A remote sensing multispectral imaging system for agricultural monitoring

Abstract

The application provides a remote sensing multispectral imaging system for agricultural monitoring, and relates to the technical field of space remote sensing.The application realizes accurate selection by determining an economic crop through an indication economic crop determination module, analyzing historical soil salt content and water shortage data, and selecting a sensitive crop; the correlation evaluation module calculates a water shortage index change rate, analyzes the correlation between historical and current data, and improves the prediction accuracy; the value range determination module divides soil salt content intervals and combines evaluation coefficient standardized management data; the indication area determination module uses multispectral images to realize accurate monitoring and enhance representativeness; the humidity evaluation module analyzes local spectral data to quantify soil humidity; the prediction model construction module uses machine learning to predict salt content, and can accurately predict the dynamic changes of soil salt content and moisture; the correlation analysis module combines the prediction model to check real-time monitoring data; and the division module optimizes soil management strategies according to the water shortage index and humidity coefficient.

Description

Technical Field

[0001] This invention relates to the field of aerospace remote sensing technology, specifically to a remote sensing multispectral imaging system for agricultural monitoring. Background Technology

[0002] With the increasing demand for precision and intelligent development in agricultural production, traditional ground monitoring methods can no longer meet the requirements for large-scale, rapid, and accurate data acquisition. Multispectral imaging technology, by capturing information from multiple spectral bands, can provide richer and more detailed information about ground features than single-spectral imaging technology, which helps to analyze the specific conditions of elements such as soil, water, and vegetation in depth.

[0003] The existing technology, with publication number CN117274585A, entitled "A Semantic Segmentation Method for Agricultural Multimodal Remote Sensing Images Based on Transformer Channel Feature Enhancement," includes the following steps: S1: Divide the agricultural semantic segmentation image dataset; S2: Define the preprocessing scheme for data reading; S3: Define a random weight sampler and a distributed data loader; S4: Load data from the data loader and start feature extraction using an encoder-decoder architecture model; S5: Enhance the fused features in the decoder using a transformer-based channel enhancement module; S6: Calculate the model's prediction map xpred,xc for each pixel of the image; S7: Calculate the loss function; S8: Repeat steps S2 to S7 until model training is complete.

[0004] Existing agricultural remote sensing monitoring systems have several shortcomings. First, traditional methods typically rely on ground sampling and experimental analysis, which is not only time-consuming and labor-intensive but also difficult to achieve large-scale real-time monitoring. Second, current multispectral imaging systems often lack full utilization of historical data, making it difficult to accurately predict dynamic changes in soil salinity and moisture, resulting in insufficient scientific rigor and effectiveness in agricultural decision-making. Furthermore, existing technologies have limitations in the correlation analysis between soil and crop physiological states, failing to fully explore the potential relationships between historical and current monitoring data.

[0005] The information disclosed in the background section above is only intended to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0006] The purpose of this invention is to provide a remote sensing multispectral imaging system for agricultural monitoring, so as to solve the problems mentioned in the background art.

[0007] To achieve the above objectives, the present invention provides the following technical solution:

[0008] A remote sensing multispectral imaging system for agricultural monitoring, specifically comprising:

[0009] Indicator crop identification module: used to acquire historical soil salinity change data of the target monitoring area in the previous year, as well as soil water shortage indicators corresponding to the historical soil salinity change data; based on the historical soil salinity change data, to identify the indicator crops that are most sensitive to soil salinity changes in the target monitoring area.

[0010] The correlation evaluation module is used to calculate the rate of change of soil water shortage index between two consecutive months in the previous year, on a monthly basis, and to conduct a correlation analysis on the historical soil salinity change data and the rate of change of soil water shortage index to obtain the correlation evaluation coefficient.

[0011] Value range determination module: Based on historical soil salinity change data, the soil salinity level is divided into several intervals, and the value range of the soil water shortage index within each soil salinity level interval is determined by combining the correlation evaluation coefficient and the change rate of the soil water shortage index.

[0012] Indicator Area Determination Module: After acquiring multispectral image data of the target monitoring area, it preprocesses the data and determines the most representative indicator area in the image data based on historical soil salinity change data and correlation evaluation coefficients, as well as the reference multispectral data corresponding to the indicator area. The indicator area is used to plant the determined indicator economic crop, and the indicator area can represent the soil salinity level of the target monitoring area.

[0013] Humidity assessment module: This module receives preprocessed reference multispectral data, extracts local spectral data from the reference multispectral data for soil moisture assessment, analyzes and processes the local spectral data, and constructs a soil moisture assessment coefficient. The soil moisture assessment coefficient is used to evaluate the humidity of the target monitoring area.

[0014] Prediction model building module: It is used to obtain multispectral data corresponding to historical soil salinity change data of indicator economic crops, process the data, build a machine learning model, and then input the processed multispectral data into the machine learning model for training and validation, and finally generate a salinity content prediction model.

[0015] The correlation analysis module is used to input the preprocessed reference multispectral data into the salt content prediction model to obtain the prediction results of soil salinity level. The prediction results are then matched with several divided soil salinity level intervals to determine the soil water shortage index of the current target monitoring area.

[0016] The soil classification module is used to construct a soil classification model by combining the soil water shortage index and soil moisture evaluation coefficient of the current target monitoring area. This model is used to classify the soil moisture demand level in order to generate farmland management strategies.

[0017] Furthermore, the rate of change of soil water shortage index between adjacent two months was calculated sequentially, specifically including:

[0018] Calculate the rate of change of soil moisture content, field capacity, permanent wilting point, and soil water tension for each month compared to the next month; let the current month be j and the next month be j+1, and the formula for the rate of change is as follows:

[0019]

[0020] in, This indicates the rate of change of soil water shortage indicators. This represents the soil water shortage index value for month j. This represents the soil water shortage index value for month j+1.

[0021] Historical soil salinity change data were paired with the monthly change rates of various soil water shortage indicators;

[0022] Calculate the Pearson correlation coefficient between historical soil salinity change data and the rate of change of various soil water shortage indicators;

[0023] The Pearson correlation coefficients calculated for each soil water deficit index are used to form a correlation evaluation coefficient matrix FJ:

[0024]

[0025] in, These are the Pearson correlation coefficients between soil salinity changes and the corresponding rates of change in soil moisture content, field capacity, permanent wilting point, and soil water tension.

[0026] Based on historical soil salinity variation data, soil salinity levels are divided into several intervals, including:

[0027] Soil is divided into three ranges based on the percentage of salinity: low salinity (0-10%), medium salinity (10%-15%), and high salinity (15%-100%).

[0028] By combining the correlation evaluation coefficient matrix FJ and the rate of change of each soil water shortage index, the value range of the soil water shortage index within each soil salinity level interval is determined, specifically including:

[0029] The formula for calculating the soil water deficit index is set as follows: Linear regression analysis was performed using historical data to obtain the regression coefficients for each soil water shortage index. and intercept YFp represents the salinity level.

[0030] The correlation coefficients in the correlation evaluation coefficient matrix FJ After standardization, the following calculation formulas are obtained:

[0031]

[0032] Will These are used as weighting coefficients for soil moisture content, field water holding capacity, permanent wilting point, and soil water tension, respectively.

[0033] The formula for calculating soil moisture content is as follows:

[0034]

[0035] in, These are the regression coefficients and intercepts for soil moisture content, respectively; YFp is calculated using the two endpoints between the current salinity zones.

[0036] The formula for calculating the value of field water holding capacity is as follows:

[0037]

[0038] in, These are the regression coefficients and intercepts for field water holding capacity, respectively.

[0039] The formula for calculating the value of the permanent withering point is:

[0040]

[0041] in, These are the regression coefficients and intercepts for the permanent wilting point, respectively;

[0042] The formula for calculating the value of soil moisture tension is as follows;

[0043]

[0044] in, These are the regression coefficients and intercepts of soil moisture tension, respectively.

[0045] Furthermore, a machine learning model is constructed, and the processed multispectral data is then input into the machine learning model for training and validation, ultimately generating a salinity prediction model, which specifically includes:

[0046] Extract key spectral features from historical multispectral data, including but not limited to: vegetation index, soil regulation index, and red edge position;

[0047] Based on the time series data, the rate of change of spectral characteristics and the rate of change of soil salinity at each time point are calculated to form time series characteristics;

[0048] A multi-source data fusion method was selected to fuse spectral features and soil salinity features; a multilayer perceptron neural network was used as the data fusion model.

[0049] The extracted spectral features and soil salinity features are input into an MLP neural network for feature fusion.

[0050] Train the MLP neural network using historical data;

[0051] The trained model is validated using a validation set, and the model's prediction accuracy and error are calculated.

[0052] Adjust the parameters of the MLP neural network based on the validation results;

[0053] The optimized MLP neural network model is saved as the final salt content prediction model.

[0054] Furthermore, the preprocessed reference multispectral data is substituted into the salinity prediction model to obtain the predicted soil salinity level. This predicted result is then matched with several defined soil salinity level intervals to determine the soil water shortage index for the current target monitoring area, specifically including:

[0055] The predicted salt content is matched with the pre-divided salt intervals to determine the salt level of the current target monitoring area;

[0056] Based on the percentage of soil salinity, the predicted salinity is matched with low salinity range (0, 10%), medium salinity range [10%-15%], and high salinity range (15%-100%) to determine the corresponding salinity range.

[0057] Based on the confirmed salinity range, the range of values ​​for soil water shortage indicators, including soil moisture content, field capacity, permanent wilting point, and soil water tension, is determined.

[0058] Furthermore, a soil classification model is constructed by combining soil water shortage indicators and soil moisture evaluation coefficients for the current target monitoring area. This model is used to classify soil moisture requirement levels to generate farmland management strategies, specifically including:

[0059] The calculation formula for the soil partitioning model is defined as follows:

[0060]

[0061] Where H is the soil moisture evaluation coefficient. These are soil water deficit indicators, including soil moisture content, field capacity, permanent wilting point, and soil water tension. For the division value;

[0062] Based on the calculated soil water shortage index Based on the soil moisture evaluation coefficient H, soil moisture requirements are divided into three levels: low moisture requirement, medium moisture requirement, and high moisture requirement.

[0063] For low moisture requirement levels;

[0064] When H < 0.35, and the soil salinity range is low (0-10%);

[0065] No additional irrigation or simple maintenance measures are required;

[0066] For medium moisture requirement levels;

[0067] when When the soil salinity range is in the medium salinity range [10%-15%];

[0068] Irrigate according to actual needs to prevent soil water shortage or over-irrigation;

[0069]

[0070] in, and These represent the maximum and minimum irrigation amounts for the medium water demand level, respectively. For water demand;

[0071] For high moisture requirement levels;

[0072] when Furthermore, the soil salinity range is high (15%-100%).

[0073] Take proactive irrigation measures to maintain soil moisture, pay attention to salt accumulation, and implement soil improvement measures.

[0074]

[0075] in, The maximum irrigation volume for high water demand levels. The percentage increase in salt accumulation.

[0076] Compared with the prior art, the beneficial effects of the present invention are: by using the indicator economic crop determination module to acquire and analyze historical soil salinity change data and soil water shortage indicators, the most sensitive indicator economic crops can be determined, thereby achieving precise crop selection and improving the targeting of monitoring;

[0077] The correlation evaluation module calculates the rate of change of soil water shortage index and obtains the correlation evaluation coefficient, enabling correlation analysis between historical data and current monitoring data to enhance prediction accuracy. The value range determination module divides soil salinity levels into intervals and determines the value range of the index in combination with the evaluation coefficient, achieving standardized management and improving data processing efficiency.

[0078] The indicator area determination module identifies representative areas based on multispectral image data, enabling precise monitoring and improving data representativeness; the humidity evaluation module analyzes local spectral data to construct humidity evaluation coefficients, enabling quantitative assessment of soil moisture and enhancing the accuracy of water management.

[0079] The prediction model building module utilizes machine learning to process multispectral data and establishes a salt content prediction model to achieve dynamic prediction and improve foresight. The correlation analysis module combines the prediction model and standard intervals to determine the current soil water shortage index, enabling real-time monitoring and prediction verification and enhancing system reliability.

[0080] By dividing the soil into modules and combining water shortage indicators and humidity evaluation coefficients, a soil classification model is constructed to classify soil moisture demand levels, optimize farmland management strategies, and improve resource utilization efficiency. Attached Figure Description

[0081] Figure 1 This is a schematic diagram of the overall system flow of the present invention;

[0082] Figure 2 This is a system module block diagram of the present invention. Detailed Implementation

[0083] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to specific embodiments.

[0084] It should be noted that, unless otherwise defined, the technical or scientific terms used in this invention should have the ordinary meaning understood by one of ordinary skill in the art to which this invention pertains. The terms "first," "second," and similar terms used in this invention do not indicate any order, quantity, or importance, but are merely used to distinguish different components. Terms such as "comprising" or "including" mean that the element or object preceding the word encompasses the elements or objects listed following the word and their equivalents, without excluding other elements or objects. Terms such as "connected" or "linked" are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. Terms such as "upper," "lower," "left," and "right" are used only to indicate relative positional relationships; when the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0085] Example 1:

[0086] Please see Figure 1 and Figure 2 The present invention provides a technical solution:

[0087] A remote sensing multispectral imaging system for agricultural monitoring, specifically comprising:

[0088] Indicator crop identification module: used to acquire historical soil salinity change data of the target monitoring area in the previous year, as well as soil water shortage indicators corresponding to the historical soil salinity change data; based on the historical soil salinity change data, to identify the indicator crops that are most sensitive to soil salinity changes in the target monitoring area.

[0089] The correlation evaluation module is used to calculate the rate of change of soil water shortage index between two consecutive months in the previous year, on a monthly basis, and to conduct a correlation analysis on the historical soil salinity change data and the rate of change of soil water shortage index to obtain the correlation evaluation coefficient.

[0090] Value range determination module: Based on historical soil salinity change data, the soil salinity level is divided into several intervals, and the value range of the soil water shortage index within each soil salinity level interval is determined by combining the correlation evaluation coefficient and the change rate of the soil water shortage index.

[0091] Indicator Area Determination Module: After acquiring multispectral image data of the target monitoring area, it preprocesses the data and determines the most representative indicator area in the image data based on historical soil salinity change data and correlation evaluation coefficients, as well as the reference multispectral data corresponding to the indicator area. The indicator area is used to plant the determined indicator economic crop, and the indicator area can represent the soil salinity level of the target monitoring area.

[0092] Humidity assessment module: This module receives preprocessed reference multispectral data, extracts local spectral data from the reference multispectral data for soil moisture assessment, analyzes and processes the local spectral data, and constructs a soil moisture assessment coefficient. The soil moisture assessment coefficient is used to evaluate the humidity of the target monitoring area.

[0093] Prediction model building module: It is used to obtain multispectral data corresponding to historical soil salinity change data of indicator economic crops, process the data, build a machine learning model, and then input the processed multispectral data into the machine learning model for training and validation, and finally generate a salinity content prediction model.

[0094] The correlation analysis module is used to input the preprocessed reference multispectral data into the salt content prediction model to obtain the prediction results of soil salinity level. The prediction results are then matched with several divided soil salinity level intervals to determine the soil water shortage index of the current target monitoring area.

[0095] The soil classification module is used to construct a soil classification model by combining the soil water shortage index and soil moisture evaluation coefficient of the current target monitoring area. This model is used to classify the soil moisture demand level in order to generate farmland management strategies.

[0096] To further explain, soil salinity change data for the previous year in the target monitoring area were directly extracted from the agricultural monitoring database. The obtained historical soil salinity change data and soil water shortage indicators were preprocessed. The data preprocessing included data cleaning, normalization, and interpolation for subsequent analysis.

[0097] Data cleaning: removing noise and outliers from data;

[0098] Data normalization: scaling data to a standard range of 0 to 1 to facilitate comparisons between different data sources;

[0099] Data interpolation: Interpolating missing data to ensure data integrity.

[0100] Soil water deficit indicators include soil moisture content, field capacity, permanent wilting point, and soil water tension, which are respectively labeled as SMC, FC, PWP, and SWT.

[0101] Soil moisture content: indicates the proportion of water in the soil, expressed as a percentage;

[0102] Field holding capacity: The maximum amount of water that soil can hold under gravity, expressed as a volume percentage;

[0103] Permanent wilting point: The lowest soil moisture content at which plants can no longer absorb water from the soil, expressed as a percentage by volume;

[0104] Soil moisture tension: the attractive force of water in the soil, specifically comparing the current soil moisture tension with the maximum moisture tension, expressed as a percentage in Pascals (Pa).

[0105] The Pearson correlation coefficient was used to calculate the correlation between soil salinity changes and corresponding multispectral data of economic crops; this was done to determine the sensitivity of historical soil salinity changes to the growth of different economic crops, specifically including:

[0106] Extract multispectral data for each economic crop in the target monitoring area;

[0107] The Pearson correlation coefficient between historical soil salinity variation data and corresponding multispectral data of economic crops is calculated using the following formula:

[0108]

[0109] in, It is the first Data points on soil salinity changes, It is the first Each economic crop corresponds to a multispectral data point. and These are the average values ​​of soil salinity variation data and the corresponding multispectral data of economic crops.

[0110] Based on the correlation analysis results, the economic crop that is positively correlated with soil salinity changes and has the largest correlation coefficient was selected as the indicator economic crop.

[0111] The correlation coefficients of each economic crop were ranked.

[0112] Select the economic crop with the largest absolute value of the correlation coefficient;

[0113] The economic crop has been identified as an indicator economic crop.

[0114] The selected indicator economic crops were validated using the model to ensure their sensitivity to changes in soil salinity. In actual monitoring, the model parameters were adjusted through periodic validation and feedback to further improve the accuracy of the model.

[0115] Detailed operation steps:

[0116] Verify the accuracy of forecasts for the selected indicator economic crops by monitoring data from the current year;

[0117] Based on the validation results, the model parameters were adjusted and feedback optimization was performed.

[0118] The selection of indicator economic crops includes a first category and a second category;

[0119] The first category selects a single type of indicator economic crop, whose multispectral data can be adapted to different soil salinity levels.

[0120] The second category selects multiple types of indicator economic crops and analyzes the multispectral data of specific indicator economic crops under different soil salinity levels.

[0121] To further explain, the rate of change of soil water shortage index between adjacent two months was calculated sequentially, specifically including:

[0122] Calculate the rate of change of soil moisture content, field capacity, permanent wilting point, and soil water tension for each month compared to the next month; let the current month be j and the next month be j+1, and the formula for the rate of change is as follows:

[0123]

[0124] in, This indicates the rate of change of soil water shortage indicators. This represents the soil water shortage index value for month j. This represents the soil water shortage index value for month j+1.

[0125] The rate of change can intuitively reflect the fluctuation of various soil water shortage indicators over time.

[0126] Historical soil salinity change data were paired with the monthly change rates of various soil water shortage indicators;

[0127] Calculate the Pearson correlation coefficient between historical soil salinity change data and the rate of change of various soil water shortage indicators; the calculation formula is the same as that used in the indicator crop identification module, and will not be repeated here:

[0128] The Pearson correlation coefficients calculated for each soil water deficit index are used to form a correlation evaluation coefficient matrix FJ:

[0129]

[0130] in, These are the Pearson correlation coefficients between soil salinity changes and the corresponding rates of change in soil moisture content, field capacity, permanent wilting point, and soil water tension.

[0131] Based on historical soil salinity variation data, soil salinity levels are divided into several intervals, including:

[0132] Soil is divided into three ranges based on the percentage of salinity: low salinity (0-10%), medium salinity (10%-15%), and high salinity (15%-100%).

[0133] By combining the correlation evaluation coefficient matrix FJ and the rate of change of each soil water shortage index, the range of values ​​for each soil water shortage index within each soil salinity level interval is determined. In this embodiment, linear regression or other statistical methods are used to calculate the range of change of each soil water shortage index under different soil salinity levels.

[0134] The range of values ​​for the soil water shortage index calculated using linear regression is as follows:

[0135] The formula for calculating the soil water deficit index is set as follows: Linear regression analysis was performed using historical data to obtain the regression coefficients for each soil water shortage index. and intercept YFp represents the salinity level.

[0136] The correlation coefficients in the correlation evaluation coefficient matrix FJ After standardization, the following calculation formulas are obtained:

[0137]

[0138] Will These are used as weighting coefficients for soil moisture content, field water holding capacity, permanent wilting point, and soil water tension, respectively.

[0139] The formula for calculating soil moisture content is as follows:

[0140]

[0141] in, These are the regression coefficients and intercepts for soil moisture content, respectively; YFp is calculated using the two endpoints between the current salinity zones.

[0142] For example, in the low salinity range (0, 10%), YFp is set to 0 and 10% respectively, resulting in two calculations. The results of the two calculations will be used as the range of values ​​for the current soil moisture content; the same applies below, and will not be elaborated further.

[0143] The formula for calculating the value of field water holding capacity is as follows:

[0144]

[0145] in, These are the regression coefficients and intercepts for field water holding capacity, respectively.

[0146] The formula for calculating the value of the permanent withering point is:

[0147]

[0148] in, These are the regression coefficients and intercepts for the permanent wilting point, respectively;

[0149] The formula for calculating the value of soil moisture tension is as follows;

[0150]

[0151] in, These are the regression coefficients and intercepts of soil moisture tension, respectively;

[0152] For low salinity ranges (0, 10%), irrigation measures are taken to maintain soil moisture;

[0153] Based on the experimental data, the regression coefficients and intercepts were determined, and the following value ranges were obtained;

[0154] Soil moisture content should be within the range of 4.5% ≤ SMC ≤ 5.2%;

[0155] Field holding capacity (FHC) should be taken in the range of 3.5% ≤ FC ≤ 4.0%;

[0156] The permanent wilting point is taken in the range of 3.0% ≤ PWP ≤ 3.5%;

[0157] Soil moisture tension should be within the range of 5.5% ≤ SWT ≤ 6.0%;

[0158] For the medium salinity range [10%-15%], irrigation should be carried out according to actual needs to prevent soil water shortage or over-irrigation; at the same time, based on the soil water shortage index range determined in the low salinity range, the soil moisture content value should be...

[0159] Soil moisture content should be within the range of 4.8% ≤ SMC ≤ 5.4%;

[0160] Field water holding capacity (FHC) is taken in the range of 4.0% ≤ FC ≤ 4.5%;

[0161] The permanent wilting point is taken in the range of 3.5% ≤ PWP ≤ 4.0%;

[0162] Soil moisture tension values ​​should be within the range of 5.8% ≤ SWT ≤ 6.2%;

[0163] For high salinity ranges (15%-100%), attention should be paid to soil salt accumulation, and soil improvement measures should be taken, such as increasing organic matter and improving irrigation techniques, to improve soil structure and water retention capacity.

[0164] Soil moisture content should be within the range of 5.0% ≤ SMC ≤ 5.6%;

[0165] Field holding capacity (FHC) is taken in the range of 4.5% ≤ FC ≤ 5.0%;

[0166] The permanent wilting point is taken in the range of 4.0% ≤ PWP ≤ 4.5%;

[0167] Soil moisture tension should be within the range of 6.0% ≤ SWT ≤ 6.5%;

[0168] The mapping relationship is shown in Table 1 below:

[0169] Table 1: Mapping Relationship Table

[0170]

[0171] Based on the above mapping relationship, the range of soil water shortage index values ​​for the target monitoring area is determined under different soil salinity levels:

[0172] Low salinity levels: When soil salinity is below 10%, soil moisture content should be maintained between 4.5% and 5.2%, field capacity should be between 3.5% and 4.0%, permanent wilting point should be between 3.0% and 3.5%, and soil water tension should be between 5.5% and 6.0%. Under these conditions, irrigation measures can be taken to maintain soil moisture.

[0173] Medium salinity level: When the soil salinity level is between 10% and 15%, the soil moisture content should be controlled between 4.8% and 5.4%, the field capacity should be between 4.0% and 4.5%, the permanent wilting point should be between 3.5% and 4.0%, and the soil water tension should be maintained between 5.8% and 6.2%. At this time, appropriate irrigation should be carried out according to actual needs to prevent soil water shortage or over-irrigation.

[0174] High salinity levels: When soil salinity is above 15%, soil moisture content should be maintained between 5.0% and 5.6%, field capacity should be between 4.5% and 5.0%, permanent wilting point should be between 4.0% and 4.5%, and soil water tension should be between 6.0% and 6.5%. In this case, it is necessary to focus on soil salt accumulation and take necessary soil improvement measures, such as increasing organic matter and improving irrigation techniques, to improve soil structure and water retention capacity.

[0175] To further explain, after acquiring multispectral image data of the target monitoring area, preprocessing is performed. Based on historical soil salinity change data and correlation evaluation coefficients, the most representative indicator area is determined in the image data, along with the corresponding reference multispectral data. This indicator area is used for planting the identified indicator economic crops, specifically including:

[0176] Use remote sensing satellites to acquire multispectral image data of the target monitoring area;

[0177] Multispectral images include multiple bands such as red, green, blue, and near-infrared, to facilitate subsequent image processing and analysis;

[0178] The acquired multispectral image data is preprocessed, including radiometric correction, geometric correction, and denoising operations, to ensure the accuracy and consistency of the image data; radiometric correction, geometric correction, and denoising operations are existing technologies and will not be described in detail.

[0179] Radiometric correction eliminates the influence of sensors and the atmosphere on image data, ensuring that the reflectance of each pixel representing a ground feature is accurate.

[0180] Geometric correction is the process of eliminating geometric distortions caused by changes in sensor angle or terrain undulations during image acquisition, so that the image is aligned with the actual geographical location.

[0181] Denoising is the process of removing noise from an image to enhance its clarity and usability.

[0182] By analyzing preprocessed multispectral images using historical soil salinity change data, current soil salinity level information can be extracted.

[0183] Soil salinity index is calculated by analyzing the near-infrared and red bands in multispectral images. The calculation formula is as follows:

[0184]

[0185] in, Here, R represents the near-infrared reflectance, and SSI represents the red reflectance; this formula is used to quantify the soil salinity level of each pixel in an image.

[0186] Obtain the correlation evaluation coefficient matrix FJ. Based on the soil salinity index and the correlation evaluation coefficient matrix FJ, determine the most representative indicator area within the target monitoring area, specifically including:

[0187] The target monitoring area is divided into several small areas, each containing several pixels;

[0188] For each pixel within a small region, calculate the weighted average of the soil salinity index:

[0189]

[0190] in, This represents the soil salinity index of the i-th pixel. Indicates the corresponding weights, This is a weighted average of the soil salinity index;

[0191] The soil water deficit index for each small area is calculated using a weighted average:

[0192]

[0193] in, This represents the soil water shortage index value of the i-th pixel. Indicates the corresponding weights, This is a weighted average of soil water shortage indicators; Based on the experimental data determined by the expert panel, further details will not be provided.

[0194] By combining the weighted average of soil salinity index and soil water deficiency index, and using the correlation evaluation coefficient matrix FJ, the comprehensive score for each sub-region is calculated:

[0195]

[0196] Where S represents the overall score. This is a weighted average of the soil salinity index; The weighted average of soil water shortage indicators is used; the small area with the highest comprehensive score is selected as the indicator area that best represents the soil salinity level and water shortage status of the target monitoring area.

[0197] To further explain, a soil moisture evaluation coefficient is constructed, which is used to evaluate the moisture conditions of the target monitoring area; specifically, it includes:

[0198] Local spectral data for soil moisture assessment are extracted from the preprocessed reference multispectral data, specifically including:

[0199] Spectral bands sensitive to soil moisture were selected for analysis and processing;

[0200] Choose the red-edge band, shortwave infrared band, and near-infrared band; these bands are more sensitive to changes in soil moisture.

[0201] The reflectance values ​​of the red-edge band, short-wave infrared band, and near-infrared band are extracted from the reference multispectral data and used as local spectral data.

[0202] The soil moisture index was calculated using the extracted local spectral data;

[0203]

[0204] in, The red-edge band represents the reflectivity, while SWIR represents the reflectivity in the short-wave infrared band. RED represents the reflectance in the near-infrared band, and a1, a2, and a3 are all positive adjustment coefficients. a1, a2, and a3 are adjusted based on actual conditions and experimental data.

[0205] The indicator area is divided into several smaller areas, each containing several pixels;

[0206] For each pixel within a small region, calculate the weighted average of the soil moisture index:

[0207]

[0208] in, This represents the soil moisture index at the u-th pixel. Indicates the corresponding weight; Based on the experimental data determined by the expert panel, further details will not be provided.

[0209] The soil moisture evaluation coefficient is defined as H, and the calculation formula is as follows:

[0210]

[0211] Where b1 and b2 are both positive adjustment coefficients, This is the weighted average of the soil moisture index. b1 and b2 are the average rainfall in the target monitoring area for the current month; b1 and b2 are determined by the expert group based on experimental data.

[0212] Based on the value of the soil moisture evaluation coefficient H, the humidity conditions of the target monitoring area are divided into low humidity, medium humidity, and high humidity levels; specifically including:

[0213] The effective range of H is limited to (0,1). If the value of H is in the range of (0,0.35), the humidity of the target monitoring area is low humidity; the corresponding average rainfall is between 0 and A1 mm, excluding A1.

[0214] If the value of H is in the range of [0.35, 0.64), the humidity of the target monitoring area is moderate; the corresponding average rainfall is between A1 and A2 mm.

[0215] If the value of H is in the range of [0.64, 1), the humidity of the target monitoring area is high humidity; the corresponding average rainfall exceeds A2 mm.

[0216] To further explain, a machine learning model is constructed, and then the processed multispectral data is input into the machine learning model for training and validation, ultimately generating a salinity prediction model, which specifically includes:

[0217] Using Geographic Information System (GIS) tools, spatial interpolation and resampling techniques were employed to spatially register multispectral data and soil salinity data, ensuring spatial consistency between the two datasets.

[0218] Extract key spectral features from historical multispectral data, including but not limited to: vegetation index, soil regulation index, and red edge position;

[0219] Based on the time series data, the rate of change of spectral characteristics and the rate of change of soil salinity at each time point are calculated to form time series characteristics;

[0220] A multi-source data fusion method was selected to fuse spectral features and soil salinity features; a multilayer perceptron (MLP) neural network was used as the data fusion model; MLP was chosen because it can handle complex nonlinear relationships and has good generalization ability.

[0221] The extracted spectral features and soil salinity features are input into the MLP neural network for feature fusion. The number of nodes in the input layer of the network is the total number of spectral features and soil salinity features. The number of nodes in the hidden layer is optimized based on experimental results. The number of nodes in the output layer is the predicted value of soil salinity.

[0222] Historical data was used to train the MLP neural network; cross-validation was used to divide the dataset into training and validation sets, with the training set used to train the model and the validation set used to evaluate the model performance; mean squared error (MSE) was chosen as the loss function and the Adam optimizer was chosen as the optimization algorithm;

[0223] The trained model is validated using a validation set, and its prediction accuracy and error are calculated. The formulas are as follows:

[0224]

[0225] in, For the sample size, For the first The actual value of each sample For the first Predicted values ​​for each sample;

[0226] Based on the validation results, adjust the parameters of the MLP neural network, including the number of hidden layer nodes, learning rate, and number of training rounds, to gradually optimize model performance.

[0227] The optimized MLP neural network model was saved as the final salinity prediction model. This model can process new multispectral data in real time and predict the soil salinity of the corresponding area.

[0228] To further explain, the preprocessed reference multispectral data is substituted into the salinity prediction model to obtain the predicted soil salinity level. This predicted result is then matched with several defined soil salinity level intervals to determine the soil water shortage index for the current target monitoring area, specifically including:

[0229] The predicted salt content is matched with the pre-divided salt intervals to determine the salt level of the current target monitoring area;

[0230] Specifically, based on the percentage of soil salinity, the predicted salinity is matched with low salinity range (0, 10%), medium salinity range [10%-15%], and high salinity range (15%-100%) to determine the corresponding salinity range.

[0231] Based on the confirmed salinity range, the range of values ​​for soil water shortage indicators, including soil moisture content, field capacity, permanent wilting point, and soil water tension, is determined.

[0232] To further explain, a soil classification model is constructed by combining soil water shortage indicators and soil moisture evaluation coefficients for the current target monitoring area. This model is used to classify soil moisture requirement levels in order to generate farmland management strategies, specifically including:

[0233] The calculation formula for the soil partitioning model is defined as follows:

[0234]

[0235] Where H is the soil moisture evaluation coefficient. These are soil water deficit indicators, including soil moisture content, field capacity, permanent wilting point, and soil water tension. For the division value;

[0236] Based on the calculated soil water shortage index ( Based on the soil moisture evaluation coefficient H, soil moisture requirements are divided into three levels: low moisture requirement level, medium moisture requirement level and high moisture requirement level.

[0237] For low moisture requirement levels;

[0238] When H < 0.35, and the soil salinity range is low (0-10%);

[0239] No additional irrigation or simple maintenance measures are required;

[0240] For medium moisture requirement levels;

[0241] when When the soil salinity range is in the medium salinity range [10%-15%];

[0242] Irrigate according to actual needs to prevent soil water shortage or over-irrigation;

[0243]

[0244] in, and These represent the maximum and minimum irrigation amounts for the medium water demand level, respectively. For water demand;

[0245] For high moisture requirement levels;

[0246] when Furthermore, the soil salinity range is high (15%-100%).

[0247] Take proactive irrigation measures to maintain soil moisture, pay attention to salt accumulation, and implement soil improvement measures.

[0248]

[0249] in, The maximum irrigation volume for high water demand levels. The percentage increase in salt accumulation.

[0250] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0251] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented in software, the above embodiments can be implemented, in whole or in part, as a computer program product. Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution.

[0252] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment, depending on actual needs.

[0253] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application.

Claims

1. A remote sensing multispectral imaging system for agricultural monitoring, characterized in that, Specifically, it includes: Indicator crop identification module: used to acquire historical soil salinity change data of the target monitoring area in the previous year, as well as soil water shortage indicators corresponding to the historical soil salinity change data; based on the historical soil salinity change data, to identify the indicator crops that are most sensitive to soil salinity changes in the target monitoring area. The correlation evaluation module is used to calculate the rate of change of soil water shortage index between two consecutive months in the previous year, on a monthly basis, and to conduct a correlation analysis on the historical soil salinity change data and the rate of change of soil water shortage index to obtain the correlation evaluation coefficient. Value range determination module: Based on historical soil salinity change data, the soil salinity level is divided into several intervals, and the value range of the soil water shortage index within each soil salinity level interval is determined by combining the correlation evaluation coefficient and the change rate of the soil water shortage index. Indicator Area Determination Module: After acquiring multispectral image data of the target monitoring area, it preprocesses the data and determines the most representative indicator area in the image data based on historical soil salinity change data and correlation evaluation coefficients, as well as the reference multispectral data corresponding to the indicator area. The indicator area is used to plant the determined indicator economic crop, and the indicator area can represent the soil salinity level of the target monitoring area. Humidity assessment module: This module receives preprocessed reference multispectral data, extracts local spectral data from the reference multispectral data for soil moisture assessment, analyzes and processes the local spectral data, and constructs a soil moisture assessment coefficient. The soil moisture assessment coefficient is used to evaluate the humidity of the target monitoring area. Prediction model building module: It is used to obtain multispectral data corresponding to historical soil salinity change data of indicator economic crops, process the data, build a machine learning model, and then input the processed multispectral data into the machine learning model for training and validation, and finally generate a salinity content prediction model. The correlation analysis module is used to input the preprocessed reference multispectral data into the salt content prediction model to obtain the prediction results of soil salinity level. The prediction results are then matched with several divided soil salinity level intervals to determine the soil water shortage index of the current target monitoring area. The soil classification module is used to construct a soil classification model by combining the soil water shortage index and soil moisture evaluation coefficient of the current target monitoring area. This model is used to classify the soil moisture demand level in order to generate farmland management strategies.

2. The remote sensing multispectral imaging system for agricultural monitoring according to claim 1, characterized in that: Data preprocessing was performed on the acquired historical soil salinity change data and soil water shortage indicators; Soil water deficit indicators include soil moisture content, field capacity, permanent wilting point, and soil water tension, which are respectively labeled as SMC, FC, PWP, and SWT. The correlation between soil salinity changes and corresponding multispectral data of economic crops was calculated using the Pearson correlation coefficient. To determine the correlation between historical soil salinity changes and the growth sensitivity of different economic crops, specifically including: Extract multispectral data for each economic crop in the target monitoring area; Calculate the Pearson correlation coefficient between historical soil salinity variation data and corresponding multispectral data of economic crops; Based on the correlation analysis results, the economic crop that is positively correlated with soil salinity changes and has the largest correlation coefficient was selected as the indicator economic crop. The selection of indicator economic crops includes a first category and a second category; The first category selects a single type of indicator economic crop, whose multispectral data can be adapted to different soil salinity levels. The second category selects multiple types of indicator economic crops and analyzes the multispectral data of specific indicator economic crops under different soil salinity levels.

3. The remote sensing multispectral imaging system for agricultural monitoring according to claim 2, characterized in that: Calculate the rate of change of soil water deficit index between two consecutive two months, specifically including: Calculate the rate of change of soil moisture content, field capacity, permanent wilting point, and soil water tension for each month compared to the next month; let the current month be j and the next month be j+1, and the formula for the rate of change is as follows: Where ΔZB represents the rate of change of the soil water shortage index, ZB j ZB represents the soil water shortage index value for month j. j+1 This represents the soil water shortage index value for month j+1. Historical soil salinity change data were paired with the monthly change rates of various soil water shortage indicators; Calculate the Pearson correlation coefficient between historical soil salinity change data and the rate of change of various soil water shortage indicators; The Pearson correlation coefficients calculated for each soil water deficit index are used to form a correlation evaluation coefficient matrix FJ: Where, r SMC r FC r PWP r SWT These are the Pearson correlation coefficients between soil salinity changes and the corresponding rates of change in soil moisture content, field capacity, permanent wilting point, and soil water tension. Based on historical soil salinity variation data, soil salinity levels are divided into several intervals, including: Soil is divided into three ranges based on the percentage of salinity: low salinity (0-10%), medium salinity (10%-15%), and high salinity (15%-100%). By combining the correlation evaluation coefficient matrix FJ and the rate of change of each soil water shortage index, the value range of the soil water shortage index within each soil salinity level interval is determined, specifically including: The formula for calculating the soil water deficit index is set as follows: Linear regression analysis was performed using historical data to obtain the regression coefficient a′ZB and intercept b′ZB for each soil water shortage index; YFp represents the salinity level. The correlation coefficient r in the correlation evaluation coefficient matrix FJ SMC r FC r PWP r SWT After standardization, the following calculation formulas are obtained: w SMC w FC w PWP w SWT These are used as weighting coefficients for soil moisture content, field water holding capacity, permanent wilting point, and soil water tension, respectively. The formula for calculating soil moisture content is as follows: SMC bh =w SMC ×(a SMC ×YFp+b SMC ) Among them, a SMC b SMC These are the regression coefficients and intercepts for soil moisture content, respectively; YFp is calculated using the two endpoints between the current salinity zones. The formula for calculating the value of field water holding capacity is as follows: FC bh =w FC ×(a FC ×YFp+b FC ) Among them, a FC b FC These are the regression coefficients and intercepts for field water holding capacity, respectively. The formula for calculating the value of the permanent withering point is: PWP bh =w PWP ×(a PWP ×YFp+b PWP ) Among them, a PWP b PWP These are the regression coefficients and intercepts for the permanent wilting point, respectively; The formula for calculating the value of soil moisture tension is as follows; SWT bh =w SWT ×(a SWT ×YFp+b SWT ) Among them, a SWT b SWT These are the regression coefficients and intercepts of soil moisture tension, respectively.

4. A remote sensing multispectral imaging system for agricultural monitoring according to claim 3, characterized in that: Use remote sensing satellites to acquire multispectral image data of the target monitoring area; By analyzing preprocessed multispectral images using historical soil salinity change data, current soil salinity level information can be extracted. Soil salinity index is calculated by analyzing the near-infrared and red bands in multispectral images. The calculation formula is as follows: Wherein, NIR is the near-infrared reflectance, R is the red reflectance, and SSI is the soil salinity index; Obtain the correlation evaluation coefficient matrix FJ. Based on the soil salinity index and the correlation evaluation coefficient matrix FJ, determine the most representative indicator area within the target monitoring area, specifically including: The target monitoring area is divided into several small areas, each containing several pixels; For each pixel within a small region, calculate the weighted average of the soil salinity index: The soil water deficit index for each small area is calculated using a weighted average: By combining the weighted average of soil salinity index and soil water deficiency index, and using the correlation evaluation coefficient matrix FJ, the comprehensive score for each sub-region is calculated: Where S represents the overall score, SSI avg ZB is the weighted average of soil salinity index. avg The weighted average of soil water shortage indicators is used; the small area with the highest comprehensive score is selected as the indicator area that best represents the soil salinity level and water shortage status of the target monitoring area.

5. A remote sensing multispectral imaging system for agricultural monitoring according to claim 4, characterized in that: A soil moisture assessment coefficient is constructed to evaluate the moisture conditions of the target monitoring area. Specifically, this includes: Local spectral data for soil moisture assessment are extracted from the preprocessed reference multispectral data, and the soil moisture index is calculated using the extracted local spectral data. The indicator area is divided into several smaller areas, each containing several pixels; For each pixel within a small region, calculate the weighted average of the soil moisture index: The soil moisture evaluation coefficient is defined as H, and the calculation formula is as follows: Where b1 and b2 are both positive adjustment coefficients, SMI avg Jy is the weighted average of the soil moisture index, and Jy is the average rainfall in the target monitoring area in the current month; the effective range of H is limited to (0,1). Based on the value of the soil moisture evaluation coefficient H, the humidity conditions of the target monitoring area are divided into low humidity, medium humidity and high humidity.

6. A remote sensing multispectral imaging system for agricultural monitoring according to claim 5, characterized in that: A machine learning model is constructed, and then the processed multispectral data is input into the machine learning model for training and validation, ultimately generating a salinity prediction model, which specifically includes: Extract key spectral features from historical multispectral data, including but not limited to: vegetation index, soil regulation index, and red edge position; Based on the time series data, the rate of change of spectral characteristics and the rate of change of soil salinity at each time point are calculated to form time series characteristics; A multi-source data fusion method was selected to fuse spectral features and soil salinity features; a multilayer perceptron neural network was used as the data fusion model. The extracted spectral features and soil salinity features are input into an MLP neural network for feature fusion. Train the MLP neural network using historical data; The trained model is validated using a validation set, and the model's prediction accuracy and error are calculated. Adjust the parameters of the MLP neural network based on the validation results; The optimized MLP neural network model is saved as the final salt content prediction model.

7. A remote sensing multispectral imaging system for agricultural monitoring according to claim 6, characterized in that: The preprocessed reference multispectral data is substituted into the salinity prediction model to obtain the predicted soil salinity level. This prediction is then matched with several defined soil salinity level intervals to determine the soil water shortage index for the current target monitoring area, specifically including: The predicted salt content is matched with the pre-divided salt intervals to determine the salt level of the current target monitoring area; Based on the percentage of soil salinity, the predicted salinity is matched with low salinity range (0, 10%), medium salinity range [10%-15%], and high salinity range (15%-100%) to determine the corresponding salinity range. Based on the confirmed salinity range, the range of values ​​for soil water shortage indicators, including soil moisture content, field capacity, permanent wilting point, and soil water tension, are determined.

8. A remote sensing multispectral imaging system for agricultural monitoring according to claim 7, characterized in that: A soil classification model is constructed by combining soil water shortage indicators and soil moisture evaluation coefficients for the current target monitoring area. This model is used to classify soil moisture requirement levels to generate farmland management strategies, specifically including: The calculation formula for the soil partitioning model is defined as follows: WDL=f(SMC bh ,FC bh ,PWP bh ,SWT bh ,H) Where H is the soil moisture evaluation coefficient, SMC bh FC bh PWP bh SWT bh These are soil water deficit indicators, including soil moisture content, field capacity, permanent wilting point, and soil water tension, with WDL being the classification value. Based on the calculated soil water deficit index {SMC bh FC bh PWP bh SWT bh Based on the soil moisture evaluation coefficient H, soil moisture requirements are divided into three levels: low moisture requirement, medium moisture requirement, and high moisture requirement. For low moisture requirement levels; When H < 0.35, and the soil salinity range is low (0-10%); No additional irrigation or simple maintenance measures are required; For medium moisture requirement levels; When 0.35≤H<0.64, and the soil salinity range is in the medium salinity range [10%-15%]; Irrigate according to actual needs to prevent soil water shortage or over-irrigation; Among them, MWD max and MWD min These represent the maximum and minimum irrigation amounts for the medium water demand level, respectively, with Water Amount being the required water volume. For high moisture requirement levels; When H ≥ 0.64, and the soil salinity range is the high salinity range (15%-100%); Take proactive irrigation measures to maintain soil moisture, pay attention to salt accumulation, and implement soil improvement measures. Water Amount=HWD max ×(1+Salt increase ) Among them, HWD max Salt is the maximum irrigation volume for high water demand levels. increase The percentage increase in salt accumulation.

Images