Saline-alkali soil improvement method and system based on big data analysis of Internet of Things

Abstract

The invention provides a saline-alkali soil improvement method and system based on Internet of Things big data analysis, and the method comprises the steps: collecting a target saline-alkali soil sample, and obtaining an improvement target and a preset improvement scheme through sample analysis and a knowledge base; randomly selecting a plurality of different types of areas, and setting sensor nodes for layered sampling; monitoring the change of the salt content and the PH value of the soil and the crop growth condition through a monitoring system; dynamically adjusting a sampling time interval; a comprehensive evaluation model is established, an improvement scheme is adjusted according to the soil salt content change value, the PH value change value and the change value of the evaluation model, and the system comprises a sampling node setting module, a sampling module and a model establishment and improvement scheme adjustment module; through the method and the system, the data volume is reduced as much as possible and the monitoring efficiency is improved on the premise of accurately setting the sampling nodes, accurately sampling and accurately and comprehensively monitoring the condition of the saline-alkali soil, and meanwhile, a monitoring model is provided to comprehensively evaluate the evaluation and improvement effect.

Description

Technical Field

[0001] The present invention relates to the technical field of saline-alkali land improvement, and in particular to a saline-alkali land improvement method and system based on Internet of Things big data analysis. Background Art

[0002] Saline-alkali land is a type of land that affects crop growth due to excessive salt content in the soil. Saline-alkali land is widely distributed around the world, which has greatly restricted agricultural production. In order to improve the utilization efficiency of saline-alkali land, many researchers are committed to the study of saline-alkali land improvement methods. However, traditional saline-alkali land improvement methods mainly rely on experience and local experiments, lacking a comprehensive understanding of the entire area and scientific planning. With the development of the Internet of Things and big data technologies, more and more researchers have begun to explore the application of these technologies in saline-alkali land improvement. Through the Internet of Things sensors, key parameters such as soil salt content and pH value can be monitored in real time. At the same time, through big data analysis, these data can be comprehensively processed to provide more scientific improvement suggestions.

[0003] However, there are still some problems with the existing saline-alkali land improvement methods based on IoT big data analysis. Because the area of ​​saline-alkali land is usually large and the IoT data is also huge, how to set up sampling nodes for accurate sampling and reduce sampling data is a problem; secondly, when monitoring soil salinity and pH value, only the data of a single sensor is often considered, which affects the accuracy and reliability of the data; in addition, the existing methods do not fully consider the impact of crop growth, so they cannot provide a comprehensive reference for improvement plans. Summary of the invention

[0004] The present invention provides a saline-alkali land improvement method based on the Internet of Things big data analysis, which is used to accurately set sampling nodes and perform accurate sampling, and to reduce the amount of data as much as possible under the premise of accurately and comprehensively monitoring the saline-alkali land conditions, improve monitoring efficiency, and provide a monitoring model to comprehensively evaluate the effect of the improvement:

[0005] The present invention proposes a saline-alkali land improvement method based on Internet of Things big data analysis, the method comprising:

[0006] S1. Obtain the location of saline-alkali land through geographic information system and remote sensing data; collect soil samples of target saline-alkali land, obtain improvement targets and preset improvement plans through sample analysis and knowledge base; divide the target saline-alkali land into multiple regional types; randomly select multiple different types of regions, and set multiple sensor nodes for stratified sampling according to the improvement targets;

[0007] S2. Obtain the soil conditions of saline-alkali land through underground sensors; perform weighted analysis on the data of multiple sensors to obtain soil salinity and pH value; monitor the changes of soil salinity and pH value through a monitoring system; obtain the crop growth conditions of saline-alkali land through image sensors; set the sampling time interval of the sensor for sampling; dynamically adjust the sampling time interval according to the sampling results;

[0008] S3. Establish a comprehensive evaluation model based on the soil conditions of saline-alkali land and crop growth conditions, and adjust the improvement plan according to the changes in soil salt content, pH value and the changes in the evaluation model.

[0009] Furthermore, a saline-alkali land improvement method based on IoT big data analysis, wherein S1 comprises:

[0010] Obtaining the location of saline-alkali lands through geographic information systems and remote sensing data;

[0011] Collect soil samples from saline-alkali land and analyze the soil samples to obtain the soil type, salt content and pH value of the corresponding saline-alkali land;

[0012] Obtain the corresponding saline-alkali land improvement goals and preliminary improvement plans through sample analysis data and knowledge base;

[0013] Dividing the corresponding saline-alkali land into a plurality of first areas; wherein the first area includes a central area and a peripheral area;

[0014] Obtaining the altitudes of different first areas of the target saline-alkali land through a geographic information system;

[0015] Divide the first area into the second area according to the altitude; obtain the highest altitude and the lowest altitude of the target saline-alkali land, divide the height difference between the highest altitude and the lowest altitude into multiple levels, and divide the second area according to the levels;

[0016] A plurality of second areas of different types are randomly selected, and sensor nodes are set for sampling according to the improved target; the number of second areas of each different type selected is N, 2≤N≤5;

[0017] The depth from 0 to 30 cm from the ground was taken as the first sampling depth, and the depth from 30 to 60 cm was taken as the second sampling depth;

[0018] Sampling points at the first sampling depth and the second sampling depth are respectively selected to set up sensor groups for soil sampling; each sampling point includes multiple types of sensors; and an image sensor is set on the ground at the same sampling point.

[0019] Furthermore, a saline-alkali land improvement method based on IoT big data analysis, wherein S2 comprises:

[0020] Obtaining soil conditions of saline-alkali land through underground sensors; the saline-alkali land soil conditions include salt content and pH value; monitoring changes in soil salt content and pH value through a monitoring system;

[0021] The soil salinity is:

[0022]

[0023] The soil pH value is:

[0024]

[0025] Among them, Y 1i is the salinity value obtained by the sensor at the first depth; 1i The pH value obtained by the sensor at the first depth; Y 2i The salinity value obtained by the sensor at the second depth; 2i is the pH value obtained by the sensor at the second depth; a1 and a2 are coefficients, ranging from (0, 1); m is the number of sampling points of the first or second depth sensor;

[0026] Obtain crop growth conditions in saline-alkali land through image sensors;

[0027] Set the sampling time interval of the underground sensor to perform sampling;

[0028] Dynamically adjust the sampling time interval according to the sampling results;

[0029] T=(1-λ×avgD)×T0

[0030]

[0031] Where T is the sampling time interval; T0 is the preset sampling time interval; C j-1 is the value sampled by a certain underground sensor last time, C j-2 is the value sampled sequentially from a certain underground sensor; ΔC y The change value is preset for this type of sensor; avgD is the average value of D obtained by multiple underground sensors; Min(E) is the minimum E value obtained by multiple underground sensors; Max(E) is the maximum E value obtained by multiple underground sensors.

[0032] Furthermore, a saline-alkali land improvement method based on IoT big data analysis, wherein S3 comprises:

[0033] Classifying the growth conditions of crops; the classifications include very poor, poor, fair, good and excellent;

[0034] Establish a comprehensive evaluation model based on the soil conditions and crop growth conditions of saline-alkali land;

[0035] H=Z×[1-λ×|avgD|×(w1×A+w2×B)]

[0036]

[0037] Among them, H is the comprehensive evaluation score; Y 1i is the salinity value obtained by the sensor at the first depth; 1i The pH value obtained by the sensor at the first depth; Y 2i The salinity value obtained by the sensor at the second depth; 2i The pH value obtained by the sensor at the second depth; Y y is the target maximum salt content; P y is the target maximum pH value; w1 and w2 are weights, ranging from (0, 1); m is the number of sampling points of the first or second depth sensor; a1 and a2 are coefficients, ranging from (0, 1); Z is the crop growth score, ranging from 2 to 10, depending on the classification; very poor is 2, poor is 4, average is 6, good is 8, and excellent is 10;

[0038] Adjust the improvement plan according to the changes in soil salinity, pH value and evaluation model.

[0039] Furthermore, a saline-alkali land improvement method based on IoT big data analysis, wherein the crop growth conditions are classified, comprises:

[0040] The growth time series is obtained according to the crop growth time, and the growth conditions of each growth time series of the same crop are classified to obtain the image features of different classifications;

[0041] Extract the growth image features and time series of the crops corresponding to the target saline-alkali land;

[0042] Compare the image features of the crop corresponding to each sampling point with the features corresponding to different classifications of the same crop in the same time series in the database, and obtain the classification corresponding to the feature with the highest similarity as the classification corresponding to the crop in this sampling point and this time series;

[0043] If different sampling points of the target saline-alkali land in the same time series correspond to multiple classification features, the same classification corresponding to the most sampling points will be used as the final classification of the crop in this time series.

[0044] The present invention proposes a saline-alkali land improvement system based on Internet of Things big data analysis, the system comprising:

[0045] Sampling node setting module: obtain the location of saline-alkali land through geographic information system and remote sensing data; collect soil samples of target saline-alkali land, obtain improvement targets and preset improvement plans through sample analysis and knowledge base, and divide the target saline-alkali land into multiple regional types; randomly select multiple different types of regions, and set multiple sensor nodes for stratified sampling according to the improvement targets;

[0046] Sampling module: obtain the soil conditions of saline-alkali land through underground sensors; perform weighted analysis on the data of multiple sensors to obtain soil salinity and pH value; monitor the changes of soil salinity and pH value through a monitoring system; obtain the growth conditions of crops in saline-alkali land through image sensors; set the sampling time interval of the sensor for sampling; dynamically adjust the sampling time interval according to the sampling results;

[0047] Model establishment and improvement plan adjustment module: Establish a comprehensive evaluation model based on the soil conditions of saline-alkali land and crop growth conditions, and adjust the improvement plan based on changes in soil salt content, pH value and evaluation model.

[0048] Furthermore, in a saline-alkali land improvement system based on IoT big data analysis, the sampling node setting module includes:

[0049] Location acquisition module: obtain the location of saline-alkali land through geographic information system and remote sensing data;

[0050] Sample analysis module: collect saline-alkali soil samples, analyze the soil samples to obtain the soil type, salt content and pH value of the corresponding saline-alkali land;

[0051] Target setting module: obtain the corresponding saline-alkali land improvement target and preliminary improvement plan through sample analysis data and knowledge base;

[0052] The first region division module is used to divide the corresponding saline-alkali land into a plurality of first regions; wherein the first region includes a central region and a peripheral region;

[0053] The second area division module: obtaining the altitudes of different first areas of the target saline-alkali land through a geographic information system; dividing the first area into second areas according to the altitudes; obtaining the highest altitude and the lowest altitude of the target saline-alkali land, dividing the height difference between the highest altitude and the lowest altitude into multiple levels, and dividing the second area according to the levels;

[0054] Region selection module: randomly select multiple second regions of different types, and set sensor nodes for sampling according to the improved target; the number of second regions of each different type is N, 2≤N≤5;

[0055] Sampling depth setting module: the depth from 0 to 30 cm from the ground is used as the first sampling depth, and the depth from 30 to 60 cm is used as the second sampling depth;

[0056] Layered sampling module: sampling points are selected at the first sampling depth and the second sampling depth to set up sensor groups for soil sampling; each sampling point includes multiple types of sensors; image sensors are set on the ground at the same sampling point.

[0057] Furthermore, a saline-alkali land improvement system based on IoT big data analysis, the sampling module includes:

[0058] Soil condition acquisition module: obtains the soil condition of saline-alkali land through underground sensors; the saline-alkali land soil condition includes salt content and pH value; monitors the changes of soil salt content and pH value through a monitoring system;

[0059] The soil salinity is:

[0060]

[0061] The soil pH value is:

[0062]

[0063] Among them, Y 1i is the salinity value obtained by the sensor at the first depth; 1i The pH value obtained by the sensor at the first depth; Y 2i The salinity value obtained by the sensor at the second depth; 2i is the pH value obtained by the sensor at the second depth; a1 and a2 are coefficients, ranging from (0, 1); m is the number of sampling points of the first or second depth sensor;

[0064] Crop growth status acquisition module: obtains crop growth status in saline-alkali land through image sensors;

[0065] Sampling time setting module: set the sampling time interval of underground sensors for sampling; dynamically adjust the sampling time interval according to the sampling results;

[0066] T=(1-λ×avgD)×T0

[0067]

[0068] Where T is the sampling time interval; T0 is the preset sampling time interval; C j-1 is the value sampled by a certain underground sensor last time, C j-2 is the value of the last sampling of a certain underground sensor; ΔC yThe change value is preset for this type of sensor; avgD is the average value of D obtained by multiple underground sensors; Min(E) is the minimum E value obtained by multiple underground sensors; Max(E) is the maximum E value obtained by multiple underground sensors.

[0069] Furthermore, a saline-alkali land improvement system based on IoT big data analysis, the model establishment and improvement scheme adjustment module includes:

[0070] Crop classification module: classify the growth conditions of crops; the classifications include very poor, poor, average, good and excellent;

[0071] Comprehensive evaluation module establishment module: establish a comprehensive evaluation model based on saline-alkali soil conditions and crop growth conditions;

[0072] H=Z×[1-λ×|avgD|×(w1×A+w2×B)]

[0073]

[0074]

[0075] Among them, H is the comprehensive evaluation score; Y 1i is the salinity value obtained by the sensor at the first depth; 1i The pH value obtained by the sensor at the first depth; Y 2i The salinity value obtained by the sensor at the second depth; 2i The pH value obtained by the sensor at the second depth; Y y is the target maximum salt content; P y is the target maximum pH value; w1 and w2 are weights, ranging from (0, 1); m is the number of sampling points of the first or second depth sensor; a1 and a2 are coefficients, ranging from (0, 1); Z is the crop growth score, ranging from 2 to 10, depending on the classification; very poor is 2, poor is 4, average is 6, good is 8, and excellent is 10;

[0076] Monitoring and optimization module: adjust the improvement plan according to the changes in soil salinity, pH value and evaluation model.

[0077] Furthermore, in a saline-alkali land improvement system based on IoT big data analysis, the crop classification module includes:

[0078] Database crop feature classification module: obtain the growth time series according to the crop growth time, and classify the growth conditions of each growth time series of the same crop to obtain different classification image features;

[0079] Target saline-alkali land crop feature extraction module: extracts the growth image features and time series of the crops corresponding to the target saline-alkali land;

[0080] Sampling point crop classification determination module: compare the image features of the crop corresponding to each sampling point with the features corresponding to different classifications of the same crop in the same time series in the database, and obtain the classification corresponding to the feature with the highest similarity as the classification corresponding to the crop in this sampling point and this time series;

[0081] Final classification module: If different sampling points of the target saline-alkali land in the same time series correspond to multiple classification features, the same classification corresponding to the most sampling points will be used as the final classification of the crop in this time series.

[0082] The beneficial effects of the present invention are as follows: through the saline-alkali land improvement method and system based on the Internet of Things big data analysis described in the present invention, the location of the saline-alkali land can be accurately obtained through the geographic information system and remote sensing data, and precise positioning is achieved; through sample analysis and knowledge base, the improvement target is obtained, the saline-alkali land is divided into different areas, and further divided according to the altitude; by setting sensor nodes for stratified sampling, the data of the target saline-alkali land can be comprehensively obtained while minimizing the sensor nodes, and the soil conditions are obtained by underground sensors and the crop growth conditions are obtained by image sensors, so that multi-dimensional data of the saline-alkali land can be obtained, providing a comprehensive information basis; according to the sampling results, the sampling time interval is adjusted to ensure the accuracy of sampling and reduce data; using a comprehensive evaluation model, the improvement plan is dynamically adjusted according to the changes in soil salt content and pH value and the changes in the evaluation model, so that the improvement measures are more flexible and effective; the method combines the Internet of Things technology and big data analysis, realizes automated data collection and analysis, can quickly obtain the status information of the saline-alkali land, and adjusts the improvement plan according to the evaluation model, thereby improving the improvement efficiency. In summary, this saline-alkali land improvement method based on IoT big data analysis can achieve benefits and effects such as precise positioning, targeted improvement, multi-dimensional data sampling, dynamic adjustment and improved improvement efficiency, which helps to improve saline-alkali land scientifically and effectively and increase land availability and crop yields. BRIEF DESCRIPTION OF THE DRAWINGS

[0083] Figure 1 This is a schematic diagram of a saline-alkali land improvement method based on IoT big data analysis. DETAILED DESCRIPTION

[0084] In order to more clearly understand the above-mentioned purpose, features and advantages of the present invention, the present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that the embodiments of the present application and the features in the embodiments can be combined with each other without conflict.

[0085] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. The embodiments described are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0086] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art of the present invention. The terms used in the specification of the present invention herein are only for the purpose of describing specific embodiments and are not intended to limit the present invention.

[0087] This embodiment provides a saline-alkali land improvement method based on Internet of Things big data analysis, the method comprising:

[0088] S1. Obtain the location of saline-alkali land through geographic information system and remote sensing data; collect soil samples of target saline-alkali land, obtain improvement targets and preset improvement plans through sample analysis and knowledge base, and divide the target saline-alkali land into multiple regional types; randomly select multiple different types of regions, set sensor nodes for stratified sampling according to the improvement targets; connect sensors to the monitoring platform through the Internet of Things;

[0089] S2. Obtain the soil conditions of saline-alkali land through underground sensors; perform weighted analysis on the data of multiple sensors to obtain soil salinity and pH value; monitor the changes of soil salinity and pH value through a monitoring system; obtain the crop growth conditions of saline-alkali land through image sensors; set the sampling time interval of the sensor for sampling; dynamically adjust the sampling time interval according to the sampling results;

[0090] S3. Establish a comprehensive evaluation model based on the soil conditions of saline-alkali land and crop growth conditions, and adjust the improvement plan according to the changes in soil salt content, pH value and the evaluation model.

[0091] The working principle of the above technical solution is as follows: The location of saline-alkali land is obtained through geographic information system and remote sensing data: Geographic information system (GIS) and remote sensing technology are used to locate the target saline-alkali land and obtain basic information such as terrain. Soil samples are collected from the target saline-alkali land, and the composition, texture, structure and other information in the soil samples are analyzed in the laboratory. Combined with the expert knowledge base, the improvement target and the preset improvement plan are determined. According to the characteristics of the target saline-alkali land, it is divided into multiple regional types, and multiple different types of regions are randomly selected. According to the improvement target, sensor nodes are set for stratified sampling to obtain more accurate soil data. Through the Internet of Things technology, underground sensors are buried in saline-alkali land to obtain soil parameters such as salt content and pH value. The data of multiple sensors are weighted and analyzed to obtain more accurate soil information. The monitoring system monitors the changes in soil salt content and pH value in real time to understand the soil conditions so as to adjust the improvement plan in time. The crop growth situation of saline-alkali land is obtained through image sensors, including crop growth, color and other characteristics. Combined with sensor data and image data, a comprehensive assessment of the growth status of crops is carried out. According to the sampling results, the sampling time interval of the sensor is dynamically adjusted to ensure the timeliness and accuracy of the data. According to the soil conditions and crop growth of saline-alkali land, a comprehensive evaluation model is established to comprehensively evaluate the effect of the improvement plan. According to the change value of soil salt content, pH value and the change value of the evaluation model, the improvement plan is adjusted and optimized to achieve the best improvement effect. The soil environment and crop growth conditions of saline-alkali land are improved by dynamically adjusting the improvement measures, such as applying appropriate improvement materials and adjusting irrigation water quality. In general, this method achieves dynamic adjustment of saline-alkali land improvement plans by obtaining saline-alkali land location information, stratified sampling of soil and crop data in different regions, and analysis based on a comprehensive evaluation model. Through this process, saline-alkali land can be improved scientifically and efficiently to increase land availability and crop yields.

[0092] The effects of the above technical scheme are as follows: through the geographic information system and remote sensing data, the location of the saline-alkali land can be accurately obtained to achieve precise positioning; through sample analysis and knowledge base, the improvement target is obtained, the saline-alkali land is divided into different areas, and further divided according to the altitude; by setting sensor nodes for stratified sampling, the data of the target saline-alkali land can be comprehensively obtained while minimizing the sensor nodes, and the soil conditions are obtained by underground sensors and the crop growth conditions are obtained by image sensors, so that multi-dimensional data of the saline-alkali land can be obtained to provide a comprehensive information basis; through the sampling results, the sampling time interval is adjusted to ensure the accuracy of sampling and reduce data; using the comprehensive evaluation model, the improvement plan is dynamically adjusted according to the changes in soil salt content and pH value and the changes in the evaluation model, so that the improvement measures are more flexible and effective; this method combines the Internet of Things technology and big data analysis to realize automated data collection and analysis, can quickly obtain the status information of the saline-alkali land, and adjust the improvement plan according to the evaluation model, thereby improving the improvement efficiency. In summary, this saline-alkali land improvement method based on IoT big data analysis can achieve benefits and effects such as precise positioning, targeted improvement, multi-dimensional data sampling, dynamic adjustment and improved improvement efficiency, which helps to improve saline-alkali land scientifically and effectively and increase land availability and crop yields.

[0093] This embodiment provides a saline-alkali land improvement method based on IoT big data analysis, wherein S1 includes:

[0094] Obtaining the location of saline-alkali lands through geographic information systems and remote sensing data;

[0095] Collect soil samples from saline-alkali land and analyze the soil samples to obtain the soil type, salt content and pH value of the corresponding saline-alkali land;

[0096] Obtain the corresponding saline-alkali land improvement goals and preliminary improvement plans through sample analysis data and knowledge base;

[0097] For example: According to the analysis results, it was found that the soil in this area is slightly to moderately salinized soil with a high salt content and a relatively alkaline pH value.

[0098] Based on sample analysis data and reference to relevant knowledge bases, the improvement goal should be to reduce soil salt content and increase soil pH to improve soil quality and increase suitability for plant growth.

[0099] Based on this information, initial improvement plans may include the following measures:

[0100] Drainage improvement: Improve soil drainage by building a drainage system to remove excess salt and water;

[0101] Salt flushing: Use desalinated water or low-salt water to flush the soil to remove the salt in the soil;

[0102] Organic matter addition: Apply organic fertilizers or organic materials such as straw to improve soil structure and texture and increase soil water retention capacity.

[0103] Gypsum addition: Add appropriate amount of gypsum and other soil conditioners to reduce the pH value of the soil and improve the acidity and alkalinity of the soil;

[0104] Suitable crop selection: Choose salt-alkali tolerant crops for planting, which can adapt to the environment of soil salinity and alkalinity;

[0105] The corresponding saline-alkali land is divided into a plurality of first regions; wherein the first region includes a central region and an edge region; the vertices of the inscribed rectangle in the region are used as intersection lines; the intersection point of the two intersection lines is used as the center, and four points are selected at 1 / 2 of the distance from the intersection point of the two intersection lines to each vertex of the inscribed rectangle to form a rectangle, and the area inside the rectangle is the central region; the outer area is the edge region; the edge region is divided into a plurality of regions according to the edge position;

[0106] Obtaining the altitudes of different first areas of the target saline-alkali land through a geographic information system;

[0107] The first area is divided into the second area according to the altitude; the highest altitude and the lowest altitude of the target saline-alkali land are obtained, the height difference between the highest altitude and the lowest altitude is divided into multiple levels, and the second area is divided according to the levels in different areas; the number of levels is h=ceil((h max -h min ) / standard floor height), the function ceil() represents rounding up a real number, the standard floor height ranges from 30 to 40, in cm;

[0108] Among them, h max is the highest altitude of the target saline-alkali land area; h min The lowest altitude of the target saline-alkali land area;

[0109] For example, if the height difference between the highest altitude and the lowest altitude of the target saline-alkali land is 50cm, then the lowest altitude in the target saline-alkali land to the lowest altitude plus 30cm is the first level, and the lowest altitude in the target saline-alkali land plus 30cm to the highest altitude is the second level; if the height difference between the highest altitude and the lowest altitude of the target saline-alkali land is 70cm, then the lowest altitude in the target saline-alkali land to the low altitude plus 35cm is the first level, and the lowest altitude in the saline-alkali land plus 35cm to the highest altitude in the saline-alkali land is the second level;

[0110] If the difference between the highest and lowest altitudes of the target saline-alkali land is less than 30 cm, it is divided into only one level;

[0111] A plurality of second areas of different types are randomly selected, and sensor nodes are set for sampling according to the improved target; the number N of the second areas of each different type is selected to be 2≤N≤5;

[0112] The depth from 0 to 30 cm from the ground was taken as the first sampling depth, and the depth from 30 to 60 cm was taken as the second sampling depth;

[0113] Select sampling points at the first sampling depth and the second sampling depth to set up sensor groups for soil sampling; each sampling point includes multiple types of sensors;

[0114] An image sensor is set up on the ground at the same sampling point.

[0115] The working principle of the above technical solution is as follows: first, the location of the target saline-alkali land is obtained through the geographic information system and remote sensing data, and the soil samples of the saline-alkali land are collected for analysis to obtain key data such as soil type, salt content and pH value of the corresponding saline-alkali land. Then, the improvement target and preliminary improvement plan of the corresponding saline-alkali land are obtained through sample analysis data and knowledge base. Then, the corresponding saline-alkali land is divided into multiple first areas, wherein the first area includes a central area and an edge area. The central area is divided into rectangles by means of intersecting lines, and the edge area is further divided into multiple areas according to the edge position. The altitudes of different first areas of the target saline-alkali land are obtained according to the geographic information system, and the first area is divided into second areas according to the altitudes. The first area is divided into second areas according to the altitudes; the highest altitude and the lowest altitude of the target saline-alkali land are obtained, and the height difference between the highest altitude and the lowest altitude is divided into multiple levels, and the second areas are divided according to the levels in different areas. Multiple second areas of different types are randomly selected, and sensor nodes are set for sampling according to the improvement target. The number of second areas selected for each different type is 2≤N≤5. The depth from 0 to 30 cm from the ground is taken as the first sampling depth, the depth from 30 to 60 cm is taken as the second sampling depth, and various types of sensors are included at each sampling point for soil sampling and monitoring.

[0116] At the same time, image sensors are set up on the ground at the same sampling points to collect and analyze data. Through the analysis and processing of the sampling data, a variety of key data such as the soil characteristics of the target saline-alkali land can be obtained, and then a more accurate saline-alkali land improvement plan can be formulated to achieve more targeted and effective saline-alkali land management.

[0117] The effect of the above technical solution is: the location of saline-alkali land can be obtained through geographic information system and remote sensing data, and the distribution range of saline-alkali land can be accurately determined. Soil samples of saline-alkali land are collected and comprehensively analyzed, including indicators such as soil type, salt content and pH value, so that the soil conditions of saline-alkali land can be fully understood. Through sample analysis data and knowledge base, combined with the goal of saline-alkali land improvement, a preliminary improvement plan is formulated. In this way, saline-alkali land can be improved in a targeted manner according to actual conditions. The saline-alkali land is divided into multiple first areas, including central areas and marginal areas, so that regional management and improvement operations can be better carried out. According to the altitude difference between different areas of the target saline-alkali land, the first area is further divided into second areas. In this way, soil conditions and improvement strategies at different levels can be considered more accurately. Multiple second areas of different types are randomly selected, and sensor nodes are set for sampling. In this way, the soil conditions in different areas can be fully understood and reference data can be provided for subsequent improvement. By setting multiple sampling depths, including 0 to 30 cm depth and 30 to 60 cm depth on the ground, information on soil at different depths can be obtained. This helps to analyze the vertical distribution of saline-alkali land soil and guide more accurate improvement strategies. By installing image sensors on the ground at the sampling point, the surface conditions can be monitored in real time. This can timely discover key information such as soil moisture distribution and vegetation growth, providing an effective basis for adjustments and judgments during the improvement process. Through the above methods, saline-alkali land can be improved more scientifically, achieving continuous improvement and restoration of soil, improving soil fertility, increasing crop yields, reducing the degree of salinization of cultivated land, and promoting sustainable agricultural development.

[0118] This embodiment provides a saline-alkali land improvement method based on IoT big data analysis, wherein S2 includes:

[0119] Obtaining soil conditions of saline-alkali land through underground sensors; the saline-alkali land soil conditions include salt content and pH value; monitoring changes in soil salt content and pH value through a monitoring system;

[0120] The soil salinity is:

[0121]

[0122] The soil pH value is:

[0123]

[0124] Among them, Y 1i is the salinity value obtained by the sensor at the first depth; 1i The pH value obtained by the sensor at the first depth; Y 2i The salinity value obtained by the sensor at the second depth; 2iis the pH value obtained by the sensor at the second depth; a1 and a2 are coefficients, ranging from (0, 1); m is the number of sampling points of the first or second depth sensor;

[0125] The crop growth conditions in saline-alkali land are obtained through image sensors; the sampling frequencies of image sensors and underground sensors are kept consistent;

[0126] Set the sampling time interval of the underground sensor to perform sampling;

[0127] Dynamically adjust the sampling time interval according to the sampling results;

[0128] T=(1-λ×avgD)×T0

[0129]

[0130] Wherein, T is the sampling time interval; T0 is the preset sampling time interval; C j-1 is the value sampled by a certain underground sensor last time, C j-2 is the value of the last sampling of a certain underground sensor; ΔC y The change value is preset for this type of sensor; avgD is the average value of D obtained by multiple underground sensors; Min(E) is the minimum E value obtained by multiple underground sensors; Max(E) is the maximum E value obtained by multiple underground sensors.

[0131] The working principle of the above technical solution is: the soil conditions of saline-alkali land, including the salt content and pH value of the soil, are obtained through underground sensors. The sensors are divided into two depths (first depth and second depth), and the salt content and pH value at the corresponding depths are obtained respectively. The changes of soil salt content and pH value are monitored in real time through the monitoring system. The monitoring system can record and display the dynamic changes of soil conditions and understand the soil conditions in real time. The sensor sampling values ​​of multiple depths obtained are weighted averaged according to the weight coefficients to obtain the overall soil salt content and pH value. Among them, the weight coefficients a1 and a2 are determined according to the actual situation and range between 0 and 1; the growth of crops on saline-alkali land is monitored in real time through image sensors. The image sensor can capture information such as the growth status, leaf color, canopy density, etc. of the plant to evaluate the improvement effect of saline-alkali land. According to the sampling results and the preset change amount, the sampling time interval of the sensor is dynamically adjusted. The stability of the soil condition is evaluated by the calculated D value (difference in change). If the stability is high, the sampling time interval can be extended; conversely, if the stability is low, the sampling time interval needs to be shortened. According to the stability evaluation results, the new sampling time interval T is calculated by the formula.

[0132] The effect of the above technical solution is: the data of soil salinity and pH value are obtained through underground sensors, and its changes are monitored in real time through the monitoring system. This enables farmers and agricultural experts to understand the soil conditions of saline-alkali land in real time and take corresponding measures in time. The overall soil salinity and pH value are obtained by calculating the weighted average of the sampling values ​​of multiple depth sensors. This accurate assessment helps to better understand the soil characteristics of saline-alkali land and provide a reference for targeted improvement. The growth of crops on saline-alkali land is monitored in real time using image sensors. By monitoring information such as the growth status, leaf color, and canopy density of plants, problems can be discovered in time and appropriate measures can be taken, which helps to improve crop yield and quality. The stability evaluation index D is calculated based on the sampling results, and the sampling time interval of the sensor is dynamically adjusted through the formula. By dynamically adjusting the sampling time interval, redundant sampling can be reduced as much as possible while ensuring data quality, thereby saving energy and storage space. Due to the reduction of redundant data, the speed of the data processing stage will also increase accordingly. The adjustment of the sampling time interval takes into account the data changes of multiple sensors. In this way, a more accurate data sampling time can be obtained, thereby improving data accuracy. This can capture changes in soil conditions more quickly, adjust measures in time, and improve improvement effects. By real-time monitoring and accurate assessment of soil conditions, as well as dynamically adjusting the sampling time interval according to the sampling results, this method can more accurately guide the improvement of saline-alkali land. Targeted measures can be taken according to specific soil conditions and crop needs to improve the improvement effect, increase crop yields, and reduce the adverse effects of soil salinity on plant growth. In short, this saline-alkali land improvement method based on IoT big data analysis can provide real-time and accurate soil information and crop growth monitoring, helping farmers and agricultural experts to effectively improve the soil conditions of saline-alkali land and increase crop yields and quality.

[0133] This embodiment provides a saline-alkali land improvement method based on IoT big data analysis, wherein S3 includes:

[0134] Classifying the growth conditions of crops; the classifications include very poor, poor, fair, good and excellent;

[0135] Establish a comprehensive evaluation model based on saline-alkali soil conditions and crop growth;

[0136] H=Z×[1-λ×|avgD|×(w1×A+w2×B)]

[0137]

[0138]

[0139] Among them, H is the comprehensive evaluation score; Y 1i is the salinity value obtained by the sensor at the first depth; 1iThe pH value obtained by the sensor at the first depth; Y 2i The salinity value obtained by the sensor at the second depth; 2i The pH value obtained by the sensor at the second depth; Y y is the target maximum salt content; P y is the target maximum pH value; w1 and w2 are weights, ranging from (0, 1); m is the number of sampling points of the first or second depth sensor; a1 and a2 are coefficients, ranging from (0, 1); Z is the crop growth score, ranging from 2 to 10, depending on the classification; very poor is 2, poor is 4, average is 6, good is 8, and excellent is 10;

[0140] C j-1 is the value sampled by a certain underground sensor last time, C j-2 is the value of the last sampling of a certain underground sensor; ΔC y The preset change value is for this type of sensor; avgD is the average value of D obtained by multiple underground sensors; Min(E) is the minimum E value obtained by multiple underground sensors; Max(E) is the maximum E value obtained by multiple underground sensors;

[0141] Adjust the improvement plan according to the changes in soil salinity, pH value and evaluation model;

[0142] Compare the changes in soil salinity at different time points or at different depths; if the salinity is found to be gradually decreasing, it means that the improvement measures may have taken effect, and the original improvement plan can be maintained and strengthened; if the change in salinity is not obvious or has increased, it may be necessary to re-evaluate the effect of the improvement plan and consider taking further measures;

[0143] If the pH value gradually approaches the target maximum value during the improvement process, it means that the improvement plan is effective in adjusting the soil pH; if the pH value change is not obvious or deviates from the target, you can consider adjusting the improvement plan in a targeted manner, such as increasing or decreasing the application of acidic and alkaline substances to achieve the ideal pH value.

[0144] By comparing the changes in the comprehensive evaluation scores of the evaluation model at different time points, we can understand the effect of the improvement plan; if the comprehensive evaluation score gradually increases, it means that the improvement plan is effective and the original improvement measures can continue to be implemented; if the change value of the evaluation model shows that the comprehensive evaluation score decreases or fluctuates, the improvement plan may need to be adjusted, such as adjusting the amount of fertilizer or water or adding other improvement measures.

[0145] The working principle of the above technical solution is: the growth conditions of crops on saline-alkali land are classified into very poor, poor, general, good and excellent grades; according to the soil conditions of saline-alkali land and the growth conditions of crops, a comprehensive evaluation model is established to evaluate the improvement effect. The model obtains a comprehensive evaluation score H by calculation, which includes two parts A and B. Part A is obtained by calculating the deviation between the salt content value and pH value obtained by the first depth sensor and the target maximum salt content and pH value, and weighted summing them according to coefficients a1 and a2. Part B is also obtained by calculating the deviation between the salt content value and pH value obtained by the second depth sensor and the target maximum salt content and pH value, and weighted summing them according to coefficients a1 and a2.

[0146] The comprehensive evaluation score H uses weights w1 and w2 to perform weighted summation on A and B to obtain the final evaluation result. According to the classification, the crop growth situation is converted into a crop growth score Z, which ranges from 2 to 10, depending on the classification. For example, very poor corresponds to 2, poor corresponds to 4, general corresponds to 6, good corresponds to 8, and excellent corresponds to 10. By analyzing the change values ​​of soil salt content, pH value, and evaluation model, the improvement plan of saline-alkali land is adjusted. According to the change trend of the comprehensive evaluation score H and the crop growth score Z, the improvement measures can be optimized, such as adjusting the irrigation amount, applying appropriate salt treatment agents, and changing the crop planting method. In short, this method evaluates the improvement effect of saline-alkali land through a comprehensive evaluation model and crop growth score, and adjusts the improvement plan based on data analysis to improve the soil quality and crop yield of saline-alkali land.

[0147] The effect of the above technical solution is: by comprehensively evaluating the soil conditions and crop growth conditions of saline-alkali land, the comprehensive evaluation score H is calculated using parameters such as the salt content value and pH value obtained by the sensor, so as to objectively and accurately evaluate the improvement effect of saline-alkali land. The parameters avgD, λ and weights w1 and w2 in the formula are dynamically adjusted according to the sampling results of the underground sensor, and the improvement of saline-alkali land can be accurately evaluated according to the actual situation, and the importance of different factors can be flexibly adjusted; through the Internet of Things technology and big data analysis, the changes in saline-alkali land soil parameters can be monitored and collected in real time, and the crop growth conditions can be evaluated in real time. In this way, problems can be discovered in time and corresponding measures can be taken to achieve refined management of saline-alkali land. Through the changing trends of the comprehensive evaluation score H and the crop growth score Z, scientific basis and reference can be provided for farmers, agricultural experts and decision makers to guide the formulation of reasonable saline-alkali land improvement plans. According to the change value of the evaluation model, the improvement measures can be adjusted, such as adjusting the irrigation amount, applying appropriate salt treatment agents, changing the crop planting method, etc., to improve the improvement effect. Through the improvement plan for saline-alkali land, parameters such as soil salt content and pH value can be adjusted, the salt content in the soil can be reduced, the soil structure and water retention capacity can be improved, and the growth environment of crops can be improved. This will help to increase the yield and quality of crops. Through accurate data analysis and evaluation, agricultural production factors such as water resources, fertilizers and pesticides can be reasonably utilized to reduce waste and pollution, so as to achieve the purpose of saving resources and protecting the environment. In summary, the saline-alkali land improvement method based on IoT big data analysis can provide accurate evaluation and guidance, realize refined management, improve soil quality and crop yield, and save resources and protect the environment.

[0148] In this embodiment, a saline-alkali land improvement method based on IoT big data analysis is provided, wherein the crop growth conditions are classified as follows:

[0149] A growth time series is obtained according to the crop growth time, and the growth conditions of each growth time series of the same crop are classified to obtain image features of different classifications, wherein the classifications include very poor, poor, average, good and excellent;

[0150] Extract the growth image features and time series of the crops corresponding to the target saline-alkali land;

[0151] Compare the image features of the crop corresponding to each sampling point with the features corresponding to different classifications of the same crop in the same time series in the database, and obtain the classification corresponding to the feature with the highest similarity as the classification corresponding to the crop in this sampling point and this time series;

[0152] If different sampling points of the target saline-alkali land in the same time series correspond to multiple classification features, the same classification corresponding to the most sampling points will be used as the final classification of the crop in this time series.

[0153] The working principle of the above technical solution is as follows: first, the growth conditions of the same crop under different growth time series are classified. This can be determined by observing and recording the growth status of the same crop at different growth stages, such as leaf color, plant height, plant morphology, etc. The classification can include very poor, poor, general, good and excellent grades. For each crop growth time series, the growth image features of the corresponding crop are extracted. This can be done by processing and analyzing crop images using computer vision technology to extract features related to crop growth status, such as texture, shape, color, etc. The extracted crop image features are compared with the features corresponding to different classifications of the same crop in the database. By calculating the similarity between the features, the classification corresponding to the features most similar to the crop features to be classified is found. If different test points correspond to multiple classification features, the same classification corresponding to the most test points is selected as the final classification result. This is to increase the accuracy and reliability of the classification, and the results of multiple test points are voted or counted, and the classification with the highest frequency is selected as the final result. Through the above steps, the method uses IoT big data analysis and image feature extraction technology to classify the growth conditions of crops in saline-alkali land. By comparing feature similarity and multi-point detection result statistics, accurate classification results can be obtained, providing a scientific basis for the improvement and management of saline-alkali land.

[0154] The effect of the above technical solution is: by classifying the crop growth time series, the growth status of the crop can be accurately evaluated and described. The crop growth conditions are divided into very poor, poor, general, good and excellent grades, which can more specifically understand the adaptability and growth status of crops in saline-alkali land. The method uses image feature extraction technology to convert the growth image of the crop into a feature vector. This classification method based on image features can more objectively and quantitatively evaluate the growth of crops, avoiding subjective and human interference factors. The image features of the crop are compared with the features in the database, and the similarity is calculated. By comparing the features corresponding to different classifications, the features most similar to the crop to be classified are found, thereby determining the classification of the crop. In this way, the accuracy and reliability of classification can be improved by taking advantage of big data analysis and machine learning algorithms. If different test points correspond to multiple classification features, the method uses the same classification corresponding to the most test points as the final classification result. This method of integrating multi-point detection results can reduce the possible misjudgment of individual test points and improve the credibility and stability of classification results. The saline-alkali land improvement method based on IoT big data analysis can provide agricultural managers and decision makers with accurate crop growth assessment and classification information, helping them better understand the status of saline-alkali land and take corresponding improvement measures. At the same time, this method is based on the analysis of big data and image features, overcoming the limitations of traditional subjective judgment and individual factors, and has higher scientificity and objectivity.

[0155] This embodiment provides a saline-alkali land improvement system based on Internet of Things big data analysis, and the system includes:

[0156] Sampling node setting module: obtain the location of saline-alkali land through geographic information system and remote sensing data; collect soil samples of target saline-alkali land, obtain improvement targets and preset improvement plans through sample analysis and knowledge base, and divide the target saline-alkali land into multiple regional types; randomly select multiple different types of regions, set sensor nodes according to the improvement targets for stratified sampling;

[0157] Sampling module: obtain the soil conditions of saline-alkali land through underground sensors; perform weighted analysis on the data of multiple sensors to obtain soil salinity and pH value; monitor the changes of soil salinity and pH value through a monitoring system; obtain the growth conditions of crops in saline-alkali land through image sensors; set the sampling time interval of the sensor for sampling; dynamically adjust the sampling time interval according to the sampling results;

[0158] Model establishment and improvement plan adjustment module: Establish a comprehensive evaluation model based on the soil conditions of saline-alkali land and crop growth conditions, and adjust the improvement plan based on changes in soil salt content, pH value and evaluation model.

[0159] The working principle of the above technical solution is as follows: The location of saline-alkali land is obtained through geographic information system and remote sensing data: Geographic information system (GIS) and remote sensing technology are used to locate the target saline-alkali land and obtain basic information such as terrain. Soil samples are collected from the target saline-alkali land, and the composition, texture, structure and other information in the soil samples are analyzed in the laboratory. Combined with the expert knowledge base, the improvement target and the preset improvement plan are determined. According to the characteristics of the target saline-alkali land, it is divided into multiple regional types, and multiple different types of regions are randomly selected. According to the improvement target, sensor nodes are set for stratified sampling to obtain more accurate soil data. Through the Internet of Things technology, underground sensors are buried in saline-alkali land to obtain soil parameters such as salt content and pH value. The data of multiple sensors are weighted and analyzed to obtain more accurate soil information. The monitoring system monitors the changes in soil salt content and pH value in real time to understand the soil conditions so as to adjust the improvement plan in time. The crop growth situation of saline-alkali land is obtained through image sensors, including crop growth, color and other characteristics. Combined with sensor data and image data, a comprehensive assessment of the growth status of crops is carried out. According to the sampling results, the sampling time interval of the sensor is dynamically adjusted to ensure the timeliness and accuracy of the data. According to the soil conditions and crop growth of saline-alkali land, a comprehensive evaluation model is established to comprehensively evaluate the effect of the improvement plan. According to the change value of soil salt content, pH value and the change value of the evaluation model, the improvement plan is adjusted and optimized to achieve the best improvement effect. The soil environment and crop growth conditions of saline-alkali land are improved by dynamically adjusting the improvement measures, such as applying appropriate improvement materials and adjusting irrigation water quality. In general, this method achieves dynamic adjustment of saline-alkali land improvement plans by obtaining saline-alkali land location information, stratified sampling of soil and crop data in different regions, and analysis based on a comprehensive evaluation model. Through this process, saline-alkali land can be improved scientifically and efficiently to increase land availability and crop yields.

[0160] The effects of the above technical scheme are as follows: through the geographic information system and remote sensing data, the location of the saline-alkali land can be accurately obtained to achieve precise positioning; through sample analysis and knowledge base, the improvement target is obtained, the saline-alkali land is divided into different areas, and further divided according to the altitude; by setting sensor nodes for stratified sampling, the data of the target saline-alkali land can be comprehensively obtained while minimizing the sensor nodes, and the soil conditions are obtained by underground sensors and the crop growth conditions are obtained by image sensors, so that multi-dimensional data of the saline-alkali land can be obtained to provide a comprehensive information basis; through the sampling results, the sampling time interval is adjusted to ensure the accuracy of sampling and reduce data; using the comprehensive evaluation model, the improvement plan is dynamically adjusted according to the changes in soil salt content and pH value and the changes in the evaluation model, so that the improvement measures are more flexible and effective; this method combines the Internet of Things technology and big data analysis to realize automated data collection and analysis, can quickly obtain the status information of the saline-alkali land, and adjust the improvement plan according to the evaluation model, thereby improving the improvement efficiency. In summary, this saline-alkali land improvement method based on IoT big data analysis can achieve benefits and effects such as precise positioning, targeted improvement, multi-dimensional data sampling, dynamic adjustment and improved improvement efficiency, which helps to improve saline-alkali land scientifically and effectively and increase land availability and crop yields.

[0161] This embodiment provides a saline-alkali land improvement system based on IoT big data analysis, and the sampling node setting module includes:

[0162] Location acquisition module: obtain the location of saline-alkali land through geographic information system and remote sensing data;

[0163] Sample analysis module: collect saline-alkali soil samples, analyze the soil samples to obtain the soil type, salt content and pH value of the corresponding saline-alkali land;

[0164] Target setting module: Obtain the corresponding saline-alkali land improvement target and preliminary improvement plan through sample analysis data and knowledge base; for example: according to the analysis results, it is found that the soil in this area is slightly to moderately salinized soil with high salt content and alkaline pH value;

[0165] Based on sample analysis data and reference to relevant knowledge bases, the improvement goal should be to reduce soil salt content and increase soil pH to improve soil quality and increase suitability for plant growth.

[0166] Based on this information, initial improvement plans may include the following measures:

[0167] Drainage improvement: Improve soil drainage by building a drainage system to remove excess salt and water;

[0168] Salt flushing: Use desalinated water or low-salt water to flush the soil to remove the salt in the soil;

[0169] Organic matter addition: Apply organic fertilizers or organic materials such as straw to improve soil structure and texture and increase soil water retention capacity.

[0170] Gypsum addition: Add appropriate amount of gypsum and other soil conditioners to reduce the pH value of the soil and improve the acidity and alkalinity of the soil;

[0171] Suitable crop selection: Choose salt-alkali tolerant crops for planting, which can adapt to the environment of soil salinity and alkalinity;

[0172] The first region division module is used to divide the corresponding saline-alkali land into a plurality of first regions; wherein the first region includes a central region and a peripheral region;

[0173] The vertices of the inscribed rectangle in the region are used as intersection lines; with the intersection point of the two intersection lines as the center, four points are selected at 1 / 2 of the distance from the intersection point of the two intersection lines to each vertex of the inscribed rectangle to make a rectangle. The area inside the rectangle is the central area; the outer area is the edge area; the edge area is divided into multiple areas according to the edge position;

[0174] The second area division module: obtain the altitudes of different first areas of the target saline-alkali land through the geographic information system; divide the first area into the second area according to the altitude; obtain the highest altitude and the lowest altitude of the target saline-alkali land, divide the height difference between the highest altitude and the lowest altitude into multiple levels, and divide the second area according to the levels; the number of levels is h=ceil((h max -h min ) / standard floor height), the function ceil() represents rounding up a real number, the standard floor height ranges from 30 to 40, in cm;

[0175] Among them, h max is the highest altitude of the target saline-alkali land area; h min The lowest altitude of the target saline-alkali land area;

[0176] For example, if the height difference between the highest altitude and the lowest altitude of the target saline-alkali land is 50cm, then the lowest altitude in the target saline-alkali land to the lowest altitude plus 30cm is the first level, and the lowest altitude in the target saline-alkali land plus 30cm to the highest altitude is the second level; if the height difference between the highest altitude and the lowest altitude of the target saline-alkali land is 70cm, then the lowest altitude in the target saline-alkali land to the low altitude plus 35cm is the first level, and the lowest altitude in the saline-alkali land plus 35cm to the highest altitude in the saline-alkali land is the second level;

[0177] If the difference between the highest and lowest altitudes of the target saline-alkali land is less than 30 cm, it is divided into only one level;

[0178] Region selection module: randomly select multiple second regions of different types, and set sensor nodes for sampling according to the improved target; the number of second regions of each different type is N, 2≤N≤5;

[0179] Sampling depth setting module: the depth from 0 to 30 cm from the ground is used as the first sampling depth, and the depth from 30 to 60 cm is used as the second sampling depth;

[0180] Layered sampling module: sampling points are selected at the first sampling depth and the second sampling depth to set up sensor groups for soil sampling; each sampling point includes multiple types of sensors; image sensors are set on the ground at the same sampling point.

[0181] The working principle of the above technical solution is as follows: first, the location of the target saline-alkali land is obtained through the geographic information system and remote sensing data, and the soil samples of the saline-alkali land are collected for analysis to obtain key data such as soil type, salt content and pH value of the corresponding saline-alkali land. Then, the improvement target and preliminary improvement plan of the corresponding saline-alkali land are obtained through sample analysis data and knowledge base. Then, the corresponding saline-alkali land is divided into multiple first areas, wherein the first area includes a central area and an edge area. The central area is divided into rectangles by means of intersecting lines, and the edge area is further divided into multiple areas according to the edge position. The altitudes of different first areas of the target saline-alkali land are obtained according to the geographic information system, and the first area is divided into second areas according to the altitudes. The first area is divided into second areas according to the altitudes; the highest altitude and the lowest altitude of the target saline-alkali land are obtained, and the height difference between the highest altitude and the lowest altitude is divided into multiple levels, and the second areas are divided according to the levels in different areas. Multiple second areas of different types are randomly selected, and sensor nodes are set for sampling according to the improvement target. The number of second areas selected for each different type is 2≤N≤5. The depth from 0 to 30 cm from the ground is taken as the first sampling depth, the depth from 30 to 60 cm is taken as the second sampling depth, and various types of sensors are included at each sampling point for soil sampling and monitoring.

[0182] At the same time, image sensors are set up on the ground at the same sampling points to collect and analyze data. Through the analysis and processing of the sampling data, a variety of key data such as the soil characteristics of the target saline-alkali land can be obtained, and then a more accurate saline-alkali land improvement plan can be formulated to achieve more targeted and effective saline-alkali land management.

[0183] The effect of the above technical solution is: the location of saline-alkali land can be obtained through geographic information system and remote sensing data, and the distribution range of saline-alkali land can be accurately determined. Soil samples of saline-alkali land are collected and comprehensively analyzed, including indicators such as soil type, salt content and pH value, so that the soil conditions of saline-alkali land can be fully understood. Through sample analysis data and knowledge base, combined with the goal of saline-alkali land improvement, a preliminary improvement plan is formulated. In this way, saline-alkali land can be improved in a targeted manner according to actual conditions. The saline-alkali land is divided into multiple first areas, including central areas and marginal areas, so that regional management and improvement operations can be better carried out. According to the altitude difference between different areas of the target saline-alkali land, the first area is further divided into the level of the second area. In this way, soil conditions and improvement strategies at different levels can be considered more accurately. Multiple different types of second areas are randomly selected, and sensor nodes are set for sampling. In this way, the soil conditions in different areas can be fully understood and reference data can be provided for subsequent improvement. By setting multiple sampling depths, including 0 to 30 cm depth and 30 to 60 cm depth on the ground, information on soil at different depths can be obtained. This helps to analyze the vertical distribution of saline-alkali land soil and guide more accurate improvement strategies. By installing image sensors on the ground at the sampling point, the surface conditions can be monitored in real time. This can timely discover key information such as soil moisture distribution and vegetation growth, providing an effective basis for adjustments and judgments during the improvement process. Through the above methods, saline-alkali land can be improved more scientifically, achieving continuous improvement and restoration of soil, improving soil fertility, increasing crop yields, reducing the degree of salinization of cultivated land, and promoting sustainable agricultural development.

[0184] This embodiment provides a saline-alkali land improvement system based on IoT big data analysis, wherein the sampling module includes:

[0185] Soil condition acquisition module: obtains the soil condition of saline-alkali land through underground sensors; the saline-alkali land soil condition includes salt content and pH value; monitors the changes of soil salt content and pH value through a monitoring system;

[0186] The soil salinity is:

[0187]

[0188] The soil pH value is:

[0189]

[0190] Among them, Y 1i is the salinity value obtained by the sensor at the first depth; 1i The pH value obtained by the sensor at the first depth; Y 2i The salinity value obtained by the sensor at the second depth; 2iis the pH value obtained by the sensor at the second depth; a1 and a2 are coefficients, ranging from (0, 1); m is the number of sampling points of the first or second depth sensor;

[0191] Crop growth status acquisition module: obtains crop growth status in saline-alkali land through image sensors; the sampling frequency of image sensors and underground sensors is consistent;

[0192] Sampling time setting module: set the sampling time interval of underground sensors for sampling; dynamically adjust the sampling time interval according to the sampling results;

[0193] T=(1-λ×avgD)×T0

[0194]

[0195] Where T is the sampling time interval; T0 is the preset sampling time interval; C j-1 is the value sampled by a certain underground sensor last time, C j-2 is the value of the last sampling of a certain underground sensor; ΔC y The change value is preset for this type of sensor; avgD is the average value of D obtained by multiple underground sensors; Min(E) is the minimum E value obtained by multiple underground sensors; Max(E) is the maximum E value obtained by multiple underground sensors.

[0196] The working principle of the above technical solution is: the soil conditions of saline-alkali land, including the salt content and pH value of the soil, are obtained through underground sensors. The sensors are divided into two depths (first depth and second depth), and the salt content and pH value at the corresponding depths are obtained respectively. The changes of soil salt content and pH value are monitored in real time through the monitoring system. The monitoring system can record and display the dynamic changes of soil conditions and understand the soil conditions in real time. The sensor sampling values ​​of multiple depths obtained are weighted averaged according to the weight coefficients to obtain the overall soil salt content and pH value. Among them, the weight coefficients a1 and a2 are determined according to the actual situation and range between 0 and 1; the growth of crops on saline-alkali land is monitored in real time through image sensors. The image sensor can capture information such as the growth status, leaf color, canopy density, etc. of the plant to evaluate the improvement effect of saline-alkali land. According to the sampling results and the preset change amount, the sampling time interval of the sensor is dynamically adjusted. The stability of the soil condition is evaluated by the calculated D value (difference in change). If the stability is high, the sampling time interval can be extended; conversely, if the stability is low, the sampling time interval needs to be shortened. According to the stability evaluation results, the new sampling time interval T is calculated by the formula.

[0197] The effect of the above technical solution is: the data of soil salinity and pH value are obtained through underground sensors, and its changes are monitored in real time through the monitoring system. This enables farmers and agricultural experts to understand the soil conditions of saline-alkali land in real time and take corresponding measures in time. The overall soil salinity and pH value are obtained by calculating the weighted average of the sampling values ​​of multiple depth sensors. This accurate assessment helps to better understand the soil characteristics of saline-alkali land and provide a reference for targeted improvement. The growth of crops on saline-alkali land is monitored in real time using image sensors. By monitoring information such as the growth status, leaf color, and canopy density of plants, problems can be discovered in time and appropriate measures can be taken, which helps to improve crop yield and quality. The stability evaluation index D is calculated based on the sampling results, and the sampling time interval of the sensor is dynamically adjusted through the formula. By dynamically adjusting the sampling time interval, redundant sampling can be reduced as much as possible while ensuring data quality, thereby saving energy and storage space. Due to the reduction of redundant data, the speed of the data processing stage will also increase accordingly. The adjustment of the sampling time interval takes into account the data changes of multiple sensors. In this way, a more accurate data sampling time can be obtained, thereby improving data accuracy. This can capture changes in soil conditions more quickly, adjust measures in time, and improve improvement effects. By real-time monitoring and accurate assessment of soil conditions, as well as dynamically adjusting the sampling time interval according to the sampling results, this method can more accurately guide the improvement of saline-alkali land. Targeted measures can be taken according to specific soil conditions and crop needs to improve the improvement effect, increase crop yields, and reduce the adverse effects of soil salinity on plant growth. In short, this saline-alkali land improvement method based on IoT big data analysis can provide real-time and accurate soil information and crop growth monitoring, helping farmers and agricultural experts to effectively improve the soil conditions of saline-alkali land and increase crop yields and quality.

[0198] This embodiment provides a saline-alkali land improvement system based on IoT big data analysis, wherein the model establishment and improvement scheme adjustment module includes:

[0199] Crop classification module: classify the growth conditions of crops; the classifications include very poor, poor, average, good and excellent;

[0200] Comprehensive evaluation module establishment module: establish a comprehensive evaluation model based on saline-alkali soil conditions and crop growth conditions;

[0201] H=Z×[1-λ×|avgD|×(w1×A+w2×B)

[0202]

[0203] Among them, H is the comprehensive evaluation score; Y 1i is the salinity value obtained by the sensor at the first depth; 1iThe pH value obtained by the sensor at the first depth; Y 2i The salinity value obtained by the sensor at the second depth; 2i The pH value obtained by the sensor at the second depth; Y y is the target maximum salt content; P y is the target maximum pH value; w1 and w2 are weights, ranging from (0, 1); m is the number of sampling points of the first or second depth sensor; a1 and a2 are coefficients, ranging from (0, 1); Z is the crop growth score, ranging from 2 to 10, depending on the classification; very poor is 2, poor is 4, average is 6, good is 8, and excellent is 10;

[0204] Monitoring and optimization module: adjust the improvement plan according to the changes in soil salinity, pH value and evaluation model.

[0205] Compare the changes in soil salinity at different time points or at different depths; if the salinity is found to be gradually decreasing, it means that the improvement measures may have taken effect, and the original improvement plan can be maintained and strengthened; if the change in salinity is not obvious or has increased, it may be necessary to re-evaluate the effect of the improvement plan and consider taking further measures;

[0206] If the pH value gradually approaches the target maximum value during the improvement process, it means that the improvement plan is effective in adjusting the soil pH; if the pH value change is not obvious or deviates from the target, you can consider adjusting the improvement plan in a targeted manner, such as increasing or decreasing the application of acidic and alkaline substances to achieve the ideal pH value.

[0207] By comparing the changes in the comprehensive evaluation scores of the evaluation model at different time points, we can understand the effect of the improvement plan; if the comprehensive evaluation score gradually increases, it means that the improvement plan is effective and the original improvement measures can continue to be implemented; if the change value of the evaluation model shows that the comprehensive evaluation score decreases or fluctuates, the improvement plan may need to be adjusted, such as adjusting the amount of fertilizer or water or adding other improvement measures.

[0208] The working principle of the above technical solution is: the growth conditions of crops on saline-alkali land are classified into very poor, poor, general, good and excellent grades; according to the soil conditions of saline-alkali land and the growth conditions of crops, a comprehensive evaluation model is established to evaluate the improvement effect. The model obtains a comprehensive evaluation score H by calculation, which includes two parts A and B. Part A is obtained by calculating the deviation between the salt content value and pH value obtained by the first depth sensor and the target maximum salt content and pH value, and weighted summing them according to coefficients a1 and a2. Part B is also obtained by calculating the deviation between the salt content value and pH value obtained by the second depth sensor and the target maximum salt content and pH value, and weighted summing them according to coefficients a1 and a2.

[0209] The comprehensive evaluation score H uses weights w1 and w2 to perform weighted summation on A and B to obtain the final evaluation result. According to the classification, the crop growth situation is converted into a crop growth score Z, which ranges from 2 to 10, depending on the classification. For example, very poor corresponds to 2, poor corresponds to 4, general corresponds to 6, good corresponds to 8, and excellent corresponds to 10. By analyzing the change values ​​of soil salt content, pH value, and evaluation model, the improvement plan of saline-alkali land is adjusted. According to the change trend of the comprehensive evaluation score H and the crop growth score Z, the improvement measures can be optimized, such as adjusting the irrigation amount, applying appropriate salt treatment agents, and changing the crop planting method. In short, this method evaluates the improvement effect of saline-alkali land through a comprehensive evaluation model and crop growth score, and adjusts the improvement plan based on data analysis to improve the soil quality and crop yield of saline-alkali land.

[0210] The effect of the above technical solution is: by comprehensively evaluating the soil conditions and crop growth conditions of saline-alkali land, the comprehensive evaluation score H is calculated using parameters such as the salt content value and pH value obtained by the sensor, so as to objectively and accurately evaluate the improvement effect of saline-alkali land. The parameters avgD, λ and weights w1 and w2 in the formula are dynamically adjusted according to the sampling results of the underground sensor, and the improvement of saline-alkali land can be accurately evaluated according to the actual situation, and the importance of different factors can be flexibly adjusted; through the Internet of Things technology and big data analysis, the changes in saline-alkali land soil parameters can be monitored and collected in real time, and the crop growth conditions can be evaluated in real time. In this way, problems can be discovered in time and corresponding measures can be taken to achieve refined management of saline-alkali land. Through the changing trends of the comprehensive evaluation score H and the crop growth score Z, scientific basis and reference can be provided for farmers, agricultural experts and decision makers to guide the formulation of reasonable saline-alkali land improvement plans. According to the change value of the evaluation model, the improvement measures can be adjusted, such as adjusting the irrigation amount, applying appropriate salt treatment agents, changing the crop planting method, etc., to improve the improvement effect. Through the improvement plan for saline-alkali land, parameters such as soil salt content and pH value can be adjusted, the salt content in the soil can be reduced, the soil structure and water retention capacity can be improved, and the growth environment of crops can be improved. This will help to increase the yield and quality of crops. Through accurate data analysis and evaluation, agricultural production factors such as water resources, fertilizers and pesticides can be reasonably utilized to reduce waste and pollution, so as to achieve the purpose of saving resources and protecting the environment. In summary, the saline-alkali land improvement method based on IoT big data analysis can provide accurate evaluation and guidance, realize refined management, improve soil quality and crop yield, and save resources and protect the environment.

[0211] This embodiment provides a saline-alkali land improvement system based on IoT big data analysis, and the crop classification module includes:

[0212] Database crop feature classification module: obtain the growth time series according to the crop growth time, and classify the growth conditions of each growth time series of the same crop to obtain different classification image features;

[0213] Target saline-alkali land crop feature extraction module: extracts the growth image features and time series of the crops corresponding to the target saline-alkali land;

[0214] Sampling point crop classification determination module: compare the image features of the crop corresponding to each sampling point with the features corresponding to different classifications of the same crop in the same time series in the database, and obtain the classification corresponding to the feature with the highest similarity as the classification corresponding to the crop in this sampling point and this time series;

[0215] Final classification module: If different sampling points of the target saline-alkali land in the same time series correspond to multiple classification features, the same classification corresponding to the most sampling points will be used as the final classification of the crop in this time series.

[0216] The working principle of the above technical solution is as follows: first, the growth conditions of the same crop under different growth time series are classified. This can be determined by observing and recording the growth status of the same crop at different growth stages, such as leaf color, plant height, plant morphology, etc. The classification can include very poor, poor, general, good and excellent grades. For each crop growth time series, the growth image features of the corresponding crop are extracted. This can be done by processing and analyzing crop images using computer vision technology to extract features related to crop growth status, such as texture, shape, color, etc. The extracted crop image features are compared with the features corresponding to different classifications of the same crop in the database. By calculating the similarity between the features, the classification corresponding to the features most similar to the crop features to be classified is found. If different test points correspond to multiple classification features, the same classification corresponding to the most test points is selected as the final classification result. This is to increase the accuracy and reliability of the classification, and the results of multiple test points are voted or counted, and the classification with the highest frequency is selected as the final result. Through the above steps, the method uses IoT big data analysis and image feature extraction technology to classify the growth conditions of crops in saline-alkali land. By comparing feature similarity and multi-point detection result statistics, accurate classification results can be obtained, providing a scientific basis for the improvement and management of saline-alkali land.

[0217] The effect of the above technical solution is: by classifying the crop growth time series, the growth status of the crop can be accurately evaluated and described. The crop growth conditions are divided into very poor, poor, general, good and excellent grades, which can more specifically understand the adaptability and growth status of crops in saline-alkali land. The method uses image feature extraction technology to convert the growth image of the crop into a feature vector. This classification method based on image features can more objectively and quantitatively evaluate the growth of crops, avoiding subjective and human interference factors. The image features of the crop are compared with the features in the database, and the similarity is calculated. By comparing the features corresponding to different classifications, the features most similar to the crop to be classified are found, thereby determining the classification of the crop. In this way, the accuracy and reliability of classification can be improved by taking advantage of big data analysis and machine learning algorithms. If different test points correspond to multiple classification features, the method uses the same classification corresponding to the most test points as the final classification result. This method of integrating multi-point detection results can reduce the possible misjudgment of individual test points and improve the credibility and stability of classification results. The saline-alkali land improvement method based on IoT big data analysis can provide agricultural managers and decision makers with accurate crop growth assessment and classification information, helping them better understand the status of saline-alkali land and take corresponding improvement measures. At the same time, this method is based on the analysis of big data and image features, overcoming the limitations of traditional subjective judgment and individual factors, and has higher scientificity and objectivity.

[0218] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A saline-alkali land improvement method based on Internet of Things big data analysis, characterized in that: The method comprises: S1. Obtain the location of saline-alkali land through geographic information system and remote sensing data; collect soil samples of target saline-alkali land, obtain improvement targets and preset improvement plans through sample analysis and knowledge base; divide the target saline-alkali land into multiple regional types; randomly select multiple different types of regions, and set multiple sensor nodes for stratified sampling according to the improvement targets; S2. Obtain the soil conditions of saline-alkali land through underground sensors; perform weighted analysis on the data of multiple sensors to obtain soil salinity and pH value; monitor the changes of soil salinity and pH value through a monitoring system; obtain the crop growth conditions of saline-alkali land through image sensors; set the sampling time interval of the sensor for sampling; dynamically adjust the sampling time interval according to the sampling results; S3. Establish a comprehensive evaluation model based on the soil conditions of saline-alkali land and crop growth conditions, and adjust the improvement plan according to the changes in soil salt content, pH value and the changes in the evaluation model.

2. A saline-alkali land improvement method based on Internet of Things big data analysis according to claim 1, characterized in that: The S1 includes: Obtaining the location of saline-alkali lands through geographic information systems and remote sensing data; Collect soil samples from saline-alkali land and analyze the soil samples to obtain the soil type, salt content and pH value of the corresponding saline-alkali land; Obtain the corresponding saline-alkali land improvement goals and preliminary improvement plans through sample analysis data and knowledge base; Dividing the corresponding saline-alkali land into a plurality of first areas; wherein the first area includes a central area and a peripheral area; Obtaining the altitudes of different first areas of the target saline-alkali land through a geographic information system; Divide the first area into the second area according to the altitude; obtain the highest altitude and the lowest altitude of the target saline-alkali land, divide the height difference between the highest altitude and the lowest altitude into multiple levels, and divide the second area according to the levels; A plurality of second areas of different types are randomly selected, and sensor nodes are set for sampling according to the improved target; the number of second areas of each different type selected is N, 2≤N≤5; The depth from 0 to 30 cm from the ground was taken as the first sampling depth, and the depth from 30 to 60 cm was taken as the second sampling depth; Sampling points at the first sampling depth and the second sampling depth are respectively selected to set up sensor groups for soil sampling; each sampling point includes multiple types of sensors; and an image sensor is set on the ground at the same sampling point.

3. The method for improving saline-alkali land based on Internet of Things big data analysis according to claim 1, characterized in that: The S2 includes: Obtaining soil conditions of saline-alkali land through underground sensors; the saline-alkali land soil conditions include salt content and pH value; monitoring changes in soil salt content and pH value through a monitoring system; The soil salinity is: The soil pH value is: Among them, Y 1i is the salinity value obtained by the sensor at the first depth; 1i The pH value obtained by the sensor at the first depth; Y 2i The salinity value obtained by the sensor at the second depth; 2i is the pH value obtained by the sensor at the second depth; a1 and a2 are coefficients, ranging from (0, 1); m is the number of sampling points of the first or second depth sensor; Obtain crop growth conditions in saline-alkali land through image sensors; Set the sampling time interval of the underground sensor to perform sampling; Dynamically adjust the sampling time interval according to the sampling results; T=(1-λ×avgD)×T0 Where T is the sampling time interval; T0 is the preset sampling time interval; C j-1 is the value sampled by a certain underground sensor last time, C j-2 is the value of the last sampling of a certain underground sensor; ΔC y The change value is preset for this type of sensor; avgD is the average value of D obtained by multiple underground sensors; Min(E) is the minimum E value obtained by multiple underground sensors; Max(E) is the maximum E value obtained by multiple underground sensors.

4. The method for improving saline-alkali land based on Internet of Things big data analysis according to claim 1, characterized in that: The S3 includes: Classifying the growth conditions of crops; the classifications include very poor, poor, fair, good and excellent; Establish a comprehensive evaluation model based on the soil conditions and crop growth conditions of saline-alkali land; H=Z×[1-λ×|avgD|×(w1×A+w2×B)] Among them, H is the comprehensive evaluation score; Y 1i is the salinity value obtained by the sensor at the first depth; 1i The pH value obtained by the sensor at the first depth; Y 2i The salinity value obtained by the sensor at the second depth; 2i The pH value obtained by the sensor at the second depth; Y y is the target maximum salt content; P y is the target maximum pH value; w1 and w2 are weights, ranging from (0, 1); m is the number of sampling points of the first or second depth sensor; a1 and a2 are coefficients, ranging from (0, 1); Z is the crop growth score, ranging from 2 to 10, depending on the classification; very poor is 2, poor is 4, average is 6, good is 8, and excellent is 10; C j-1 is the value sampled by a certain underground sensor last time, C j-2 is the value of the last sampling of a certain underground sensor; ΔC y The preset change value is for this type of sensor; avgD is the average value of D obtained by multiple underground sensors; Min(E) is the minimum E value obtained by multiple underground sensors; Max(E) is the maximum E value obtained by multiple underground sensors; Adjust the improvement plan according to the changes in soil salinity, pH value and evaluation model.

5. A saline-alkali land improvement method based on Internet of Things big data analysis according to claim 4, characterized in that: The classification of crop growth conditions includes: The growth time series is obtained according to the crop growth time, and the growth conditions of each growth time series of the same crop are classified to obtain the image features of different classifications; Extract the growth image features and time series of the crops corresponding to the target saline-alkali land; Compare the image features of the crop corresponding to each sampling point with the features corresponding to different classifications of the same crop in the same time series in the database, and obtain the classification corresponding to the feature with the highest similarity as the classification corresponding to the crop in this sampling point and this time series; If different sampling points of the target saline-alkali land in the same time series correspond to multiple classification features, the same classification corresponding to the most sampling points will be used as the final classification of the crop in this time series.

6. A saline-alkali land improvement system based on IoT big data analysis, characterized in that: The system comprises: Sampling node setting module: obtain the location of saline-alkali land through geographic information system and remote sensing data; collect soil samples of target saline-alkali land, obtain improvement targets and preset improvement plans through sample analysis and knowledge base, and divide the target saline-alkali land into multiple regional types; randomly select multiple different types of regions, and set multiple sensor nodes for stratified sampling according to the improvement targets; Sampling module: obtain the soil conditions of saline-alkali land through underground sensors; perform weighted analysis on the data of multiple sensors to obtain soil salinity and pH value; monitor the changes of soil salinity and pH value through a monitoring system; obtain the growth conditions of crops in saline-alkali land through image sensors; set the sampling time interval of the sensor for sampling; dynamically adjust the sampling time interval according to the sampling results; Model establishment and improvement plan adjustment module: Establish a comprehensive evaluation model based on the soil conditions of saline-alkali land and crop growth conditions, and adjust the improvement plan based on changes in soil salt content, pH value and evaluation model.

7. The saline-alkali land improvement system based on Internet of Things big data analysis according to claim 6 is characterized in that: The sampling node setting module includes: Location acquisition module: obtain the location of saline-alkali land through geographic information system and remote sensing data; Sample analysis module: collect saline-alkali soil samples, analyze the soil samples to obtain the soil type, salt content and pH value of the corresponding saline-alkali land; Target setting module: obtain the corresponding saline-alkali land improvement target and preliminary improvement plan through sample analysis data and knowledge base; The first region division module is used to divide the corresponding saline-alkali land into a plurality of first regions; wherein the first region includes a central region and a peripheral region; The second area division module: obtaining the altitudes of different first areas of the target saline-alkali land through a geographic information system; dividing the first area into second areas according to the altitudes; obtaining the highest altitude and the lowest altitude of the target saline-alkali land, dividing the height difference between the highest altitude and the lowest altitude into multiple levels, and dividing the second area according to the levels; Region selection module: randomly select multiple second regions of different types, and set sensor nodes for sampling according to the improved target; the number of second regions of each different type is N, 2≤N≤5; Sampling depth setting module: the depth from 0 to 30 cm from the ground is used as the first sampling depth, and the depth from 30 to 60 cm is used as the second sampling depth; Layered sampling module: sampling points are selected at the first sampling depth and the second sampling depth to set up sensor groups for soil sampling; each sampling point includes multiple types of sensors; image sensors are set on the ground at the same sampling point.

8. The saline-alkali land improvement system based on Internet of Things big data analysis according to claim 6 is characterized in that: The sampling module comprises: Soil condition acquisition module: obtains the soil condition of saline-alkali land through underground sensors; the saline-alkali land soil condition includes salt content and pH value; monitors the changes of soil salt content and pH value through a monitoring system; The soil salinity is: The soil pH value is: Among them, Y 1i is the salinity value obtained by the sensor at the first depth; 1i The pH value obtained by the sensor at the first depth; Y 2i The salinity value obtained by the sensor at the second depth; 2i is the pH value obtained by the sensor at the second depth; a1 and a2 are coefficients, ranging from (0, 1); m is the number of sampling points of the first or second depth sensor; Crop growth status acquisition module: obtains crop growth status in saline-alkali land through image sensors; Sampling time setting module: set the sampling time interval of underground sensors for sampling; dynamically adjust the sampling time interval according to the sampling results; T=(1-λ×avgD)×T0 Where T is the sampling time interval; T0 is the preset sampling time interval; C j-1 is the value sampled by a certain underground sensor last time, C j-2 is the value of the last sampling of a certain underground sensor; ΔC y The change value is preset for this type of sensor; avgD is the average value of D obtained by multiple underground sensors; Min(E) is the minimum E value obtained by multiple underground sensors; Max(E) is the maximum E value obtained by multiple underground sensors.

9. The saline-alkali land improvement system based on Internet of Things big data analysis according to claim 6 is characterized in that: The model building and improvement scheme adjustment module includes: Crop classification module: classify the growth conditions of crops; the classifications include very poor, poor, average, good and excellent; Comprehensive evaluation module establishment module: establish a comprehensive evaluation model based on saline-alkali soil conditions and crop growth conditions; H=Z×[1-λ×|avgD|×(w1×A+w2×B)] Among them, H is the comprehensive evaluation score; Y 1i is the salinity value obtained by the sensor at the first depth; 1i The pH value obtained by the sensor at the first depth; Y 2i The salinity value obtained by the sensor at the second depth; 2i The pH value obtained by the sensor at the second depth; Y y is the target maximum salt content; P y is the target maximum pH value; w1 and w2 are weights, ranging from (0, 1); m is the number of sampling points of the first or second depth sensor; a1 and a2 are coefficients, ranging from (0, 1); Z is the crop growth score, ranging from 2 to 10, depending on the classification; very poor is 2, poor is 4, average is 6, good is 8, and excellent is 10; C j-1 is the value sampled by a certain underground sensor last time, C j-2 is the value of the last sampling of a certain underground sensor; ΔC y The preset change value is for this type of sensor; avgD is the average value of D obtained by multiple underground sensors; Min(E) is the minimum E value obtained by multiple underground sensors; Max(E) is the maximum E value obtained by multiple underground sensors; Monitoring and optimization module: adjust the improvement plan according to the changes in soil salinity, pH value and evaluation model.

10. The saline-alkali land improvement system based on Internet of Things big data analysis according to claim 9 is characterized in that: The crop classification module comprises: Database crop feature classification module: obtain the growth time series according to the crop growth time, and classify the growth conditions of each growth time series of the same crop to obtain different classification image features; Target saline-alkali land crop feature extraction module: extracts the growth image features and time series of the crops corresponding to the target saline-alkali land; Sampling point crop classification determination module: compare the image features of the crop corresponding to each sampling point with the features corresponding to different classifications of the same crop in the same time series in the database, and obtain the classification corresponding to the feature with the highest similarity as the classification corresponding to the crop in this sampling point and this time series; Final classification module: If different sampling points of the target saline-alkali land in the same time series correspond to multiple classification features, the same classification corresponding to the most sampling points will be used as the final classification of the crop in this time series.

Images