliability of soil fertility quality assessment, but also provides continuous support and basis for subsequent decision-making. This innovative method will provide technical support for the sustainable development of agriculture, make the management of soil resources more scientific and efficient, and promote the further development of precision agriculture.

[0170] See also Figure 2 , a soil fertility quality analysis system based on corn-growing cultivated land, comprising:

[0171] The data acquisition module is used to collect soil samples from the corn-planting farmland to be evaluated, with a sampling depth of 15-20 cm, collect visible-near infrared spectral images of the soil samples, and extract reflectance data of key bands from the spectral images, and calculate the relative contents of nitrogen, phosphorus, and potassium, microbial biomass, and organic matter in the soil according to the reflected wave data, wherein the nitrogen, phosphorus, and potassium are fast-acting nitrogen, fast-acting phosphorus, and fast-acting potassium;

[0172] An index calculation module is used to process soil samples to obtain soil conductivity, pH value and soil moisture content data of the soil samples, and calculate the soil health index based on the conductivity, microbial biomass, pH value and relative content of organic matter after dimensionless processing, and calculate the nutrient balance index based on the relative contents of nitrogen, phosphorus and potassium after dimensionless processing;

[0173] A comprehensive scoring module, used to calculate a comprehensive score of soil fertility quality for comprehensive evaluation of soil fertility quality based on a soil health index, a nutrient balance index and soil moisture content;

[0174] An evaluation module is used to compare the comprehensive score of soil fertility quality with a preset threshold value, and evaluate the soil of the corn-planting arable land to be evaluated based on the comparison result;

[0175] The model building module is used to introduce machine learning models, train and predict historical soil data, build a quality analysis model based on the random forest algorithm, put the trained quality analysis model into use, and continuously learn new data to dynamically adjust and update it.

[0176] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0177] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination thereof. When implemented by software, the above embodiments may be implemented in whole or in part in the form of a computer program product. Those skilled in the art may appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein may be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed by hardware or software methods depends on the specific application and design constraints of the technical solution.

[0178] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, and may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0179] The above description is only a specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application.

Claims

1. A soil fertility quality analysis method based on corn-growing farmland, characterized in that: The specific steps include: Step 1: Soil samples are collected from the corn-planting farmland to be evaluated, with a sampling depth of 15-20 cm, visible-near infrared spectral images of the soil samples are collected, and reflectance data of key bands are extracted from the spectral images. The relative contents of nitrogen, phosphorus, and potassium, microbial biomass, and organic matter in the soil are calculated based on the reflected wave data, wherein the nitrogen, phosphorus, and potassium are available nitrogen, available phosphorus, and available potassium; Step 2: Process the soil samples to obtain the soil conductivity, pH value and soil moisture content data of the soil samples, and calculate the soil health index based on the dimensionless conductivity, microbial biomass, pH value and relative content of organic matter, and calculate the nutrient balance index based on the dimensionless relative contents of nitrogen, phosphorus and potassium; Step 3: Calculate the soil fertility quality comprehensive score for comprehensive evaluation of soil fertility quality based on the soil health index, nutrient balance index and soil moisture content; Step 4: Compare the comprehensive score of soil fertility quality with the preset threshold, and evaluate the soil of the corn-planting arable land to be evaluated based on the comparison results; Step 5: Introduce a machine learning model to train and predict historical soil data, build a quality analysis model based on the random forest algorithm, put the trained quality analysis model into use, and continuously learn new data to dynamically adjust and update it; The soil health index is calculated based on the following formula: Where SHI is the soil health index, |OM-OM opt | is the absolute difference between the relative content of organic matter and the organic matter threshold, the organic matter threshold OM opt That is the relative content value of the optimal organic matter for corn growth, |EC-EC opt | is the absolute difference between conductivity and conductivity threshold, conductivity threshold EC opt This is the optimal conductivity value for corn growth, pH max and pH min are the maximum and minimum values ​​of soil pH, respectively. max =8.0, pH min =5.0, pH is the pH value of the planting soil, pH opt is the optimal pH value for corn growth; α, β, and γ are preset proportional coefficients, and α>β>γ>0; the formula for calculating the nutrient balance index is as follows: In the formula, NBI is the nutrient balance index, N is the relative content of nitrogen, opt is the optimal nitrogen content for corn growth, P is the relative content of phosphorus, and P opt is the optimal phosphorus content for corn growth, K is the relative content of potassium, and K opt The optimal potassium content for corn growth. is the proportionality coefficient, and Based on the soil health index, nutrient balance index and soil moisture content, the soil fertility quality comprehensive score for comprehensive evaluation of soil fertility quality is calculated: In the formula, SOI is the comprehensive score of soil fertility quality, e is a natural constant, SHI is the soil health index, NBI is the nutrient balance index, S is the soil moisture content, and S opt is the most suitable soil moisture content for corn planting, ω1, ω2, and ω3 are proportional coefficients, which are obtained according to the analytic hierarchy process.

2. The method for analyzing soil fertility quality based on corn-growing farmland according to claim 1, characterized in that: The key band refers to the band that can reflect the relative content of nitrogen, phosphorus, potassium, microbial biomass and organic matter in the soil in the spectral image; The specific logic for obtaining soil conductivity is as follows: The soil samples were dried naturally in a cool place, avoiding direct sunlight. The dried soil samples were sieved through a 2 mm mesh to remove large particles and impurities. 10 g of soil sample was placed in a measuring cylinder, 100 ml of distilled water was added, and the mixture was stirred to make a soil suspension. The conductivity meter probe was inserted into the soil suspension, and the conductivity value was read after the instrument was stabilized, which was recorded as EC. The relative content of nitrogen is calculated based on the reflected wave data according to the following formula: Where N is the relative content of nitrogen, P N is the reflectivity in the 970nm band, R green is the reflectivity in the 550nm band, k N is a constant, which is used to reflect the effect of soil nitrogen content on reflectance ratio; The relative content of phosphorus is calculated based on the reflected wave data according to the following formula: In the formula, P is the relative content of phosphorus, R P is the reflectivity in the 520nm band, R green is the reflectivity in the 550nm band, k P is a constant, which is used to reflect the effect of soil phosphorus content on reflectance ratio; The relative potassium content is calculated based on the reflected wave data according to the following formula: In the formula, K is the relative content of potassium, R K is the reflectivity in the 850nm band, R green is the reflectivity in the 550nm band, k K is a constant, which is used to reflect the effect of potassium content in soil on reflectance ratio; The relative content of organic matter is calculated based on the reflected wave data according to the following formula: In the formula, OM is the relative content of organic matter, R OM is the reflectivity in the 750nm band, R ref is the reference band, here is the reflectivity of the 550nm band, k OM is a constant, which is used to reflect the effect of soil organic matter content on reflectance ratio; The microbial biomass is calculated based on the reflected wave data according to the following formula: In the formula, MB is microbial biomass, which reflects the relative amount of organisms in the soil and is used to characterize the relative changes in microbial activity and abundance in different soil samples. MB is the reflectivity in the 1450nm band, R ref is the reference band, here is the reflectivity of the 550nm band, k MB is a constant, which is used to reflect the effect of soil microbial biomass on the reflectance ratio; Weigh the soil sample after removing impurities and record the wet weight, which is recorded as W wet Then put the sample into the oven and set the temperature to 105℃ to dry it until the weight of the sample no longer changes. Record the dry weight and record it as W dry , calculate the soil moisture content according to the following formula: Where S is the soil moisture content, W wet is the wet weight of the sample, W dry is the sample dry weight; Use a pH sensor to obtain the pH value of the planting soil, recorded as pH.

3. The soil fertility quality analysis method based on corn-growing farmland according to claim 1, characterized in that: The specific logic for determining the proportionality coefficient by analytic hierarchy process is as follows: The soil health index SHI, nutrient balance index NBI and (SS opt ) 2 The three indicators are marked, and the relative importance between them is determined by the nine-scale method to construct a judgment matrix, in which the index of SHI is marked as 1, the index of NBI is marked as 2, and (SS opt ) 2 The index mark is 3, and the constructed judgment matrix is: Where f and v both represent indexes of the exponent, and f∈[1,3], v∈[1,3], b fv Indicates the importance of the index with index f relative to the index with index v. The importance is scaled from 1 to 9, and b fv The larger the value, the more important the index f is compared to the index v, and b ff =1, Divide each element value in the judgment matrix by the sum of its columns to obtain the normalized judgment matrix. Calculate the mean of each row of element values ​​in the normalized judgment matrix, and use the mean of the first row of element values ​​as the weight of SHI, the mean of the second row of element values ​​as the weight of NBI, and the mean of the third row of element values ​​as (SS opt ) 2 The weights are scaled in equal proportions with the constraint that the sum of the scaled values ​​is 1, and the scaled weights are used as the proportional coefficients of the corresponding exponents.

4. The method for analyzing soil fertility quality based on corn-growing farmland according to claim 1, characterized in that: The soil fertility quality comprehensive score is compared with the preset threshold, based on the following specific logic: If SOI>SOI threshold , it means that the soil fertility quality of the cultivated land is good, suitable for corn growth, and has high production potential; If 0.5*SOI threshold <SOI≤SOI threshold , it means that the soil fertility quality of the cultivated land is at a medium level and is within an acceptable range, and needs to be monitored regularly; If SOI≤0.5*SOI threshold , it means that the soil fertility quality of the cultivated land is poor and improvement measures are needed to improve the soil fertility quality; In the formula, SOI is the comprehensive score of soil fertility quality, SOI threshold is the preset quality assessment threshold.

5. The method for analyzing soil fertility quality based on corn-growing farmland according to claim 1, characterized in that: A machine learning model was introduced to train and predict historical soil data, and the performance indicators of the model were calculated, specifically: Collect a batch of historical soil key parameters, including soil nitrogen, phosphorus, potassium concentration, microbial biomass, organic matter content, soil conductivity, pH value and soil moisture content data, calculate the soil fertility quality comprehensive score corresponding to each set of historical soil key parameters, and take each set of historical soil key parameters and the corresponding soil fertility quality comprehensive score as a set of data to form a data set; Divide the data set into training set and test set, 70% as training set and 30% as test set; Use Z-score normalization to transform each feature into a standard normal distribution: In the formula, Z is the standardized value, X is the original value, μ is the mean of the feature, and σ is the standard deviation of the feature. The calculation formula is: Where μ is the mean of the feature, σ is the standard deviation of the feature, N is the total number of observations of the feature, and X i is the observed value of the i-th feature; Construct a random forest model and determine the feature vector Y, which is expressed as: Y=[S,pH,EC,N,P,K,MB,OM] Among them, S is soil moisture content, pH is soil pH, EC is electrical conductivity, N is the relative content of nitrogen, P is the relative content of phosphorus, K is the relative content of potassium, MB is microbial biomass, and OM is the relative content of organic matter; Randomly select samples from the training set to generate T decision trees. Each tree randomly selects features to make decisions when the node splits. The prediction results of each tree are averaged to obtain the final prediction result: Among them, Prediction is the final prediction result, that is, the model prediction value, T is the total number of decision trees constructed, and t is the index. is the predicted output of the tth tree; The mean square error and R are calculated according to the following formula: 2 Coefficient of determination Among them, MSE is the mean square error, N test is the number of samples in the test set, is the model prediction value, which is the above-mentioned Prediction, y k is the true value, k is the index of the sample in the test set; Among them, R 2 is the coefficient of determination, is the model prediction value, which is the above-mentioned Prediction, y k is the true value, is the mean of the true values ​​in the test set, and k is the index of the sample in the test set.

6. A soil fertility quality analysis system based on corn-growing farmland, characterized by: The soil fertility quality analysis system based on corn-planting farmland is used to execute the soil fertility quality analysis method based on corn-planting farmland described in any one of claims 1 to 5, comprising: The data acquisition module is used to collect soil samples from the corn-planting farmland to be evaluated, with a sampling depth of 15-20 cm, collect visible-near infrared spectral images of the soil samples, and extract reflectance data of key bands from the spectral images, and calculate the relative contents of nitrogen, phosphorus, and potassium, microbial biomass, and organic matter in the soil according to the reflected wave data, wherein the nitrogen, phosphorus, and potassium are fast-acting nitrogen, fast-acting phosphorus, and fast-acting potassium; An index calculation module is used to process soil samples to obtain soil conductivity, pH value and soil moisture content data of the soil samples, and calculate the soil health index based on the conductivity, microbial biomass, pH value and relative content of organic matter after dimensionless processing, and calculate the nutrient balance index based on the relative contents of nitrogen, phosphorus and potassium after dimensionless processing; A comprehensive scoring module, used to calculate a comprehensive score of soil fertility quality for comprehensive evaluation of soil fertility quality based on a soil health index, a nutrient balance index and soil moisture content; An evaluation module is used to compare the comprehensive score of soil fertility quality with a preset threshold value, and evaluate the soil fertility of the corn-planting arable land to be evaluated based on the comparison result; The model building module is used to introduce machine learning models, train and predict historical soil data, build a quality analysis model based on the random forest algorithm, put the trained quality analysis model into use, and continuously learn new data to dynamically adjust and update it.

Images