Big-data-based smart-farming intelligent monitoring system and method

Abstract

The present application relates to the technical field of smart farming. Disclosed are a big-data–based smart-farming intelligent monitoring system and method. The system comprises: a farming data monitoring module, a crop growth prediction module, and an agricultural-machinery harvesting assignment module, wherein the farming data monitoring module uses remote sensing technology to monitor image information of all farmland and crops planted within a supervised region, perform grid-based partitioning on the farmland, and analyzes and calculates crop maturity; the crop growth prediction module is used for constructing a crop growth state prediction model, and predicting an estimated crop maturity of farmland crops at a future time point; and the agricultural-machinery harvesting assignment module calculates the harvesting performance of agricultural machinery in respective harvesting regions assigned thereto, and performs analysis to acquire an agricultural-machinery assignment strategy that maximizes a total harvesting performance for all farmland within the supervised region.

Description

A Smart Agriculture Intelligent Monitoring System and Method Based on Big Data Technical Field

[0001] This invention relates to the field of smart agriculture technology, specifically a smart agriculture intelligent monitoring system and method based on big data. Background Technology

[0002] The continuous advancement of agricultural mechanization and management modernization is driving agriculture towards clustered and large-scale production. However, with the ongoing trend of agricultural land transfer in my country, and the large-scale centralized and small-scale dispersed farmland structure, how to make precise harvesting decisions for mainstream crops such as rice and wheat based on different planting times through intelligent monitoring of agricultural crop information, and maximize harvesting efficiency, is a critical issue in the current development of modern agriculture. Summary of the Invention

[0003] The purpose of this invention is to provide a smart agriculture intelligent monitoring system and method based on big data, so as to solve the problems raised in the prior art.

[0004] To achieve the above objectives, the present invention provides the following technical solution: A smart agriculture intelligent monitoring method based on big data, comprising the following analysis steps: Step S100: Set the width of the farmland grid unit, divide all farmland into grid units, monitor farmland image information through remote sensing technology, and measure the fill degree in each farmland grid unit; Step S200: Obtain historical remote sensing image data of each time period during crop growth in each farmland grid unit through remote sensing technology, record the acquisition time point of the historical remote sensing image data, synchronously obtain the final harvest time data of crops in the historical remote sensing image data, calculate the crop maturity data of each farmland grid unit in the historical remote sensing image data, and construct a crop growth status prediction model; Step S300: Predict the expected maturity period of crops in each farmland grid unit, obtain meteorological parameter forecast information for future time periods, and analyze the harvest deadline of crops in each farmland grid unit; Step S400: Use an external navigation interface to monitor and summarize the location data of all agricultural machinery and the information of traversable roads within the management range in real time, and analyze the time required for each agricultural machinery to reach each farmland grid unit; Step S500: Calculate the harvesting efficiency of each agricultural machine allocated to each farmland grid unit at each time node, and use dynamic programming to analyze the strategy that maximizes the total harvesting efficiency of all farmland in the monitored area.

[0005] In the above technical solution, step S100 includes the following: setting the width 'a' of the farmland grid cell, wherein the farmland grid cell is square with a side length of 'a', and tiling all farmland areas to ensure that all farmland areas are covered by the farmland grid cells; acquiring remote sensing image data of the farmland areas, and for irregular areas at the edges of the farmland areas, according to the formula: Calculate the fill degree of the grid cells covering the farmland in this area; where r is the fill degree of the grid cells covering the farmland in this area, and n f To cover the number of pixels contained in the farmland area in the remote sensing image of the farmland grid cell, where 'a' is the width of the farmland grid cell and 'n' is the number of pixels in the grid cell. u This refers to the number of pixels contained in a unit area of ​​remote sensing image. By using remote sensing technology to acquire images of farmland within the monitored area and segment them into grids, the accurate collection and analysis of farmland area information under the trend of large-scale agricultural development is ensured, providing a good data foundation for subsequent strategic decisions. When processing farmland area image information, pixel-level fill degree calculation is adopted, resulting in more detailed feedback on farmland area information.

[0006] In the above technical solution, step S200 includes the following: Extracting historical remote sensing image data of crop growth in each farmland grid cell, obtaining the reflectance of each band's reflectance spectrum, calculating the crop maturity in each farmland grid cell from the historical remote sensing image data at different time points, and for any farmland grid cell x, according to the formula: Among them, M x Let n be the crop maturity level of the farmland grid unit x. f,x Let be the number of pixels contained in the farmland region within the farmland grid cell x, i be the pixel number contained in the farmland region within the farmland grid cell x, G be the digital amplification gain, and NIR be the digital gain. i Let RED be the near-infrared reflectance of pixel i. i For pixel i, the red band reflectance is BLUE. i Let M be the blue light band reflectance of pixel i, C1 and C2 refer to the atmospheric scattering correction coefficients for the red and blue light bands, respectively, and L be the ground reflectance correction coefficient. For different crop types, crop growth status curves are fitted using the interval between remote sensing data acquisition time points and final harvest time points, as well as crop maturity, to obtain a crop growth status prediction model. The model is as follows: M pre,x =A×e k×t +B; where M pre，x The model predicts crop maturity for each farmland grid cell x, where A and B are prediction model parameters, e is the natural logarithm, k is the current crop growth rate coefficient, and t is the time difference between the prediction time and the final harvest. The model uses spectral splitting to extract reflected light wave information from different bands of the farmland crop remote sensing image. Combined with the enhanced vegetation index, it analyzes the growth and maturity status of the farmland crops, providing effective image feedback on the gradual maturation process of the crops. Compared to the normalized vegetation index, it also has stronger resistance to atmospheric interference, further enabling accurate assessment of crop maturity.

[0007] In the above technical solution, step S300 includes the following: Setting crop maturity threshold ranges according to crop type; for any farmland grid unit, calculating the crop maturity at the current time point; using the crop growth status prediction model to calculate the estimated time difference between crop maturity and harvest, and then analyzing the estimated crop maturity at future time points; calculating the earliest time point in the current farmland grid unit where the estimated crop maturity falls within the maturity threshold range, and setting this time point as the initial maturity time point of the farmland grid unit; calculating the latest time point in the current farmland grid unit where the estimated crop maturity falls within the maturity threshold range, and setting this time point as the cutoff maturity time point of the farmland grid unit; setting the period from the initial maturity time point to the cutoff maturity time point as farmland grid unit x cutoff harvest period T. x , denoted as T x [t x，1 , t x，2 ]; where t x，1 The left threshold for the harvest deadline of farmland grid cell x represents the initial maturity time of farmland grid cell x, t. x,2 The left threshold for the harvest deadline of farmland grid unit x represents the cutoff maturity time of farmland grid unit x. Meteorological parameter forecast information for all monitored farmland areas within the harvest deadline is obtained, and parameters affecting the harvesting of crops in each farmland grid unit are screened, setting interference harvesting threshold ranges. For any farmland grid unit, if the value of any screened parameter at any time point within the harvest deadline falls within the interference harvesting threshold range, the current farmland grid unit's harvest deadline is corrected to the time point from the initial maturity time to the time point where the parameter value falls within the interference harvesting threshold range. The harvesting window period for crop maturity in each farmland area is fully considered, and the negative interference of weather changes on crop harvesting is analyzed simultaneously, further limiting the crop harvesting time and providing an analytical data basis for subsequent agricultural machinery harvesting allocation.

[0008] In the above technical solution, step S500 includes the following: Obtaining the harvest deadline data for each monitored farmland grid unit, and classifying all farmland grid units according to the left threshold t of the harvest deadline. x，1The process involves sorting the farmland grid units and setting harvest time windows. The sorted grid units are then selected using these windows, and those within the same window are grouped into the same harvest batch. Furthermore, geographically adjacent grid units within the same harvest batch are grouped into the same harvest area. During harvest strategy analysis, the harvest time windows for each harvest area are obtained. When a time point falls within any given window, the machinery allocation strategy for all harvest areas within that window is analyzed. For machinery currently harvesting, harvesting effectiveness is calculated only after the expected completion time for the harvesting operation. Geographically adjacent grid units with similar maturity times are grouped into regions to ensure detailed crop harvesting allocation while reducing redundant travel between grid units by machinery. This prevents overcorrection caused by precise harvesting strategies, which could lead to redundant traffic losses for machinery traveling between grid units.

[0009] In the above technical solution, the method for calculating the harvesting effectiveness of different agricultural machines allocated to each harvesting area at each time node and the method for analyzing the strategy to maximize the total harvesting effectiveness in step S500 are as follows: Calculate the harvesting effectiveness of different agricultural machines allocated to each harvesting area at each time node. For any agricultural machine m and any harvesting area p, according to the formula: W tra =k1×t tra +k2×W F Among them, W m，p The harvesting efficiency of agricultural machinery m allocated to harvesting area p, W tra Let ε be the traffic loss of agricultural machinery m traveling to the nearest farmland grid cell in the harvesting area p, ε be a constant parameter to avoid a denominator of 0, e be the natural logarithm, α be the time scale scaling factor, and t be the distance between the agricultural machinery m and the harvesting area p. gap S is the time difference between the current time point and the earliest time point of the right threshold of the harvest deadline for the farmland grid cell in the harvesting area p. p Let n be the area of ​​harvesting region a, j be the farmland grid cell number in harvesting region p, and n be the area of ​​harvesting region a. j Let be the number of farmland grid cells in the harvesting area p, 'a' be the side length of the farmland grid cell, 'T' be the time taken by agricultural machinery m to harvest one farmland grid cell in days, and 'r' be the number of farmland grid cells in the harvesting area p. j Let k1 and k2 be the fill degree of farmland grid cell j, and k1 and k2 be the traffic loss assessment parameters for agricultural machinery m. tra W represents the travel time, in hours, for agricultural machinery m to reach the nearest farmland grid cell in the harvesting area p. FThe fuel consumption of agricultural machinery m when traveling to the nearest farmland grid unit in harvesting area p is calculated. Then, the harvesting results of all harvesting areas are summed to obtain the total harvesting result within the monitored area, and dynamic programming is used to analyze the strategy for maximizing the total harvesting result. Combining the travel time and fuel consumption of agricultural machinery when assigned to different harvesting areas, the area of ​​each assigned harvesting area is analyzed simultaneously to maximize the harvesting area completed by the agricultural machinery in a single area transfer. A parameter t is also added. gap This ensures that when analyzing the harvesting results of each harvesting area simultaneously, losses caused by prolonged periods of unharvested crops after they have matured are avoided.

[0010] This invention relates to a smart agriculture intelligent monitoring system based on big data, employing the big data-based intelligent monitoring method described in the above technical solution. The system comprises: an agricultural data monitoring module, a crop growth prediction module, and an agricultural machinery harvesting allocation module. The agricultural data monitoring module uses remote sensing technology to monitor image information of all farmland and planted crops within the monitored area, divides the farmland into grids, and performs crop maturity analysis and calculation. The crop growth prediction module constructs a crop growth status prediction model and predicts the expected crop maturity at future time points, analyzing the harvest deadline for each farmland grid unit. The agricultural machinery harvesting allocation module calculates the harvesting effectiveness of each agricultural machine allocated to its respective harvesting area and analyzes the agricultural machinery allocation strategy that maximizes the total harvesting effectiveness of all farmland within the monitored area.

[0011] In the above technical solution, the agricultural data monitoring module includes: a remote sensing monitoring unit, a farmland area segmentation unit, and a crop monitoring unit; the remote sensing monitoring unit is used to acquire remote sensing image data of farmland areas; the farmland area segmentation unit is used to perform grid-based segmentation of farmland areas; and the crop monitoring unit is used to monitor the crop growth status within the monitored area and calculate crop maturity data.

[0012] In the above technical solution, the crop growth prediction module includes: a model building unit, a crop maturity prediction unit, and a harvest deadline analysis unit; the model building unit is used to build a crop growth status prediction model; the crop maturity prediction unit is used to calculate and analyze the expected maturity of the crop; and the harvest deadline analysis unit is used to correct the harvest deadline of the grown crop based on the future weather parameters of each farmland grid unit.

[0013] In the above technical solution, the agricultural machinery harvesting allocation module includes: a harvesting area division unit, a harvesting effectiveness calculation unit, and an agricultural machinery allocation unit; the harvesting area division unit divides the farmland within the monitoring area into different harvesting areas according to the expected maturity of crops in each farmland grid unit; the harvesting effectiveness calculation unit is used to calculate the harvesting effectiveness of different agricultural machinery allocated to each harvesting area; the agricultural machinery allocation unit is used to analyze the strategy for maximizing the total harvesting effectiveness of all farmland within the monitoring area and to allocate agricultural machinery.

[0014] Compared with existing technologies, the beneficial effects of this invention are as follows: This invention uses remote sensing technology to conduct large-scale monitoring of farmland within a monitored area. By dividing the farmland into grids, it achieves area-based monitoring and management of crops. Then, based on the processing of remote sensing image data, spectral reflectance analysis results are obtained to refine the analysis and reflect the growth and maturity status of crops in the farmland. In this invention, a model is constructed based on the growth and maturity status of crops in each farmland grid unit to predict the expected maturity of crops in each grid unit, thereby predicting the maturity time of crops in each grid unit. Furthermore, by combining future meteorological parameter changes, the harvest deadline for each grid unit is limited, providing a data foundation for timed harvesting of crops in different farmland areas. Specifically, by setting time windows to select and divide farmland grid units into harvesting areas, precise grid division is avoided. Grid analysis leads to over-allocation of agricultural machinery, causing meaningless back-and-forth travel between areas and resulting in unnecessary traffic losses. This invention also calculates the harvesting effectiveness of agricultural machinery allocated to each harvesting area at different time points. It introduces the time difference between the calculation time point and the earliest right threshold of the harvest deadline of the grid unit in the harvesting area, as well as the area of ​​the harvesting area, as calculation parameters. This ensures that the harvesting area with a shorter harvesting time and a larger overall area has a higher allocation weight when calculating the harvesting effectiveness and allocating agricultural machinery. This allows the invention to take into account both the overall and regional harvesting effectiveness in large-scale farmland harvesting, making it more applicable and practical. Attached Figure Description

[0015] Figure 1 is a flowchart of a smart agriculture intelligent monitoring method based on big data according to the present invention; Figure 2 is an organizational structure diagram of a smart agriculture intelligent monitoring system based on big data according to the present invention. Detailed Implementation

[0016] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0017] Example: Please refer to Figures 1-2. This invention provides a technical solution. As shown in Figure 1, this invention provides a smart agriculture intelligent monitoring method based on big data, including the following analysis steps: Step S100: Set the width of the farmland grid unit, divide all farmland into grid units, monitor farmland image information through remote sensing technology, and measure the fill degree in each farmland grid unit; Step S200: Obtain historical remote sensing image data of each time period during crop growth in each farmland grid unit through remote sensing technology, record the time point of remote sensing image historical data acquisition, synchronously obtain the final harvest time data of crops in the historical remote sensing image data, calculate the crop maturity data of each farmland grid unit in the historical remote sensing image data, and construct a crop growth status prediction model; Step S300: Predict the expected maturity period of crops in each farmland grid unit, obtain meteorological parameter forecast information for future time periods, and analyze the harvest deadline of crops in each farmland grid unit; Step S400: Use an external navigation interface to monitor and summarize the location data of all agricultural machinery and the information of traversable roads within the management range in real time, and analyze the time required for each agricultural machinery to reach each farmland grid unit; Step S500: Calculate the harvesting efficiency of each agricultural machine allocated to each farmland grid unit at each time node, and use dynamic programming to analyze the strategy that maximizes the total harvesting efficiency of all farmland in the monitored area.

[0018] Step S100 includes the following: Setting the width 'a' of the farmland grid cells, where each farmland grid cell is a square with a side length of 'a', and tiling all farmland areas to ensure all farmland areas are covered by the farmland grid cells; Acquiring remote sensing image data of the farmland areas, and for irregular areas at the edges of the farmland areas, according to the formula: Calculate the fill degree of the grid cells covering the farmland in this area; where r is the fill degree of the grid cells covering the farmland in this area, and n f To cover the number of pixels contained in the farmland area in the remote sensing image of the farmland grid cell, where 'a' is the width of the farmland grid cell and 'n' is the number of pixels in the grid cell. u This refers to the number of pixels contained in a unit area of ​​remote sensing image. In practical implementation, the side length of farmland grid units can be set according to the accuracy of the remote sensing technology used and the crop planting habits to ensure that the number of remote sensing image pixels contained in a single farmland grid unit is within the range of 100-500. In some irregularly shaped farmland plots, farmland grid units are used to densely cover the plots, and then the farmland area fill degree in each unit is calculated.

[0019] Step S200 includes the following: Extracting historical remote sensing image data of crop growth in each farmland grid cell, obtaining the reflectance of each band's reflectance spectrum, calculating the crop maturity in each farmland grid cell from the historical remote sensing image data at different time points, and for any farmland grid cell x, according to the formula: Among them, Mx Let n be the crop maturity level of the farmland grid unit x. f，x Let be the number of pixels contained in the farmland region within the farmland grid cell x, i be the pixel number contained in the farmland region within the farmland grid cell x, G be the digital amplification gain, and NIR be the digital gain. i Let RED be the near-infrared reflectance of pixel i. i For pixel i, the red band reflectance is BLUE. i Let represent the blue light band reflectance of pixel i, C1 and C2 refer to the atmospheric scattering correction coefficients for the red and blue light bands, respectively, and L be the ground reflectance correction coefficient. In practice, the near-infrared band reflectance, red band reflectance, and blue band reflectance can be obtained from satellite remote sensing multispectral images or from measurements taken by a UAV carrying a multispectral sensor. In the above calculation formula, the gain factor G is generally taken as 2.5, the atmospheric correction coefficients C1 and C2 are used to eliminate atmospheric scattering interference on the red and blue light bands, and are generally taken as 6 and 7.5, respectively, and the ground reflectance correction coefficient L is used to reduce noise interference caused by ground reflection, and is generally taken as 1.

[0020] For different crop types, crop growth status curves were fitted using the interval between remote sensing data acquisition time points and final harvest time points, as well as crop maturity, to obtain crop growth status prediction models. The models are as follows: M pre,x =A×e k×t +B; where M pre,x The crop maturity is predicted for each farmland grid cell x, where A and B are the prediction model parameters, e is the natural logarithm, k is the current crop growth rate coefficient, and t is the time difference between the prediction time and the final harvest. In practice, for crops such as rice and wheat that gradually change from green to withered yellow as they grow, the chlorophyll content decreases in a uniform trend, and the farmland reflectance spectrum changes more significantly. Therefore, using remote sensing images for maturity assessment is more accurate. At the same time, an exponential model is used to predict maturity. In this model, the parameter t decreases over time, so the maturity value decreases over time. Generally, rice and wheat are considered mature and ready for harvest when the maturity value drops to 0.2-0.4. Since the actual harvest time is chosen to allow the grains to dry slightly in the field after maturity, the subsequent maturity threshold can be adjusted and lowered accordingly.

[0021] Step S300 includes the following: Setting crop maturity threshold ranges according to crop type; for any farmland grid cell, calculating the crop maturity at the current time point; using the crop growth status prediction model to calculate the estimated time difference between crop maturity and harvest, and then analyzing the estimated crop maturity at future time points; calculating the earliest time point in the current farmland grid cell where the estimated crop maturity falls within the maturity threshold range, and setting this time point as the initial maturity time point of the farmland grid cell; calculating the latest time point in the current farmland grid cell where the estimated crop maturity falls within the maturity threshold range, and setting this time point as the cutoff maturity time point of the farmland grid cell; setting the period from the initial maturity time point to the cutoff maturity time point as the cutoff harvest period T for the current farmland grid cell. x , denoted as T x [t x,1 ,t x,2 ]; where t x,1 The left threshold for the harvest deadline of farmland grid cell x represents the initial maturity time of farmland grid cell x, t. x,2 The left threshold for the harvest deadline of farmland grid unit x represents the cutoff maturity time of farmland grid unit x. Meteorological parameter forecasts for all monitored farmland areas within the harvest deadline are obtained, parameters affecting the harvest of crops in each farmland grid unit are screened, and interference harvest threshold ranges are set. For any farmland grid unit, if the value of any screened parameter at any time point within the harvest deadline falls within the interference harvest threshold range, the harvest deadline of the current farmland grid unit is corrected to the time point from the initial maturity time to the time point where the parameter value falls within the interference harvest threshold range. In specific implementation, taking rice and wheat as examples, a certain amount of rainfall is required during the grain-filling stage, while the water demand gradually decreases after actual maturity. Excessive water can lead to problems such as disease, mold, and premature germination of grains in the field. Therefore, by obtaining climate parameters during the harvest period, damage prevention is carried out to avoid losses in the final harvest due to abnormal changes in field water volume.

[0022] Step S500 includes the following: Obtaining the harvest deadline data for each monitored farmland grid unit, and classifying all farmland grid units according to the left threshold t of the harvest deadline. x，1Sort the data, then set harvest time windows. Select the sorted farmland grid units by the harvest time window, and set farmland grid units in the same harvest time window as the same harvest batch. Then, divide the geographically adjacent farmland grid units in the same harvest batch into the same harvest area. When performing harvesting strategy analysis, obtain the harvest time window of each harvest area. When the time point reaches any time window, perform agricultural machinery allocation strategy analysis on all harvest areas within that time window. For agricultural machinery that is currently harvesting, calculate the harvesting effect after the expected time point of completion of the harvesting operation is reached. The harvesting effect calculation and the strategy analysis method for maximizing the total harvesting effect of different agricultural machinery allocated to each harvesting area within each time node in step S500 are as follows: Calculate the harvesting effect of different agricultural machinery allocated to each harvesting area within each time node. For any agricultural machinery m and any harvesting area p, according to the formula: W tra =k1×t tra +k2×W F Among them, W m,p The harvesting efficiency of agricultural machinery m allocated to harvesting area p, W tra Let ε be the traffic loss of agricultural machinery m traveling to the nearest farmland grid cell in the harvesting area p, ε be a constant parameter to avoid a denominator of 0, e be the natural logarithm, α be the time scale scaling factor, and t be the distance between the agricultural machinery m and the harvesting area p. gap S is the time difference between the current time point and the earliest time point of the right threshold of the harvest deadline for the farmland grid cell in the harvesting area p. p Let n be the area of ​​harvesting region a, j be the farmland grid cell number in harvesting region p, and n be the area of ​​harvesting region a. j Let be the number of farmland grid cells in the harvesting area p, 'a' be the side length of the farmland grid cell, 'T' be the time taken by agricultural machinery m to harvest one farmland grid cell in days, and 'r' be the number of farmland grid cells in the harvesting area p. j Let k1 and k2 be the fill degree of farmland grid cell j, and k1 and k2 be the traffic loss assessment parameters for agricultural machinery m. tra W represents the travel time, in hours, for agricultural machinery m to reach the nearest farmland grid cell in the harvesting area p. F The fuel consumption of agricultural machinery m traveling to the nearest farmland grid cell in the harvesting area p is given. Based on the above calculation formula, the following calculation example is provided: Assuming the side length of the farmland grid cell is 10 meters, the harvesting time for one farmland grid cell is 0.5 hours, and the calculated harvesting area contains 5 farmland grid cells with fill percentages of 80%, 100%, 100%, 100%, and 60%, respectively, and the traffic loss assessment parameters k1 = 0.6 and k2 = 0.4, the time taken for the agricultural machinery to travel to the nearest farmland grid cell in the harvesting area is 1.5 hours, with a fuel consumption of 5 liters. gap=8, the unit is days, the time scale scaling factor α = -0.1, ε = 0.001, the calculated harvest area is 440 square meters, the traffic loss is 2.9, and the harvesting effect of the agricultural machinery allocated to this harvest area is 136.5; then the harvesting effects of all harvest areas are added together to obtain the total harvesting effect of farmland in the monitored area, and dynamic programming is used to analyze the strategy to maximize the total harvesting effect.

[0023] As shown in Figure 2, this invention also provides a smart agriculture intelligent monitoring system based on big data. The system includes: an agricultural data monitoring module, a crop growth prediction module, and an agricultural machinery harvesting allocation module. The agricultural data monitoring module uses remote sensing technology to monitor image information of all farmland and crops planted within the monitored area, divides the farmland into grids, and performs crop maturity analysis and calculation. The crop growth prediction module is used to construct a crop growth status prediction model and predict the expected crop maturity of farmland crops at future time points, and analyze the harvest deadline for each farmland grid unit. The agricultural machinery harvesting allocation module calculates the harvesting effectiveness of each agricultural machine allocated to each harvesting area and analyzes the agricultural machinery allocation strategy that maximizes the total harvesting effectiveness of all farmland within the monitored area.

[0024] The agricultural data monitoring module includes: a remote sensing monitoring unit, a farmland area segmentation unit, and a crop monitoring unit; the remote sensing monitoring unit is used to acquire remote sensing image data of farmland areas; the farmland area segmentation unit is used to segment farmland areas into grids; and the crop monitoring unit is used to monitor the growth status of crops within the monitored area and calculate crop maturity data.

[0025] The crop growth prediction module includes: a model building unit, a crop maturity prediction unit, and a harvest deadline analysis unit. The model building unit is used to build a crop growth status prediction model; the crop maturity prediction unit is used to calculate and analyze the expected maturity of the crop; and the harvest deadline analysis unit is used to correct the harvest deadline of the grown crop based on future weather parameters for each farmland grid unit.

[0026] The agricultural machinery harvesting allocation module includes: a harvesting area division unit, a harvesting effectiveness calculation unit, and an agricultural machinery allocation unit; the harvesting area division unit divides the farmland within the monitored area into different harvesting areas based on the expected maturity of crops in each farmland grid unit; the harvesting effectiveness calculation unit is used to calculate the harvesting effectiveness of different agricultural machinery allocated to each harvesting area; the agricultural machinery allocation unit is used to analyze the strategy for maximizing the total harvesting effectiveness of all farmland within the monitored area and to allocate agricultural machinery.

[0027] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered in all respects as exemplary and non-limiting, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the present invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

Claims

1. A smart agriculture intelligent monitoring method based on big data, characterized in that... The method includes the following analysis steps: Step S100: Set the width of the farmland grid cell, divide all farmland into grid cells, monitor farmland image information through remote sensing technology, and measure the fill degree in each farmland grid cell; Step S200: Use remote sensing technology to acquire historical remote sensing image data of each time period during the crop growth process of each farmland grid unit, record the acquisition time point of the historical remote sensing image data, synchronously acquire the final harvest time data of the crop in the historical remote sensing image data, calculate the crop maturity data of each farmland grid unit in the historical remote sensing image data, and construct a crop growth status prediction model. Step S300: Predict the expected maturity date of crops in each farmland grid unit, obtain meteorological parameter forecast information for future time periods, and analyze the cut-off harvest period of crops in each farmland grid unit; Step S400: Use the external navigation interface to monitor and summarize the location data of all agricultural machinery and the information on traversable roads within the management range in real time, and analyze the time required for each agricultural machine to reach each farmland grid unit; Step S500: Calculate the harvesting efficiency of each agricultural machine allocated to each farmland grid unit at each time node, and use dynamic programming to analyze the strategy that maximizes the total harvesting efficiency of all farmland in the monitored area.

2. The intelligent monitoring method for smart agriculture based on big data as described in claim 1, characterized in that... Step S100 includes the following: Set the width 'a' of the farmland grid cells, where the shape of the farmland grid cells is a square with a side length of 'a', and tile all farmland areas so that all farmland areas are covered by the farmland grid cells; To acquire remote sensing image data of farmland areas, for irregular areas at the edges of farmland areas, the formula is: Calculate the fill degree of the grid cells covering the farmland in this area; Where r is the fill degree of the grid cells covering the farmland in the area, and n f To cover the number of pixels contained in the farmland area in the remote sensing image of the farmland grid cell, where 'a' is the width of the farmland grid cell and 'n' is the number of pixels in the grid cell. u This represents the number of pixels contained in a unit area of ​​a remote sensing image.

3. The intelligent monitoring method for smart agriculture based on big data as described in claim 1, characterized in that... Step S200 includes the following: Historical remote sensing image data of crop growth process was extracted from each farmland grid cell. The reflectance of each band was obtained, and the crop maturity of each farmland grid cell in the historical remote sensing image data at different time points was calculated. For any farmland grid cell x, the following formula was used: Among them, M x Let n be the crop maturity level of the farmland grid unit x. f,x Let be the number of pixels contained in the farmland region within the farmland grid cell x, i be the pixel number contained in the farmland region within the farmland grid cell x, G be the digital amplification gain, and NIR be the digital gain. i Let RED be the near-infrared reflectance of pixel i. i For pixel i, the red band reflectance is BLUE. i denoted as the blue light band reflectance of pixel i, C1 and C2 refer to the atmospheric scattering correction coefficients for the red and blue light bands, respectively, and L is the ground reflectance correction coefficient. For different crop types, crop growth status curves were fitted using the interval between remote sensing data acquisition time points and final harvest time points, as well as crop maturity, to obtain crop growth status prediction models. The models are as follows: M pre，x =A×e k×t +B; Among them, M pre，x Let x be the predicted crop maturity of the farmland grid cell, A and B be the prediction model parameters, e be the natural logarithm, k be the growth rate coefficient of the current crop type, and t be the time difference between the prediction time and the final harvest of the crop.

4. The intelligent monitoring method for smart agriculture based on big data as described in claim 1, characterized in that... Step S300 includes the following: Crop maturity threshold ranges are set according to crop type. For any farmland grid cell, the crop maturity at the current time point is calculated. The crop growth status prediction model is used to calculate the expected time difference between the crop and maturity for harvest, and then the expected crop maturity at future time points is analyzed. Calculate the earliest time point in the current farmland grid cell where the expected crop maturity is within the maturity threshold range, and set this time point as the initial maturity time point of the farmland grid cell; calculate the latest time point in the current farmland grid cell where the expected crop maturity is within the maturity threshold range, and set this time point as the cutoff maturity time point of the farmland grid cell. Let the time from the initial maturity point to the end maturity point be defined as farmland grid unit x harvest deadline T. x , denoted as T x [t x,1 ,t x,2 ]; where t x,1 The left threshold for the harvest deadline of farmland grid cell x represents the initial maturity time of farmland grid cell x, t. x，2 The left threshold for the harvest deadline of farmland grid unit x represents the harvest time point of farmland grid unit x; obtain the meteorological parameter forecast information within the harvest deadline of all monitored farmland areas, filter the parameters that may affect the harvest of crops in each farmland grid unit, and set the interference harvest threshold range; For any farmland grid cell, if the value of the selected parameter at any point within the harvest deadline is within the range of the interference harvest threshold, the harvest deadline of the current farmland grid cell is corrected to the point from the initial maturity time to the point where the parameter value is within the range of the interference harvest threshold.

5. The intelligent monitoring method for smart agriculture based on big data as described in claim 4, characterized in that... Step S500 includes the following: Obtain the harvest deadline data for each monitored farmland grid unit, and then classify all farmland grid units according to the left threshold t of the harvest deadline. x，1 Sort the data, then set the harvest time window. Select the sorted farmland grid units by the harvest time window, set the farmland grid units in the same harvest time window to the same harvest batch, and then divide the geographically adjacent farmland grid units in the same harvest batch into the same harvest area. When performing harvesting strategy analysis, the harvesting time window for each harvesting area is obtained. When the time point reaches any time window, the agricultural machinery allocation strategy analysis is performed on all harvesting areas within that time window. For agricultural machinery that is currently harvesting, the harvesting effect is calculated only after the time point reaches the expected time point for completing the harvesting operation.

6. The intelligent monitoring method for smart agriculture based on big data as described in claim 5, characterized in that... The method for calculating the harvesting effectiveness of different agricultural machines allocated to each harvesting area at each time point in step S500 and the method for analyzing the strategy to maximize the total harvesting effectiveness are as follows: To calculate the harvesting effectiveness of different agricultural machines allocated to different harvesting areas at each time point, for any agricultural machine m and any harvesting area p, according to the formula: IN tra =k1×t tra +k2×W F ; Among them, W m，p The harvesting efficiency of agricultural machinery m allocated to harvesting area p, W tra Let ε be the traffic loss of agricultural machinery m traveling to the nearest farmland grid cell in the harvesting area p, ε be a constant parameter to avoid a denominator of 0, e be the natural logarithm, α be the time scale scaling factor, and t be the distance between the agricultural machinery m and the harvesting area p. gap S is the time difference between the current time point and the earliest time point of the right threshold of the harvest deadline for the farmland grid cell in the harvesting area p. p Let n be the area of ​​harvesting region a, j be the farmland grid cell number in harvesting region p, and n be the area of ​​harvesting region a. j Let be the number of farmland grid cells in the harvesting area p, 'a' be the side length of the farmland grid cell, 'T' be the time taken by agricultural machinery m to harvest one farmland grid cell in days, and 'r' be the number of farmland grid cells in the harvesting area p. j Let k1 and k2 be the fill degree of farmland grid cell j, and k1 and k2 be the traffic loss assessment parameters for agricultural machinery m. tra W represents the travel time, in hours, for agricultural machinery m to reach the nearest farmland grid cell in the harvesting area p. F Fuel consumption for agricultural machinery m as it travels to the nearest farmland grid unit in the harvesting area p; Then, the harvesting results of all harvested areas are summed to obtain the total harvesting results of farmland within the monitored area, and dynamic programming is used to analyze the strategy to maximize the total harvesting results.

7. A smart agriculture intelligent monitoring system based on big data, applying the smart agriculture intelligent monitoring method based on big data as described in any one of claims 1-6, characterized in that... The system includes: an agricultural data monitoring module, a crop growth prediction module, and an agricultural machinery harvesting and distribution module; The agricultural data monitoring module uses remote sensing technology to monitor image information of all farmland and crops within the monitored area, divides the farmland into grids, and performs crop maturity analysis and calculation. The crop growth prediction module is used to construct a crop growth status prediction model and predict the expected crop maturity of farmland crops at future time points, and analyzes the harvest deadline for each farmland grid unit. The agricultural machinery harvesting allocation module calculates the harvesting effectiveness of each agricultural machine allocated to each harvesting area and analyzes the agricultural machinery allocation strategy that maximizes the total harvesting effectiveness of all farmland within the monitored area.

8. The intelligent monitoring system for smart agriculture based on big data according to claim 7, characterized in that... The agricultural data monitoring module includes: a remote sensing monitoring unit, a farmland area segmentation unit, and a crop monitoring unit; The remote sensing monitoring unit is used to acquire remote sensing image data of farmland areas; the farmland area segmentation unit is used to segment farmland areas into grids; and the crop monitoring unit is used to monitor the crop growth status within the monitored area and calculate crop maturity data.

9. A smart agriculture intelligent monitoring system based on big data according to claim 7, characterized in that... The crop growth prediction module includes: a model building unit, a crop maturity prediction unit, and a harvest deadline analysis unit. The model building unit is used to build a crop growth status prediction model; the crop maturity prediction unit is used to calculate and analyze the expected maturity of the crop; and the harvest deadline analysis unit is used to correct the harvest deadline of the grown crop based on the future weather parameters of each farmland grid unit.

10. The intelligent monitoring system for smart agriculture based on big data according to claim 7, characterized in that... The agricultural machinery harvesting and allocation module includes: a harvesting area division unit, a harvesting effect calculation unit, and an agricultural machinery allocation unit; The harvesting area division unit divides farmland within the monitored area into different harvesting areas based on the expected maturity of crops in each farmland grid unit; the harvesting effectiveness calculation unit is used to calculate the harvesting effectiveness of different agricultural machines allocated to each harvesting area; the agricultural machine allocation unit is used to analyze the strategy for maximizing the total harvesting effectiveness of all farmland within the monitored area and to allocate agricultural machines.

Images