A rice yield prediction system based on remote sensing data analysis

Abstract

The invention belongs to the field of rice yield prediction, relates to remote sensing data analysis technology, and is used to solve the problem of low efficiency of the rice yield prediction analysis process in the prior art. Specifically, it is a rice yield prediction system based on remote sensing data analysis, comprising a rice field basic information module, a remote sensing data analysis module, a historical data matching module, a yield prediction module and a planting optimization module; the rice field basic information module, the remote sensing data analysis module, the historical data matching module, the yield prediction module and the planting optimization module are sequentially communicated and connected; the rice field basic information module is used to collect basic information of the rice field where rice yield estimation is required: the rice planting area is marked as a prediction area, and the planting parameters of the prediction area are obtained; the invention can draw a growth prediction curve of a growth coefficient and a monitoring time point, simplify the traditional rice yield prediction analysis process by comparing the growth prediction curve, and improve the efficiency of the prediction analysis.

Description

Technical Field

[0001] The invention belongs to the field of rice yield prediction, relates to data analysis technology, and specifically is a rice yield prediction system based on remote sensing data analysis. Background Art

[0002] Different rice varieties have significant differences in adaptability, stress resistance and yield. With the continuous advancement of agricultural science and technology, high-yield cultivation techniques are becoming more and more popular, and suitable climatic conditions are conducive to the growth and development of rice, thereby increasing yield.

[0003] Existing rice yield prediction systems often use complex data processing models for prediction and analysis. Complex data calculations are required at each growth stage of the entire rice growth cycle to obtain the final prediction results, resulting in inefficient rice yield prediction and analysis process. In addition, existing technologies are unable to directly perform optimization decision analysis when it is detected that the yield prediction results do not meet expectations, resulting in complicated and inefficient optimization processing procedures when the rice yield prediction results are abnormal.

[0004] In view of the above problems, this application proposes a solution. Summary of the invention

[0005] The purpose of the present invention is to provide a rice yield prediction system based on remote sensing data analysis, which is used to solve the problem of low efficiency of the rice yield prediction and analysis process in the prior art;

[0006] The technical problem to be solved by the present invention is: how to provide a rice yield prediction system based on remote sensing data analysis that can quickly predict the yield of rice at each growth stage.

[0007] The purpose of the present invention can be achieved through the following technical solutions:

[0008] A rice yield prediction system based on remote sensing data analysis, comprising a rice field basic information module, a remote sensing data analysis module, a historical data matching module, a yield prediction module and a planting optimization module; the rice field basic information module, the remote sensing data analysis module, the historical data matching module, the yield prediction module and the planting optimization module are sequentially connected in communication;

[0009] The rice field basic information module is used to collect basic information of the rice fields that need to be estimated for rice yield: after rice sowing and transplanting are completed, the rice planting area is marked as a prediction area, and the planting parameters of the prediction area are obtained;

[0010] The remote sensing data analysis module is used to monitor and analyze the rice growth in the predicted area: generate a rice growth monitoring cycle, set a monitoring time point T in the rice growth monitoring cycle j, j = 1, 2, ..., k, k is a positive integer, at the monitoring time point T j Get the growth coefficient SZ of the prediction area j ; By monitoring time point T j The growth coefficient is plotted to predict the area at the monitoring time point T j Growth prediction curve Q at time

[0011] The historical data matching module is used to analyze and match the historical growth data in the database and obtain a matching historical growth curve PL;

[0012] The yield prediction module is used to predict the rice yield in the prediction area according to the matching historical growth curve PL and obtain the yield prediction value YC of the prediction area;

[0013] The planting optimization module is used to optimize the rice planting process in the prediction area according to the yield prediction value YC.

[0014] Furthermore, the planting parameters include the planting area S, direct comparison parameters and indirect comparison parameters: the direct comparison parameters include the rice variety and soil type, and the indirect comparison parameters include the altitude value and the average annual precipitation.

[0015] Furthermore, the monitoring time point T j Growth coefficient SZ j The acquisition process includes: dividing the prediction area into analysis areas i, i = 1, 2, ..., n, n is a positive integer, at the monitoring time point T j Through the remote sensing platform, the monitoring time point T of each analysis area i is obtained j The ear distance, the length of the mature leaf and the plant height were calculated by summing up the ear distance, the length of the mature leaf and the plant height, and the average value was calculated to obtain the predicted rice in the monitoring area at the monitoring time point T. j The average value of the leaf ear distance YJ j , average leaf length of fixed leaves DY j And the average plant height ZG j , through the formula SZ j =k1*YJ j +k2*DY j +k3*ZG j The predicted area at the monitoring time point T is obtained through numerical calculation. j Growth coefficient SZ j .

[0016] Furthermore, the prediction area is at the monitoring time point T j The specific process of drawing the growth prediction curve includes: j As the Y axis of the coordinate system, the rice growth monitoring period as the X axis of the coordinate system to establish a rectangular coordinate system.j The execution time in the rice growth monitoring cycle is taken as the horizontal axis, and the monitoring time point T j The corresponding growth coefficient SZ j Mark j monitoring points in the rectangular coordinate system as the ordinate, use a smooth curve to connect all the monitoring points, and mark the connected curve as the growth prediction curve Q of the prediction area.

[0017] Furthermore, the specific process of the historical data matching module analyzing and matching the historical growth data in the database includes: retrieving the historical growth data of all rice from the database, the historical growth data including the historical planting parameters of the rice-growing area, the historical growth curve of rice and the final actual rice yield; comparing the planting parameters of the predicted area with the historical planting parameters: wherein, the comparison method of the direct comparison parameters is the type comparison method, if the type data names of the direct comparison parameters are consistent, the first-level matching data of the historical growth data is obtained; the comparison method of the indirect comparison parameters is the numerical interval comparison method, the planting parameters are * the relevant confidence interval (1±a%) to obtain the relevant parameter range, if the indirect comparison parameters in the first-level matching data are within the relevant parameter range, the second-level matching data of the historical growth data is obtained, and the historical growth curve corresponding to the second-level matching data is marked as the matching historical growth curve PL.

[0018] Furthermore, the process of obtaining the predicted yield value YC of the predicted area includes: intercepting the matching historical growth curve PL from left to right, and the interception time is the monitoring time point T j The time difference between the start time of the rice growth monitoring period and the historical growth curve PL after interception and the growth prediction curve Q are plotted in the same coordinate system, and the coincidence coefficient of all the historical growth curves PL after interception and the growth prediction curve Q is calculated: select several points on the X-axis and mark them as the comparison time point h, h = 1, 2, ... s, s is a positive integer, and obtain the historical growth coefficient LS of the matching historical growth curve PL at the time point h h The growth coefficient SZ at time point h with the predicted growth curve Q h , through the formula , the historical growth coefficient LS h and growth coefficient SZ h Numerical calculation is performed to obtain the overlap coefficient CH between each matching historical growth curve PL and the predicted growth curve Q. The smaller the overlap coefficient CH, the higher the overlap. The actual rice yield corresponding to the matching historical growth curve PL with the highest overlap is marked as the unit yield prediction value DY of the prediction area. The unit yield prediction value DY and the planting area S are multiplied to obtain the yield prediction value YC of the prediction area.

[0019] Furthermore, the specific process of the planting optimization module optimizing the rice planting process in the prediction area according to the yield prediction value YC includes: judging whether to optimize the planting process by comparing the yield prediction value YC with the preset yield prediction threshold value YZ, and optimizing the planting process when the yield prediction value does not meet the requirements: arranging all the matching historical growth curves PL in the prediction area in order of the overlap coefficient CH with the prediction growth curve Q from small to large, selecting the M matching historical growth curves PL with the top overlap coefficient CH and obtaining the corresponding actual rice yields, marking the matching historical growth curve PL with the largest actual rice yield as the optimization curve of the prediction area, and setting the current monitoring time point T of the rice growth monitoring period to the optimal time point of the prediction area. j Marked as the optimization time point, will be at the optimization time point T j The period from D1 day before to D2 day after is marked as the optimization analysis period, and all processing means implemented in the optimization analysis period in the historical growth data corresponding to the optimization curve are marked as optimization means for the prediction area, and the optimization means are sent to the terminal of the manager.

[0020] Furthermore, the specific process of judging whether to optimize the planting process includes: if the yield prediction value YC is greater than or equal to the yield prediction threshold YZ, it is determined that the yield prediction result meets the requirements and there is no need to optimize the planting process; if the yield prediction value YC is less than the yield prediction threshold YZ, it is determined that the yield prediction result does not meet the requirements and planting process optimization is required.

[0021] The present invention has the following beneficial effects:

[0022] The rice field basic information module can be used to obtain the planting area, rice variety and natural conditions of the rice field planting area, such as soil type, altitude and average precipitation. Then, according to the planting information, the data most similar to the rice planting environment for predicted yield can be screened from the historical planting data, eliminating the interference of external factors on rice yield prediction, and improving the accuracy of rice yield prediction. In the rice growth monitoring cycle, the remote sensing platform is used to collect, process and analyze the high-resolution image data of the rice planting area to obtain the growth coefficient of the rice planting area at the monitoring time point, and the growth prediction curve of the growth coefficient and the monitoring time point is drawn. The traditional rice yield prediction analysis process is simplified by comparing the growth prediction curve, and the efficiency of prediction analysis is improved. The planting optimization module can be used to determine whether the yield prediction value has reached the yield expectation. If not, the planting method optimization is triggered, and all the processing methods implemented during the optimization analysis period of the historical growth data with the highest overlap are retrieved as the optimization methods for the current rice planting area, ensuring that the rice is in the best growth state during the rice life monitoring cycle to achieve the yield expectation. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0024] Figure 1 The figure is a system block diagram of the present invention as a whole. DETAILED DESCRIPTION

[0025] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only 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.

[0026] like Figure 1 As shown, a rice yield prediction system based on remote sensing data analysis includes a rice field basic information module, a remote sensing data analysis module, a historical data matching module, a yield prediction module and a planting optimization module; the rice field basic information module, the remote sensing data analysis module, the historical data matching module, the yield prediction module and the planting optimization module are communicatively connected in sequence.

[0027] The rice field basic information module is used to collect basic information of rice fields that need to estimate rice yield: after rice sowing and transplanting, the rice planting area is marked as a prediction area, and the planting parameters of the prediction area are obtained. The planting parameters include the planting area S, direct comparison parameters and indirect comparison parameters: direct comparison parameters include rice variety and soil type, and indirect comparison parameters include altitude and average annual precipitation; for example, the rice variety planted is indica rice, the soil type is black soil, the altitude is 500m, and the average precipitation is 1200mm.

[0028] The remote sensing data analysis module is used to monitor and analyze the rice growth in the predicted area: generate a rice growth monitoring cycle, set the monitoring time point T within the rice growth monitoring cycle j , j = 1, 2, ..., k, k is a positive integer, the prediction area is divided into analysis areas i, i = 1, 2, ..., n, n is a positive integer, at the monitoring time point T jThe high-resolution image data of rice in each analysis area i is obtained through the remote sensing platform. The high-resolution rice data is preprocessed and feature extracted to obtain reflectance data, leaf area index and vegetation index in different bands. The ear distance and the length of the fixed leaf are obtained through inversion of the physical model. The height of the rice plant is obtained by the visible light camera equipped with the UAV platform. j The leaf ear distance, the length of the mature leaf and the plant height are summed and averaged to obtain the rice height at the monitoring time point T in the predicted area. j The average value of the leaf ear distance YJ j , average leaf length of fixed leaves DY j And the average plant height ZG j , through the formula SZ j =k1*YJ j +k2*DY j +k3*ZG j The numerical calculation shows that the rice in the predicted area at the monitoring time point T j Growth coefficient SZ j ; The growth coefficient SZ j As the Y axis of the coordinate system, the rice growth monitoring period as the X axis of the coordinate system to establish a rectangular coordinate system. j The execution time in the rice growth monitoring cycle is taken as the horizontal axis, and the monitoring time point T j The corresponding growth coefficient SZ j Mark j monitoring points in the rectangular coordinate system as the ordinate, use a smooth curve to connect all the monitoring points, and mark the connected curve as the growth prediction curve Q of the prediction area.

[0029] The historical data matching module is used to analyze and match the historical growth data in the database: retrieve all the historical growth data of rice from the database, the historical growth data includes the historical planting parameters of the rice planting area, the historical growth curve of rice and the final actual rice yield. In particular, the historical growth curve is a smooth curve drawn by the growth coefficient of all monitoring time points in the rice growth monitoring cycle and all monitoring time points. The final actual rice yield is the ratio of the total yield corresponding to the end of rice growth to the planting area; compare the planting parameters of the predicted area with the historical planting parameters: among them, the comparison method of the direct comparison parameters is the type comparison method. If the type data name of the direct comparison parameters is the same, then The primary matching data of the historical growth data is obtained; the indirect matching parameter is matched by the numerical interval matching method, the planting parameter * the relevant confidence interval (1±a%) is used to obtain the relevant parameter range, if the indirect matching parameter in the primary matching data is within the relevant parameter range, the secondary matching data of the historical growth data is obtained, and the historical growth curve corresponding to the secondary matching data is marked as the matching historical growth curve PL; illustratively, the altitude value HB* (1-a%) is used to obtain the low altitude value, and the altitude value HB* (1+a%) is used to obtain the high altitude value, the interval included by the low altitude value and the high altitude value is the relevant parameter range of the altitude value, wherein a is the numerical constant corresponding to the relevant confidence interval of the altitude value, and the value range is (5-15).

[0030] The yield prediction module is used to predict the rice yield of the prediction area according to the matching historical growth curve PL: the matching historical growth curve PL is intercepted from left to right, and the interception time is the monitoring time point T j The time difference between the start time of the rice growth monitoring period and the historical growth curve PL after interception and the growth prediction curve Q are plotted in the same coordinate system, and the coincidence coefficient of all the historical growth curves PL after interception and the growth prediction curve Q is calculated: select several points on the X-axis and mark them as the comparison time point h, h=1, 2, ...s, s is a positive integer, and obtain the historical growth coefficient LS of the matching historical growth curve PL at the time point h h The growth coefficient SZ at time point h with the predicted growth curve Q h , through the formula , the historical growth coefficient LS h and growth coefficient SZ h Numerical calculation is performed to obtain the overlap coefficient CH between each matching historical growth curve PL and the predicted growth curve Q. The smaller the overlap coefficient CH, the higher the overlap. The actual rice yield corresponding to the matching historical growth curve PL with the highest overlap is marked as the unit yield prediction value DY of the prediction area. The unit yield prediction value DY and the planting area S are multiplied to obtain the yield prediction value YC of the prediction area.

[0031] The planting optimization module is used to optimize the rice planting process in the prediction area according to the yield prediction value YC: by comparing the yield prediction value YC with the preset yield prediction threshold value YZ, it is judged whether to optimize the planting process: if the yield prediction value YC is greater than or equal to the yield prediction threshold value YZ, it is judged that the yield prediction result meets the requirements and there is no need to optimize the planting process; if the yield prediction value YC is less than the yield prediction threshold value YZ, it is judged that the yield prediction result does not meet the requirements and it is necessary to optimize the planting process; all matching historical growth curves PL in the prediction area are arranged in order from small to large according to the overlap coefficient CH with the prediction growth curve Q, and the M matching historical growth curves PL with the top overlap coefficient CH are selected and the corresponding actual rice yields are obtained, and the matching historical growth curve PL with the largest actual rice yield is marked as the optimization curve of the prediction area, and the current monitoring time point T of the rice growth monitoring period is set as the optimization curve. j Marked as the optimization time point, will be at the optimization time point T j The first D1 day to the last D2 days are marked as the optimization analysis period, and all the treatment means implemented in the optimization analysis period in the historical growth data corresponding to the optimization curve are marked as the optimization means of the prediction area. Exemplarily, the treatment means include: irrigation, drainage, insecticide, weeding and application of organic fertilizer.

[0032] A rice yield prediction system based on remote sensing data analysis, when working, collects basic information of paddy fields for which rice yield estimation is required; after rice sowing and transplanting are completed, monitors and analyzes the rice growth situation in the prediction area during the rice growth monitoring period and obtains a growth prediction curve Q of the prediction area; performs planting parameter analysis and matching on historical growth data in a database and obtains a matching historical growth curve PL; predicts the rice yield in the prediction area and obtains a yield prediction value YC of the prediction area; and optimizes the rice planting process in the prediction area according to the yield prediction value YC.

[0033] The above contents are merely examples and explanations of the structure of the present invention. The technicians in this technical field may make various modifications or additions to the specific embodiments described or replace them in a similar manner. As long as they do not deviate from the structure of the invention or exceed the scope defined by the claims, they should all fall within the protection scope of the present invention.

[0034] The above formulas are obtained by collecting a large amount of data and performing software simulation to select a formula close to the actual value. The coefficients in the formula are set by technicians in this field according to actual conditions; for example: Formula SZ j =k1*YJ j +k2*DY j +k3*ZG j; A technician in this field collects multiple groups of sample data and sets a corresponding growth coefficient for each group of sample data; Substitute the set growth coefficient and the collected sample data into the formula, any three formulas constitute a three-variable linear equation system, screen the calculated coefficients and take the average, and obtain the values ​​of k1, k2 and k3 as 3.49, 2.82 and 2.67 respectively;

[0035] The size of the coefficient is to quantify each parameter to obtain a specific value for subsequent comparison. The size of the coefficient depends on the amount of sample data and the initial setting of the corresponding growth coefficient for each set of sample data by technical personnel in this field; as long as it does not affect the proportional relationship between the parameter and the quantified value, such as the growth coefficient is proportional to the average value of the leaf ear distance.

[0036] In the description of this specification, the description with reference to the terms "one embodiment", "example", "specific example", etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.

[0037] The preferred embodiments of the present invention disclosed above are only used to help explain the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to only specific implementation methods. Obviously, many modifications and changes can be made according to the content of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present invention, so that those skilled in the art can understand and use the present invention well. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. A rice yield prediction system based on remote sensing data analysis, characterized in that: It includes a rice field basic information module, a remote sensing data analysis module, a historical data matching module, a yield prediction module and a planting optimization module; the rice field basic information module, the remote sensing data analysis module, the historical data matching module, the yield prediction module and the planting optimization module are sequentially connected in communication; The rice field basic information module is used to collect basic information of the rice fields that need to be estimated for rice yield: after rice sowing and transplanting are completed, the rice planting area is marked as a prediction area, and the planting parameters of the prediction area are obtained; The remote sensing data analysis module is used to monitor and analyze the rice growth in the predicted area: generate a rice growth monitoring cycle, set a monitoring time point T in the rice growth monitoring cycle j , j = 1, 2, ..., k, k is a positive integer, at the monitoring time point T j Get the growth coefficient SZ of the prediction area j ; By monitoring time point T j The growth coefficient is plotted to predict the area at the monitoring time point T j Growth prediction curve Q at time The historical data matching module is used to analyze and match the historical growth data in the database and obtain a matching historical growth curve PL; The yield prediction module is used to predict the rice yield in the prediction area according to the matching historical growth curve PL and obtain the yield prediction value YC of the prediction area; The planting optimization module is used to optimize the rice planting process in the prediction area according to the yield prediction value YC; Monitoring time point T j The growth coefficient SZ j The acquisition process includes: dividing the prediction area into analysis areas i, i = 1, 2, ..., n, n is a positive integer, at the monitoring time point T j Through the remote sensing platform, the monitoring time point T of each analysis area i is obtained j The ear distance, the length of the mature leaf and the plant height were calculated by summing up the ear distance, the length of the mature leaf and the plant height, and the average value was calculated to obtain the predicted rice in the monitoring area at the monitoring time point T. j The average value of the leaf ear distance YJ j , average leaf length of fixed leaves DY j And the average plant height ZG j , through the formula SZ j =k1*YJ j +k2*DY j +k3*ZG j The predicted area at the monitoring time point T is obtained through numerical calculation. j The growth coefficient SZ j ; The specific process of the historical data matching module analyzing and matching the historical growth data in the database includes: retrieving all the historical growth data of rice from the database, the historical growth data including the historical planting parameters of the rice planting area, the historical growth curve of rice and the final actual rice yield; comparing the planting parameters of the predicted area with the historical planting parameters: wherein, the comparison method of the direct comparison parameters is the type comparison method, if the type data names of the direct comparison parameters are consistent, the first-level matching data of the historical growth data is obtained; the comparison method of the indirect comparison parameters is the numerical interval comparison method, the planting parameters are * the relevant confidence interval (1±a%) to obtain the relevant parameter range, if the indirect comparison parameters in the first-level matching data are within the relevant parameter range, the second-level matching data of the historical growth data is obtained, and the historical growth curve corresponding to the second-level matching data is marked as the matching historical growth curve PL; The process of obtaining the predicted yield value YC of the predicted area includes: intercepting the matching historical growth curve PL from left to right, and the interception time is the monitoring time point T j The time difference between the start time of the rice growth monitoring period and the time difference between the intercepted matching historical growth curve PL and the growth prediction curve Q is plotted in the same coordinate system, and the coincidence coefficient CH of all intercepted matching historical growth curves PL and the growth prediction curve Q is calculated: select several points on the X-axis and mark them as comparison time points h, h=1, 2, …s, s is a positive integer, and obtain the historical growth coefficient LS of the matching historical growth curve PL at time point h h The growth coefficient SZ at time point h with the predicted growth curve Q h , through the formula , the historical growth coefficient LS h and growth coefficient SZ h Numerical calculation is performed to obtain the overlap coefficient CH between each matching historical growth curve PL and the predicted growth curve Q. The actual rice yield corresponding to the matching historical growth curve PL with the smallest overlap coefficient CH is marked as the unit yield prediction value DY of the prediction area. The unit yield prediction value DY and the planting area S are multiplied to obtain the yield prediction value YC of the prediction area.

2. A rice yield prediction system based on remote sensing data analysis according to claim 1, characterized in that: Planting parameters include planting area S, direct comparison parameters and indirect comparison parameters: direct comparison parameters include rice variety and soil type, and indirect comparison parameters include altitude and average annual precipitation.

3. A rice yield prediction system based on remote sensing data analysis according to claim 2, characterized in that: Prediction area at monitoring time point T j The specific process of drawing the growth prediction curve includes: j As the Y axis of the coordinate system, the rice growth monitoring period as the X axis of the coordinate system to establish a rectangular coordinate system. j The execution time in the rice growth monitoring cycle is taken as the horizontal axis, and the monitoring time point T j The corresponding growth coefficient SZ j Mark j monitoring points in the rectangular coordinate system as the ordinate, use a smooth curve to connect all the monitoring points, and mark the connected curve as the growth prediction curve Q of the prediction area.

4. A rice yield prediction system based on remote sensing data analysis according to claim 3, characterized in that: The specific process of the planting optimization module optimizing the rice planting process in the prediction area according to the yield prediction value YC includes: judging whether to optimize the planting process by comparing the yield prediction value YC with the preset yield prediction threshold value YZ, and optimizing the planting process when the yield prediction value does not meet the requirements: arranging all the matching historical growth curves PL in the prediction area in order of the overlap coefficient CH with the prediction growth curve Q from small to large, selecting the M matching historical growth curves PL with the top overlap coefficient CH and obtaining their corresponding actual rice yields, marking the matching historical growth curve PL with the largest actual rice yield as the optimization curve of the prediction area, and setting the current monitoring time point T of the rice growth monitoring period as the optimization curve. j Marked as the optimization time point, will be at the optimization time point T j The period from D1 day before to D2 day after is marked as the optimization analysis period, and all processing means implemented in the optimization analysis period in the historical growth data corresponding to the optimization curve are marked as optimization means for the prediction area, and the optimization means are sent to the terminal of the manager.

5. A rice yield prediction system based on remote sensing data analysis according to claim 4, characterized in that: The specific process of judging whether to optimize the planting process includes: if the yield prediction value YC is greater than or equal to the yield prediction threshold YZ, it is determined that the yield prediction result meets the requirements and there is no need to optimize the planting process; if the yield prediction value YC is less than the yield prediction threshold YZ, it is determined that the yield prediction result does not meet the requirements and planting process optimization is required.

Images