Terrain analysis method and system based on unmanned aerial vehicle remote sensing monitoring decision

Abstract

The invention belongs to the field of agricultural terrain decision, and particularly relates to a terrain analysis method and system based on unmanned aerial vehicle remote sensing monitoring decision, and the method comprises the steps: firstly obtaining various data of a to-be-mined region, constructing a three-dimensional terrain visualization and planting simulation model, configuring a crop planting attribute library, and simulating planting to obtain a score; secondly, screening crops according to a score threshold value, and performing secondary planning planting if a condition is met; the area needing to be transformed is transformed according to attributes, and difficulty is evaluated; and configuring a modification difficulty threshold according to the planting demand and cost of the user, and planting, modifying and planning the crops meeting the conditions according to the conditions. If the crops do not meet the requirements but the transformation difficulty is proper, selecting crops with low estimated cost for transformation and planting; and finally, when the crop planting difficulty is large and the growth vigor is poor, determining whether to transform planting according to conditions by analyzing attributes, thereby providing a precise decision for agricultural planting planning.

Description

Technical Field

[0001] The present invention belongs to the field of agricultural terrain decision-making, and in particular relates to a terrain analysis method and system based on unmanned aerial vehicle (UAV) remote sensing monitoring and decision-making. Background Art

[0002] In precision agriculture, terrain analysis is a key component, involving the understanding and application of farmland topographic features such as slope, aspect, and altitude. Proper terrain analysis can help farmers optimize irrigation system layout, select appropriate crop varieties, and plan agricultural machinery operation routes, thereby achieving optimal resource allocation and environmentally friendly agricultural practices. However, traditional terrain analysis methods often rely on ground measurements or satellite imagery, which suffer from slow data updates, low spatial resolution, and high costs. Using terrain feature analysis to conduct agricultural monitoring and analysis on different types of land, design reasonable agricultural development strategies, and plant the most suitable crops is a major challenge in current agricultural development.

[0003] For example, the Chinese patent with authorization announcement number CN118503658B discloses a method for extracting piedmont water systems based on optical remote sensing data and digital elevation models. The method first obtains and preprocesses the optical remote sensing and digital elevation model data of the piedmont area, then extracts the water body and terrain feature data sets respectively to obtain the corresponding extraction index. The extraction index is then comprehensively analyzed to obtain the accuracy index, which is compared with the threshold to optimize the preprocessing status of the piedmont water system data extraction.

[0004] The above existing technologies have the following problems: there are still many areas that can be used for agricultural planting but have not been developed, and the traditional manual measurement method is inefficient and has limited accuracy, and is greatly restricted by manpower and geographical environment. In terms of regional terrain modeling, the lack of intelligent means makes it difficult to accurately simulate and analyze complex terrain, and it is impossible to fully consider the comprehensive interaction between multiple factors such as soil type and hydrological conditions and terrain, making it difficult to accurately match the direction of agricultural development, which can easily lead to resource mismatch and low agricultural production efficiency. For this reason, the present invention provides a terrain analysis method and system based on drone remote sensing monitoring decision-making. Summary of the Invention

[0005] To address the shortcomings of existing technologies, the present invention proposes a terrain analysis method and system based on drone remote sensing monitoring and decision-making. The system includes the following steps: first, acquiring multiple data points for the area to be mined, constructing a three-dimensional terrain visualization and planting simulation model, configuring a crop planting attribute library, and simulating planting to obtain scores; second, screening crops based on score thresholds, and performing secondary planting planning if conditions are met; then, re-planning areas for transformation based on their attributes and assessing the difficulty; then, configuring a transformation difficulty threshold based on user planting requirements and costs, and planting, transforming, and planning crops that meet the requirements accordingly. If a crop does not meet the requirements but the transformation difficulty is appropriate, a crop with a low estimated cost is selected for transformation and planting; finally, if the crop is difficult to plant and its growth is poor, the attributes are analyzed and the conditions are used to determine whether to transform the crop, thereby providing accurate decision-making for agricultural planting planning.

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

[0007] Terrain analysis methods based on UAV remote sensing monitoring and decision-making include:

[0008] S1. Obtain multi-angle terrain images, laser measurement data, meteorological data, hydrological data, and user planting requirements and estimated cost data for the current area. Build a regional 3D terrain planting simulation model with hydrological and meteorological data by combining simulation algorithms with 3D generation algorithms.

[0009] S2. Configure a crop planting attribute library. Based on the planting requirement parameters of each crop in the library, perform planting simulation and evaluation using a regional three-dimensional terrain planting simulation model with hydrological and meteorological data and a crop growth factor function for each growth period. This results in a planting difficulty score and simulated growth score for each crop in the current region.

[0010] S3. Set the planting difficulty score threshold and simulated growth score threshold , when there is a crop planting difficulty score less than And the simulated growth score is greater than the simulated growth score threshold When the number of types is greater than or equal to 1, the crops that meet the conditions are selected as the crops planted in the current area, and according to the user's planting needs, the current area is secondary planted according to the number of crop types that meet the conditions and the crop planting attributes;

[0011] S4. When the planting difficulty scores of all crops are greater than or equal to the planting difficulty score threshold , but there are crops with simulated growth scores greater than the simulated growth score threshold If the species is greater than the simulated growth score threshold, The crop planting attributes of the conditions are evaluated and judged for transformation of the current area, the current area after transformation is obtained, and secondary planting planning is carried out.

[0012] Specifically, the steps for evaluating and judging the transformation of the current area include:

[0013] S5. Set a transformation difficulty threshold based on the user's planting needs and estimated cost data, and evaluate the difficulty of environmental transformation for the crops that meet the conditions in S4 when planted in the current area, to obtain the environmental transformation difficulty for the crops.

[0014] S6. When there is at least one crop that meets the user's planting requirements and the environmental transformation difficulty is less than the transformation difficulty threshold, the crop that meets the conditions is selected for planting, and the current area is secondary transformed based on the number of crop types that meet the conditions, the corresponding crop planting attributes, and the current area's environmental attributes. At the same time, the S3 process is repeated to perform secondary planting planning on the transformed current area.

[0015] S7. When all crops do not meet the user's planting requirements, but there are crops that meet the corresponding environmental modification difficulty and are less than the modification difficulty threshold, the crop with the lowest estimated cost is selected as the corresponding crop for the current area, and the S6 process is repeated to perform environmental modification and planting planning for the current area based on the planting requirement parameters corresponding to the crops that meet the user's planting requirements.

[0016] S8. When the planting difficulty scores of all crops are greater than or equal to the planting difficulty score threshold , and the simulated growth score is less than or equal to the simulated growth score threshold When analyzing all crop attributes, if there is a crop with the lowest planting difficulty score and the highest simulated growth score, the corresponding crop type is the user's planting requirement and the transformation cost is less than or equal to the estimated cost, the current area will be transformed and planted. Otherwise, no planting plan will be made for the current area.

[0017] Specifically, the steps of constructing a regional three-dimensional terrain planting simulation model with hydrological and meteorological data include:

[0018] S101, generating a three-dimensional terrain sub-model of the current area using a three-dimensional generation algorithm based on the acquired multi-angle terrain image of the current area and the slope and terrain height data in the laser measurement data;

[0019] S102: Pre-train the configured weather forecast model based on the acquired current regional weather data to obtain a regional weather simulation sub-model, and output the simulated weather status for each time period in the current region;

[0020] S103. Based on the current regional hydrological data, the configured hydrological model is trained to obtain a regional hydrological simulation sub-model, and the distance between the water source and the current region, the water flow path, and the soil moisture are simulated.

[0021] S104, integrating the current regional three-dimensional terrain sub-model, the regional meteorological simulation sub-model, and the regional hydrological simulation sub-model through an integration algorithm to obtain a regional three-dimensional terrain model with hydrological and meteorological features;

[0022] S105: constructing a segmented growth period growth factor function based on the planting environment conditions and simulated meteorological conditions of each crop growth period, the distance between the water source and the current area, the water flow path, and the soil moisture. The segmented growth period growth factor function is embedded in the constructed crop segmented growth simulation sub-model for training to obtain the corresponding growth state of each crop under different growth conditions.

[0023] S106. Integrate the pre-trained crop segmented growth simulation sub-model into the regional three-dimensional terrain model with hydrometeorology to obtain a regional three-dimensional terrain planting simulation model with hydrometeorology. When the current region changes, repeat the S101-S106 process to obtain a fine-tuned regional three-dimensional terrain planting simulation model with hydrometeorology for each region, and dynamically annotate the simulated meteorological, hydrological and crop growth status according to the time dimension.

[0024] Specifically, the steps for constructing the growth factor function for the segmented growth period are:

[0025] S201. Set the crop growth stages to include the sowing stage, seedling stage, vegetative growth stage, reproductive growth stage, and maturity stage. Based on the relationship between each crop's parameters at different growth stages and meteorological and hydrological parameters, use a factor analysis algorithm to analyze and obtain meteorological and hydrological parameter spaces whose contribution is greater than a contribution threshold.

[0026] S202. Based on the meteorological and hydrological parameter spaces whose contribution is greater than the contribution threshold, a segmented growth period growth factor function corresponding to each crop is constructed by combining a multivariate regression algorithm with a genetic optimization algorithm.

[0027] Specifically, the steps for obtaining the crop planting difficulty score include:

[0028] S211. Obtain a comprehensive terrain transformation difficulty score based on the current area slope, terrain height difference, and terrain complexity obtained through the evaluation algorithm;

[0029] S212. Obtain a meteorological adaptability score for the current region using an evaluation algorithm based on the current average temperature, precipitation, and wind speed of the region.

[0030] S213, obtaining a hydrological adaptability score for the current region through an evaluation algorithm based on the distance between the water source and the current region, the water flow path, and soil moisture information;

[0031] S214. Obtain the crop planting difficulty score for the current region by weighted average method based on the comprehensive terrain transformation difficulty score, meteorological adaptability score, and hydrological adaptability score for the current region.

[0032] Specifically, the steps of performing secondary planting planning for the current area in S3 include:

[0033] S301. When only one type of crop meets the conditions, directly plant the corresponding type of crop in the current area;

[0034] S302. When more than one crop meets the requirements, the shade tolerance, drought tolerance, and lodging resistance of each crop type are obtained based on the number of crop types that meet the requirements, as well as the terrain height, slope, soil moisture, distance to water source, water flow path direction, historical wind speed and direction, and planned planting ratio of each crop type that meets the user's planting requirements. The current area is divided using a particle swarm optimization grid algorithm to obtain crop planting sub-areas of the corresponding crop type.

[0035] S303: Plant the corresponding crop in the crop planting sub-area, and obtain the crop growth score and corresponding yield during the corresponding planting period;

[0036] S304: Feedback the growth score and yield of the corresponding type of crops to the particle swarm optimization grid algorithm to make real-time adjustments to the crop planting area.

[0037] The terrain analysis system based on UAV remote sensing monitoring decision-making includes: simulation module, initial decision-making judgment module and secondary decision-making judgment module;

[0038] The simulation module includes a data acquisition unit, a model building unit and a simulation evaluation unit;

[0039] Data acquisition unit, used to obtain multi-angle terrain images of the current area, laser measurement data, meteorological data and hydrological data, as well as user planting requirements and estimated cost data;

[0040] The model building unit is used to build a regional three-dimensional terrain planting simulation model with hydrological and meteorological data through a simulation algorithm and a three-dimensional generation algorithm according to the data obtained by the data acquisition unit;

[0041] The simulation evaluation unit is used to simulate and evaluate the planting of the corresponding crops in the crop planting attribute library through a regional three-dimensional terrain planting simulation model with hydrological and meteorological conditions and a crop segmented growth period growth factor function, so as to obtain the planting difficulty score and simulated growth score of each crop in the current area.

[0042] Specifically, the initial decision discrimination module includes a first initial decision unit and a second initial decision unit;

[0043] The first initial decision unit is used to score the planting difficulty according to the set threshold. and simulated growth score threshold ,select the crops that meet the first initial decision conditions as the crops to be planted in the current area, and according to the user's planting needs, perform secondary planting planning for the current area based on the number of crop types that meet the conditions and the crop planting attributes;

[0044] The first initial decision condition is: there is a crop planting difficulty score less than And the simulated growth score is greater than types, and the number of types is greater than or equal to 1;

[0045] The second initial decision unit is used to select crops that meet the second initial decision conditions as crops to be planted in the current area, and to evaluate and determine the current area based on the planting attributes of the crops that meet the second initial decision conditions, obtain the current area after the transformation, and perform secondary planting planning;

[0046] The second initial decision condition is: the difficulty scores of all crops are greater than or equal to , but there are crops with simulated growth scores greater than kind of.

[0047] Specifically, the secondary decision-making and discrimination module includes a transformation evaluation unit, a first transformation decision unit, a second transformation decision unit, and a third transformation decision unit;

[0048] A transformation evaluation unit is used to set a transformation difficulty threshold based on the user's planting needs and estimated cost data, and evaluate the difficulty of environmental transformation corresponding to the planting of the corresponding crop in the current area that meets the second initial decision condition, thereby obtaining the environmental transformation difficulty of the corresponding crop;

[0049] The first transformation decision unit is used to select crops that meet the conditions for planting according to the first transformation decision conditions, and perform a secondary transformation on the current area based on the number of crop types that meet the conditions, the corresponding crop planting attributes, and the current area environmental attributes. At the same time, the S3 process is repeated to perform a secondary planting plan on the transformed current area;

[0050] The first transformation decision condition is: there is at least one crop that meets the user's planting needs and the environmental transformation difficulty is less than the transformation difficulty threshold;

[0051] The second transformation decision unit is used to select the crop with the lowest estimated cost as the corresponding planting crop in the current area according to the second transformation decision condition, and repeat the S6 process to perform environmental transformation and planting planning in the current area;

[0052] The second transformation decision condition is: all crops do not meet the user's planting requirements, but there are crops that meet the environmental transformation difficulty threshold;

[0053] The third transformation decision unit is used to select, based on the third transformation decision condition, a crop with the smallest planting difficulty score, the largest simulated growth score, a corresponding crop type that meets the user's planting requirements, and a transformation cost that is less than or equal to the estimated cost as the crop to be planted in the current area, and to perform transformation and planting planning for the current area based on the selected crop;

[0054] The third transformation decision condition is: the planting difficulty scores of all crops are greater than or equal to the planting difficulty score threshold, and the simulated growth scores are less than or equal to the simulated growth score threshold.

[0055] Compared with the prior art, the present invention has the following beneficial effects:

[0056] In response to the shortcomings of the existing technology, the present invention constructs a high-precision regional three-dimensional terrain planting simulation model with hydrological and meteorological data through multi-angle terrain images of undeveloped areas, laser measurement data, meteorological data and hydrological data, combined with simulation and three-dimensional generation algorithms, overcoming the problems of low efficiency and limited accuracy of traditional manual measurement methods. Secondly, through the configured crop planting attribute library, the regional three-dimensional terrain planting simulation model with hydrological and meteorological data combined with the crop segmented growth period growth factor function to perform planting simulation, accurately evaluate the planting difficulty and simulated growth score corresponding to each crop in each growth stage in the undeveloped area, and provide accurate analysis results for scientific seed selection and agricultural development in the current area; dynamically set thresholds to screen suitable crops to ensure the selection of the most suitable crops, while considering the user's planting needs and budget constraints. For areas that do not meet direct planting conditions, environmental transformation optimization plans are proposed so that areas that were originally unsuitable for planting can also be efficiently utilized within the budget. This process not only avoids resource mismatch but also further improves agricultural production efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] Figure 1 This is a flow chart of a terrain analysis method based on UAV remote sensing monitoring and decision-making according to Example 1 of the present invention;

[0058] Figure 2 This is a flow chart of regional transformation determination in Example 1 of the present invention;

[0059] Figure 3 This is a module diagram of a terrain analysis system based on UAV remote sensing monitoring and decision-making according to Example 2 of the present invention. DETAILED DESCRIPTION

[0060] Example 1:

[0061] See also Figure 1The present invention provides an embodiment of a terrain analysis method based on UAV remote sensing monitoring decision-making, which is used to evaluate crop planting planning in areas that have not undergone agricultural development, comprising the following steps:

[0062] S1. Obtain multi-angle terrain images, laser measurement data, meteorological data, hydrological data, and user planting requirements and estimated cost data for the current area. Build a regional 3D terrain planting simulation model with hydrological and meteorological data by combining simulation algorithms with 3D generation algorithms.

[0063] Furthermore, this embodiment utilizes a DJI M3M (Mavic 3 Multispectral Edition) to capture multi-angle terrain images of the current area. The main flight angle can be adjusted to align with the long edge, while other parameters remain at default. When planning the capture mission boundaries, consider surrounding high-altitude obstacles, such as high-voltage power lines and windmills, to determine whether they will affect the flight altitude. When planning plots, it is important to note that since we are viewing an elevation map, any other factors beyond the field, such as high-voltage power lines, operating agricultural machinery, and ridges along the edges of the fields, should be automatically removed using a noise segmentation algorithm. Failure to remove these elements will affect local elevation determination. After capture, image and point cloud stitching is performed to generate a high-precision digital terrain model of the farmland, allowing for elevation analysis of the target plot. Furthermore, in this embodiment, meteorological and hydrological data, as well as user planting requirements and estimated costs, are used simultaneously to further refine the accuracy of the land topography analysis.

[0064] Furthermore, in this embodiment, the steps of constructing a regional three-dimensional terrain planting simulation model with hydrological and meteorological features include:

[0065] S101, generating a three-dimensional terrain sub-model of the current area using a three-dimensional generation algorithm based on the acquired multi-angle terrain image of the current area and the slope and terrain height data measured in the laser data; the three-dimensional generation algorithm is preferably implemented by combining a Delaunay triangulation algorithm with a multi-view stereo (MVS) reconstruction algorithm;

[0066] S102. Pre-training a configured meteorological forecast model based on the acquired current regional meteorological data to obtain a regional meteorological simulation sub-model, and outputting simulated meteorological conditions for each time period in the current region; the meteorological forecast model is preferably constructed using a statistical downscaling model or a mesoscale meteorological model;

[0067] S103. Based on the current regional hydrological data, the configured hydrological model is trained to obtain a regional hydrological simulation sub-model, and the distance between the water source and the current region, the water flow path, and the soil moisture are simulated. The hydrological model is preferably a distributed hydrological model SWAT.

[0068] S104, integrating the current regional three-dimensional terrain sub-model, the regional meteorological simulation sub-model, and the regional hydrological simulation sub-model through an integration algorithm to obtain a regional three-dimensional terrain model with hydrological and meteorological features;

[0069] S105: constructing a segmented growth period growth factor function based on the planting environment conditions and simulated meteorological conditions of each crop growth period, the distance between the water source and the current area, the water flow path, and the soil moisture. The segmented growth period growth factor function is embedded in the constructed crop segmented growth simulation sub-model for training to obtain the corresponding growth state of each crop under different growth conditions.

[0070] S106. Integrate the pre-trained crop segmented growth simulation sub-model into the regional three-dimensional terrain model with hydrometeorology to obtain a regional three-dimensional terrain planting simulation model with hydrometeorology. When the current region changes, repeat the S101-S106 process to obtain a fine-tuned regional three-dimensional terrain planting simulation model with hydrometeorology for each region, and dynamically annotate the simulated meteorological, hydrological and crop growth status according to the time dimension.

[0071] This process generates a high-precision three-dimensional terrain sub-model by acquiring multi-angle terrain images and laser measurement data, overcoming the problems of low efficiency and limited accuracy of traditional manual measurement. Secondly, the meteorological forecast model and hydrological model are pre-trained to obtain a regional meteorological simulation sub-model and a regional hydrological simulation sub-model, respectively, which can accurately predict meteorological changes and water flow paths in the region, providing a scientific basis for irrigation system design. On this basis, the three sub-models are fused through an integrated algorithm to form a comprehensive three-dimensional terrain model, which comprehensively considers the interactive effects of multiple factors such as soil type and hydrological conditions on the terrain. Furthermore, in step S105, a segmented growth period growth factor function is introduced and trained in combination with the specific environmental conditions of each crop growth period to ensure accurate simulation of different growth stages. Finally, the crop segmented growth simulation sub-model is integrated into the overall model to realize dynamic labeling of crop growth status. When the current area changes, the above process can be repeated for fine-tuning and optimization.

[0072] S2. Configure a crop planting attribute library. Based on the planting requirement parameters of each crop in the library, perform planting simulation and evaluation using a regional three-dimensional terrain planting simulation model with hydrological and meteorological data and a crop growth factor function for each growth period. This results in a planting difficulty score and simulated growth score for each crop in the current region.

[0073] Furthermore, in this embodiment, the steps for constructing the segmented growth factor function are as follows:

[0074] S201. Set the crop growth stages to include the sowing stage, seedling stage, vegetative growth stage, reproductive growth stage, and maturity stage. Based on the relationship between each crop's parameters at different growth stages and meteorological and hydrological parameters, use a factor analysis algorithm to analyze and obtain meteorological and hydrological parameter spaces whose contribution is greater than a contribution threshold.

[0075] S202. Based on the meteorological and hydrological parameter spaces whose contribution is greater than the contribution threshold, a segmented growth period growth factor function corresponding to each crop is constructed by combining a multivariate regression algorithm with a genetic optimization algorithm.

[0076] Furthermore, in this embodiment, the segmented growth period growth factor function corresponding to each crop is constructed by combining a multivariate regression algorithm with a genetic optimization algorithm and the meteorological and hydrological parameter space with a contribution greater than a contribution threshold in each growth period.

[0077] It should be further explained that the specific implementation process of the multivariate regression algorithm combined with the genetic optimization algorithm in this embodiment includes:

[0078] After selecting meteorological and hydrological parameters with contributions greater than a contribution threshold for each growth period using a factor analysis algorithm, the filtered parameters were first subjected to dimensionality reduction processing using a principal component analysis algorithm to eliminate multicollinearity between variables and obtain uncorrelated principal component variables. An initial multivariate linear regression model was then constructed using the ordinary least squares algorithm, with growth factors as dependent variables and principal component variables as independent variables, to preliminarily establish a quantitative relationship between the parameters and growth factors. A genetic optimization algorithm was then introduced, using the model's coefficient of determination (R²) as the fitness function. The coefficients of the regression model were iteratively optimized through genetic operations such as selection, crossover, and mutation. During the iterative process, parameters with nonlinear relationships, such as wind speed and distance to water sources, were converted into linear forms using logarithmic transformation or polynomial expansion algorithms and incorporated into the regression model. Multiple rounds of genetic iterations were performed until the fitness function converged, and the regression coefficients with the optimal fitting effect were obtained. Finally, a segmented growth factor function was constructed that incorporates the key meteorological and hydrological parameters for each growth period and conforms to the biological mechanisms of crop growth.

[0079] Furthermore, in this embodiment, the specific reasons for selecting the parameters corresponding to the segmented growth stage growth factor function for each crop include:

[0080] During the sowing period, the parameters of the growth factor function corresponding to the segmented growth period include variables such as temperature, water and soil moisture, because seed germination requires suitable temperature, water and soil aeration; at this time, temperature and soil moisture are key factors, and too high or too low temperature may lead to a decrease in seed germination rate, and too dry or too wet soil will affect seed water absorption and respiration; in addition, during the sowing period, drought-tolerant crop varieties still have a certain germination ability when the soil moisture is low, which is reflected by the interaction term between crop drought tolerance and soil moisture; when the soil moisture is low, the drought-tolerant varieties still have a certain germination ability when the soil moisture is low. The value is large, and the interaction term between crop drought tolerance and soil moisture contributes relatively more to the growth factor, that is, drought-tolerant crop varieties can alleviate the inhibitory effect of low soil moisture on germination to a certain extent, on the contrary, drought-intolerant crop varieties The smaller the value, the lower the ability of the corresponding crop variety to alleviate the soil moisture lower than the corresponding required growth moisture, which makes the growth state of the corresponding crop variety subject to a greater inhibitory effect; represents the drought tolerance score of the i-th crop. This embodiment uses a comprehensive evaluation algorithm combined with the drought tolerance attribute parameters of the corresponding crop to perform a comprehensive evaluation;

[0081] During the seedling stage, the parameters of the growth factor function corresponding to the segmented growth period include: photoperiod, soil moisture, wind speed, shade-tolerant variety attributes, and the interaction term of light and soil moisture. For photoperiod-sensitive crops, the photoperiod function of each crop adjusts the effect of light on the growth factor; long-day crops promote growth when the light duration exceeds the threshold in the early growth stage, because long daylight meets their growth and development needs, promoting photosynthesis and seedling morphology; short-day crops promote growth when the light duration is lower than the threshold; soil moisture continues to affect root growth and nutrient absorption; during the seedling stage, the growth factor function corresponding to the segmented growth period has an inverse influence relationship with wind speed, because strong winds are not good for seedlings. The greater the wind speed, the more obvious the inhibitory effect on seedling growth. For the interaction term of light and soil moisture, when light is sufficient, appropriate soil moisture promotes photosynthesis, while insufficient soil moisture limits light utilization efficiency, and vice versa. The interaction term reflects this mutual constraint relationship. Shade-tolerant varieties can still maintain a certain growth under weak light conditions. When light is insufficient, the basic growth maintenance parameter value of shade-tolerant varieties is larger. At the same time, the interaction term of light and soil moisture contributes relatively more to the growth factor. This means that shade-tolerant varieties can maintain their growth momentum under low light conditions by enhancing the compensatory effect of soil moisture. However, the interaction term contributes less to shade-intolerant varieties, and their growth is more dependent on sufficient light conditions.

[0082] From a biological perspective, shade-tolerant varieties increase their chlorophyll b content in low-light environments, improving their weak light absorption efficiency and allocating more photosynthate to their roots. This enhances the interaction effect by increasing their water and nutrient absorption capacity. In mathematical models, shade-tolerant varieties exhibit larger interaction coefficients and basal growth maintenance parameters. This characteristic has important applications in agricultural production, such as optimizing intercropping patterns, reducing energy consumption for supplemental lighting in protected agriculture, and providing a reference indicator for the selection and breeding of shade-tolerant varieties.

[0083] During the vegetative growth period, the parameters of the growth factor function corresponding to the segmented growth period include: When other variables remain unchanged, the distance to the water source is increased. During the growth stage, a large amount of water is required. If the distance to the water source is far, water transportation is difficult, which limits growth. The limiting effect of the distance to the water source is reflected;

[0084] During the reproductive growth period, the segmented growth period growth factor function adds a new water flow path complexity parameter, which is quantified by combining indicators such as the number of tributaries, curvature, and water flow bifurcation coefficient with a fuzzy evaluation algorithm. The water flow path complexity parameter affects the uniformity of water and nutrient distribution, which in turn affects the development of reproductive organs; too many tributaries will cause the water flow at the end to attenuate, too large a curvature will easily cause water shortage on the convex bank and water accumulation on the concave bank, and a high water flow bifurcation coefficient will lead to differences in nutrient concentration gradients, all of which will cause uneven development of reproductive organs, such as incomplete grains and low fruit set rates; therefore, it is necessary to optimize the water flow path and control the number of tributaries, curvature, and water flow bifurcation coefficient within a reasonable threshold to achieve uniform distribution of water and nutrients in the field, reduce the risk of falling flowers and fruits, and ensure the normal development of reproductive organs;

[0085] During the maturity period, the parameters of the segmented growth factor function include radiation intensity, precipitation, soil moisture, wind speed and distance to water source. Each parameter affects the maturity process through a specific physiological mechanism. Radiation intensity and temperature synergistically regulate the material transformation and accumulation during the crop maturity process. The principle is that radiation is the energy source for photosynthesis, and its intensity directly affects the amount of photosynthetic products produced, while temperature controls the synthesis and conversion rate of storage substances such as sugar and starch by regulating enzyme activity; precipitation has a dual impact on crop quality in a linear relationship. Appropriate precipitation can maintain plant cell turgor and promote the transport of metabolic substances, while excessive precipitation will cause pathogens to grow due to high field humidity. The principle is that a high-humidity environment provides a suitable environment for fungi, bacteria and other pests and diseases. Suitable breeding conditions; soil moisture acts on the physiological activities of crops in the later stage in a linear form. The principle is that suitable soil moisture can ensure the water absorption function of the root system, maintain the balance between leaf transpiration and nutrient transport, and thus affect the ripening process such as grain filling or fruit swelling; wind speed and distance to water source reflect the inhibitory effect on ripening in an inverse form. Among them, strong winds in the ripening period can easily cause crops to fall over. The principle is that the mechanical tissue of the plant stem ages and the wind resistance decreases in the mature stage. When the mechanical stress generated by strong wind exceeds the threshold that the stem can withstand, it causes lodging. The distance to the water source will affect the water supply. The principle is that the resistance to the transport of soil water to the root system increases with the distance, resulting in a decrease in the amount of water available to the plant, affecting the water metabolism balance in the late stage of maturity. These parameters jointly affect the yield and quality formation of crops in the mature stage by regulating physiological processes such as material accumulation efficiency and environmental stress resistance.

[0086] Among the above variables, none of them has a single impact on crops, so corresponding interaction factors are added here to reflect the relationship between the variables.

[0087] Furthermore, in this embodiment, the steps for obtaining the crop planting difficulty score include:

[0088] S211. Obtain a comprehensive terrain transformation difficulty score based on the current area slope, terrain height difference, and terrain complexity obtained through the evaluation algorithm;

[0089] It should be further explained that, in this embodiment, one implementation method of the comprehensive terrain transformation difficulty score is as follows:

[0090] Based on regional DEM data, the slope value of each sampling point is extracted through the terrain analysis algorithm, and then the minimum-maximum normalization algorithm is used to map the slope to the range of 0-1 to obtain the slope normalization feature; then the altitude difference between the highest and lowest points in the region is calculated, and the height difference is converted into the corresponding transformation difficulty coefficient based on the threshold piecewise function algorithm; secondly, terrain characteristic parameters such as gully density and ridge undulation are extracted from the current regional digital elevation model data, and the fractal dimension is calculated using the box dimension algorithm to quantify the terrain complexity; finally, the slope normalization value, altitude difference difficulty coefficient, and terrain complexity fractal dimension are linearly weighted and summed according to preset weights, and the terrain transformation difficulty score is obtained through a comprehensive evaluation model.

[0091] Furthermore, in this embodiment, the difficulty of terrain transformation is graded according to the size of the slope S; When the terrain transformation difficulty coefficient , indicating that the terrain is relatively flat and the transformation is easy; when hour, ;when hour, , indicating that the slope is large and the transformation is difficult, such as the need to build terraces and other large-scale projects; when hour, , indicating that the terrain is steep and transformation is extremely difficult, which may require special engineering measures and high cost investment;

[0092] Furthermore, in this embodiment, the height difference between the highest point and the lowest point in the area is calculated from the laser measurement data and recorded as ;

[0093] when Meters, the difficulty coefficient of the terrain height difference ;when Meter hour, ;when Meter hour, ;when Meter hour, The greater the height difference, the more complicated the renovation work, such as leveling the land and building irrigation and drainage facilities, and the greater the difficulty coefficient.

[0094] Furthermore, in this embodiment, the terrain complexity is calculated by using methods such as fractal dimension on the multi-angle terrain image, which is recorded as ;

[0095] when When the terrain complexity coefficient ;when hour, ;when hour, ;when hour, The more complex the terrain, such as the presence of a large number of gullies and undulations, the more difficult it will be in land planning and mechanical operations.

[0096] S212. Obtain a meteorological adaptability score for the current region using an evaluation algorithm based on the current average temperature, precipitation, and wind speed of the region.

[0097] It should be further explained that, in this embodiment, one implementation method of the current regional weather adaptability score is as follows:

[0098] Based on meteorological parameters such as the current regional average temperature, precipitation, and wind speed, the Gaussian membership function algorithm is first used to calculate the degree of deviation between the actual temperature value and the optimal temperature for crops to obtain the temperature suitability score; secondly, the piecewise linear transformation algorithm is used to evaluate precipitation. When the precipitation is lower than the lower limit or higher than the upper limit, the precipitation suitability score is obtained through a linear decreasing function; then the wind speed is converted into an adaptability index through the inverse transformation algorithm. The greater the wind speed, the lower the score, and the wind speed suitability score is obtained; then the minimum-maximum normalization algorithm is used to normalize the scores of each factor to eliminate the dimensional effect; finally, the weight of each meteorological factor is calculated based on the entropy weight method, and the standardized factor score is multiplied and accumulated with the corresponding weight through a linear weighted aggregation model to obtain the meteorological adaptability score.

[0099] In this embodiment, the current regional historical meteorological data output by the regional meteorological simulation sub-model, such as multi-year average temperature, precipitation, wind speed, etc., as well as future predicted meteorological data, are compared with the suitable range of meteorological conditions for crops in the crop planting attribute library. The absolute value of the temperature difference, that is, the absolute value of the difference between the suitable temperature range of the crop and the actual average temperature, the absolute value of the precipitation difference, that is, the absolute value of the difference between the suitable precipitation range of the crop and the actual average precipitation, and the wind speed influence coefficient are calculated. The coefficient is determined based on the crop's tolerance to wind damage. For example, for crops sensitive to wind damage, when the wind speed exceeds a certain threshold, the wind speed influence coefficient takes a larger value.

[0100] S213, obtaining a hydrological adaptability score for the current region through an evaluation algorithm based on the distance between the water source and the current region, the water flow path, and soil moisture information;

[0101] It should be further explained that, in this embodiment, one implementation method of the current regional hydrological adaptability score is as follows:

[0102] Based on the actual distance data between the water source and the region, the distance is converted into a water supply convenience index using the inverse distance decay algorithm; based on characteristic parameters such as the number of tributaries and river channel curvature of the water flow path, the path complexity index is calculated using the network complexity analysis algorithm in graph theory; based on the measured soil moisture value and the field water holding capacity threshold, the linear membership function algorithm is used to calculate the humidity consistency score between the actual humidity and the suitable range; the water supply convenience index, path complexity index, and humidity consistency score are dimensionlessly processed using the minimum-maximum normalization algorithm, and then the objective weight of each indicator is determined based on the entropy weight method. Finally, the hydrological adaptability score is obtained using the linear weighted aggregation algorithm.

[0103] In this embodiment, the regional hydrological simulation submodel is used to obtain information about the distance between the water source and the current region, the water flow path, and soil moisture. The water source distance difficulty coefficient is calculated. For example, the greater the distance, the greater the water source distance difficulty coefficient. Different coefficient values can be set based on the distance range. The water flow path complexity coefficient is determined based on the tortuosity of the water flow path and the presence of obstacles. The more complex the water flow path, the greater the water flow path complexity coefficient. The absolute value of the soil moisture difference, that is, the absolute value of the difference between the crop-suitable soil moisture range and the actual soil moisture, is used to calculate this data and use it to obtain the hydrological adaptability score for the current region using an evaluation algorithm.

[0104] S214. Obtain the crop planting difficulty score for the current region by weighted average method based on the comprehensive terrain transformation difficulty score, meteorological adaptability score, and hydrological adaptability score for the current region.

[0105] It should be further explained that, in this embodiment, one implementation method of the current region crop planting difficulty score is as follows:

[0106] Based on the three evaluation scores of terrain transformation difficulty, meteorological adaptability, and hydrological adaptability, a judgment matrix was first constructed through the hierarchical analysis method, and the weight system of each indicator was obtained after eigenvalue calculation and consistency test; then, considering the positive correlation between the terrain transformation difficulty score and planting difficulty, and the negative correlation between the meteorological adaptability score and the hydrological adaptability score and the planting difficulty, a weighted aggregation algorithm was used to linearly combine the three scores, among which the meteorological adaptability score and the hydrological adaptability score needed to be adjusted to indicators positively correlated with planting difficulty through the polarity conversion algorithm; finally, the comprehensive score was mapped to the range of 0-100 points through the linear transformation algorithm to form a crop planting difficulty score that quantitatively represents the severity of planting conditions.

[0107] Furthermore, in this embodiment, the weighted coefficients of the corresponding variables in the weighted average method are obtained through analysis of a large number of actual cases and expert experience.

[0108] This process ensures the accuracy and applicability of the model by establishing a parameter space between crops at different growth stages and meteorological and hydrological parameters, and constructing segmented growth factor functions for each crop. These functions comprehensively consider key factors such as temperature, humidity, precipitation, light intensity, wind speed, distance to water sources and water flow paths, as well as the crop's drought tolerance, photoperiod function, and shade tolerance parameters, to accurately assess crop growth conditions at different growth stages. Furthermore, based on the configuration of a crop planting attribute library, planting simulation and evaluation are carried out using a regional three-dimensional terrain planting simulation model with hydrological and meteorological data and crop segmented growth factor functions, obtaining a planting difficulty score and simulated growth score for the crops in the current region. This process not only comprehensively considers the difficulty of terrain transformation, meteorological adaptability and hydrological adaptability, but also conducts a comprehensive assessment through the weighted average method to ensure the comprehensiveness and accuracy of the assessment results; specifically, the terrain transformation difficulty score is graded according to slope, terrain height difference and terrain complexity to ensure the feasibility of the terrain transformation plan; the meteorological adaptability score calculates the absolute value of temperature difference, the absolute value of precipitation difference and the wind speed influence coefficient by comparing historical meteorological data with the suitable range of crops, thereby ensuring the adaptability of crops to meteorological conditions; the hydrological adaptability score calculates the water source distance difficulty coefficient, the water flow path complexity coefficient and the absolute value of soil moisture difference through water source distance, water flow path and soil moisture information to ensure the adaptability of crops to moisture conditions; finally, the weighted average method is used to comprehensively evaluate each score to obtain the current regional crop planting difficulty score and simulated growth score, providing farmers with intelligent and personalized decision-making support and promoting the refined development of modern agricultural management.

[0109] S3. Set the planting difficulty score threshold and simulated growth score threshold , when there is a crop planting difficulty score less than And the simulated growth score is greater than the simulated growth score threshold When the number of types is greater than or equal to 1, the crops that meet the conditions are selected as the crops to be planted in the current area, and according to the user's planting needs, the current area is secondary planted according to the number of crop types that meet the conditions and the crop planting attributes. It should be noted that the current area in this embodiment is the area to be developed in the user's planting needs.

[0110] S4. When the planting difficulty scores of all crops are greater than or equal to the planting difficulty score threshold , but there are crops with simulated growth scores greater than the simulated growth score threshold If the species is greater than the simulated growth score threshold, The crop planting attributes of the conditions are evaluated and judged for transformation of the current area, the current area after transformation is obtained, and secondary planting planning is carried out.

[0111] Furthermore, in this embodiment, the step of performing secondary planting planning on the current area includes:

[0112] S301. When only one type of crop meets the conditions, directly plant the corresponding type of crop in the current area;

[0113] S302. When more than one crop meets the requirements, the shade tolerance, drought tolerance, and lodging resistance of each crop type are obtained based on the number of crop types that meet the requirements, as well as the terrain height, slope, soil moisture, distance to water source, water flow path direction, historical wind speed and direction, and planned planting ratio of each crop type that meets the user's planting requirements. The current area is divided using a particle swarm optimization grid algorithm to obtain crop planting sub-areas of the corresponding crop type.

[0114] It should be further explained that the specific implementation method of the particle swarm optimization grid algorithm in this embodiment to divide the current area is as follows:

[0115] First, the current area is divided into regular grid cells. Based on the environmental parameters of each grid cell, such as the average slope and soil moisture in the grid obtained by spatial interpolation algorithm, the matching scores of each environmental parameter and crop attributes are calculated using the hierarchical analysis method to construct the environment-crop matching matrix. Second, the particle swarm is initialized. Each particle represents a regional division scheme. The particle position corresponds to the planting allocation of different crops in each grid cell, and the particle velocity represents the adjustment direction and amplitude of the allocation scheme. In the iterative process, the fitness of each particle is obtained through the objective function according to the planned planting ratio constraint. Third, based on the particle's own history, the particle's own fitness is obtained through the objective function. The particle swarm optimization algorithm iteratively updates the particles based on the optimal position and the global optimal position, and adjusts the crop planting allocation of each grid unit using the speed-position update formula of the particle swarm optimization algorithm. When the iteration reaches the preset number of times or the fitness converges, the optimal crop planting sub-area division scheme is output. In the actual planting stage, based on crop growth monitoring data, such as the vegetation index obtained by the drone multispectral image analysis algorithm, and the yield estimation model to obtain the growth score and yield of each crop, these feedback data are used as new constraints or optimization targets, and the particle swarm optimization algorithm is used again to dynamically adjust the planting area, thereby achieving continuous optimization of the planting plan. The objective function is constructed by maximizing the overall crop yield and minimizing the weighted value of environmental risk in combination with the least squares method. Specifically, based on historical yield data and environmental parameters of the current planting area, a yield estimation model for each crop is constructed using the random forest regression algorithm to obtain the expected yield of each crop in different sub-regions. At the same time, based on environmental parameters such as terrain height, slope, and soil moisture, a fuzzy comprehensive evaluation algorithm is used to evaluate the environmental risks of planting different crops in each sub-region, such as the soil erosion risk of planting drought-intolerant crops in high-slope areas and the lodging risk of crops with weak lodging resistance in high-wind speed areas, and corresponding weights are assigned according to the risk type. Secondly, the least squares method is used to construct the objective function with maximizing the overall crop yield and minimizing the weighted value of environmental risk as the optimization direction. Specifically, the sum of the expected yields of crops in each sub-region is used as the yield optimization item, and the sum of the environmental risk value and the corresponding weight product is used as the risk penalty item. The model parameters are adjusted by the least squares method to maximize the weighted difference between the yield optimization item and the risk penalty item, ultimately forming an objective function that balances yield and risk.

[0116] It should be further explained that the lodging resistance in this embodiment refers to the ability of crops to resist the lodging of stems caused by strong winds. Combined with historical wind speed and direction data, it is used to plan the planting of crops with strong lodging resistance in high wind speed areas to reduce the risk of lodging during maturity.

[0117] Terrain altitude refers to the regional altitude parameter, which affects the vertical distribution of temperature, duration of sunlight and drainage conditions. High-altitude areas may have lower temperatures and large temperature differences between day and night, which are suitable for the cultivation of cool-loving crops.

[0118] The slope reflects the degree of surface inclination, which affects the water flow rate, soil erosion risk and the convenience of mechanical operation. Areas with large slopes need to be matched with crops with well-developed root systems or drought-resistant crops to reduce soil and water loss.

[0119] The planned planting ratio refers to the proportion of each crop planting area set by the user based on market demand, crop rotation plan, etc. It serves as a constraint condition for the particle swarm optimization algorithm to ensure that the optimized regional division results meet the planting planning goals.

[0120] S303: Plant the corresponding crop in the crop planting sub-area, and obtain the crop growth score and corresponding yield during the corresponding planting period;

[0121] S304: Feedback the growth score and yield of the corresponding type of crops to the particle swarm optimization grid algorithm to make real-time adjustments to the crop planting area.

[0122] This process accurately screens crops suitable for planting in the current area by setting planting difficulty scores and simulated growth score thresholds, and conducts secondary planting planning based on user planting needs, significantly improving the scientific nature and economic benefits of agricultural production. First, when only one crop meets the conditions, directly planting that crop simplifies the decision-making process and improves efficiency. When there are multiple crops that meet the conditions, the particle swarm optimization grid algorithm is used to combine crop characteristics such as shade tolerance and drought tolerance; as well as environmental factors such as terrain height, slope, and soil moisture to finely divide the area to ensure that each crop can grow in the optimal environment. This not only optimizes space utilization but also enhances crop resistance and yield stability. Furthermore, by obtaining the growth score and yield of the crop during the corresponding planting period and feeding it back to the particle swarm optimization grid algorithm, dynamic adjustment of the planting area is achieved, enabling the model to be optimized in real time according to actual growth conditions. This method not only improves the accuracy of crop growth predictions but also enhances the system's adaptability, ensuring long-term stable high yield and high efficiency.

[0123] Further, see Figure 2 In this embodiment, the steps of evaluating and judging the transformation of the current area include:

[0124] S5. Set a transformation difficulty threshold based on the user's planting needs and estimated cost data, and evaluate the difficulty of environmental transformation for the crops that meet the conditions in S4 when planted in the current area, to obtain the environmental transformation difficulty for the crops.

[0125] S6. When there is at least one crop that meets the user's planting requirements and the environmental transformation difficulty is less than the transformation difficulty threshold, the crop that meets the conditions is selected for planting, and the current area is secondary transformed based on the number of crop types that meet the conditions, the corresponding crop planting attributes, and the current area's environmental attributes. At the same time, the S3 process is repeated to perform secondary planting planning on the transformed current area.

[0126] S7. When all crops do not meet the user's planting requirements, but there are crops that meet the corresponding environmental modification difficulty and are less than the modification difficulty threshold, the crop with the lowest estimated cost is selected as the corresponding crop for the current area, and the S6 process is repeated to perform environmental modification and planting planning for the current area based on the planting requirement parameters corresponding to the crops that meet the user's planting requirements.

[0127] S8. When the planting difficulty scores of all crops are greater than or equal to the planting difficulty score threshold , and the simulated growth score is less than or equal to the simulated growth score threshold When analyzing all crop attributes, if there is a crop with the lowest planting difficulty score and the highest simulated growth score, the corresponding crop type is the user's planting requirement and the transformation cost is less than or equal to the estimated cost, the current area will be transformed and planted. Otherwise, no planting plan will be made for the current area.

[0128] It should be further explained that, in this embodiment, one implementation method for evaluating and judging the transformation of the current area is as follows:

[0129] Based on the user's planting demand parameters and historical estimated cost data, a judgment matrix is constructed and weights are calculated through the hierarchical analysis method to obtain a scientific transformation difficulty threshold. At the same time, based on the terrain transformation difficulty assessment algorithm and the hydrological adaptability assessment algorithm, the environmental parameters required for crop growth are compared with the measured data of the current area, and the environmental transformation difficulty score of the corresponding crop is obtained through a linear weighted model. When there is at least one crop that meets the planting requirements and the transformation difficulty is lower than the threshold, the region is gridded based on the planting attributes of the crops that meet the conditions, such as shade tolerance and lodging resistance, and environmental attributes such as regional slope and soil moisture, using the multi-objective particle swarm optimization algorithm to determine the planting sub-areas of each crop. Then, based on the three-dimensional terrain planting simulation model and the regional meteorological simulation sub-model, S3 is repeated. Secondary planting planning is carried out in the process; if all crops do not meet the planting requirements but there are crops with transformation difficulty lower than the threshold, based on the material cost, labor cost and other data of the environmental transformation of each crop, the crop with the lowest estimated cost is screened out through a greedy algorithm, and then the transformation and planning are carried out according to the S6 process; when the planting difficulty score of all crops exceeds the threshold and the growth score is lower than the threshold, based on the TOPSIS multi-attribute decision algorithm, the attributes such as crop planting difficulty, simulated growth, and transformation cost are standardized and weighted sorted to screen out the crop with the lowest planting difficulty and the highest growth score. If the crop meets the user's planting requirements and the transformation cost does not exceed the estimated amount, regional transformation and planting planning are carried out based on the hydrological model and terrain transformation algorithm, otherwise the planning process is terminated.

[0130] This process sets a transformation difficulty threshold based on the user's planting needs and estimated costs, and evaluates the environmental transformation difficulty of crops that meet the conditions, ensuring the feasibility and cost-effectiveness of the transformation plan; when there is at least one crop that meets the user's planting needs and has an environmental transformation difficulty lower than the transformation difficulty threshold, this crop is selected for planting, and secondary transformation and planting planning are carried out in combination with crop characteristics and environmental attributes, achieving optimal resource allocation and efficient utilization; for cases where all crops do not fully meet the user's planting needs but the transformation difficulty meets the conditions, the crop with the lowest estimated cost is selected for transformation and planting, reducing cost risks and improving decision-making flexibility; finally, when the planting difficulty and simulated growth scores of all crops are unsatisfactory, by analyzing all crop attributes, crops with transformation costs within the budget and the best overall performance are prioritized for transformation and planting planning, avoiding blind investment and ensuring the sustainability of the project; this method not only optimizes the planting planning process, but also enhances the system's adaptive ability, and can dynamically adjust the plan according to different conditions, accurately matching user needs, reasonably controlling transformation costs, and maximizing crop growth potential.

[0131] Example 2:

[0132] See also Figure 3, another embodiment provided by the present invention: a terrain analysis system based on UAV remote sensing monitoring decision-making, comprising: a simulation module, an initial decision discrimination module and a secondary decision discrimination module;

[0133] The simulation module is used to simulate and analyze the current planting status of the area and obtain the crop planting difficulty score and simulated growth score of the current area;

[0134] The simulation module includes a data acquisition unit, a model building unit and a simulation evaluation unit;

[0135] The data acquisition unit is used to acquire multi-angle terrain images of the current area, laser measurement data, meteorological data and hydrological data, and user planting requirements and estimated cost data;

[0136] The model building unit is used to build a regional three-dimensional terrain planting simulation model with hydrological and meteorological data through a simulation algorithm and a three-dimensional generation algorithm based on the data acquired by the data acquisition unit;

[0137] The simulation evaluation unit is used to simulate and evaluate the planting of the corresponding crops in the crop planting attribute library using a regional three-dimensional terrain planting simulation model with hydrological and meteorological data and a crop segmented growth period growth factor function, thereby obtaining a planting difficulty score and a simulated growth score for each crop in the current region;

[0138] The initial decision-making and discrimination module is used to make planting and transformation decisions in the current area based on the planting difficulty score and simulated growth score of each crop in the current area and the set decision threshold;

[0139] The initial decision-making module includes a first initial decision-making unit and a second initial decision-making unit;

[0140] The first initial decision unit is used to determine the planting difficulty score threshold according to the set planting difficulty score threshold. and simulated growth score threshold , select the crops that meet the first initial decision condition as the crops to be planted in the current area, and according to the user's planting needs, perform secondary planting planning for the current area based on the number of crop types that meet the conditions and the crop planting attributes; the first initial decision condition is: there are crops with a planting difficulty score less than And the simulated growth score is greater than types, and the number of types is greater than or equal to 1;

[0141] The second initial decision unit is used to select crops that meet the second initial decision conditions as crops to be planted in the current area, and to evaluate and determine the current area based on the planting attributes of the crops that meet the second initial decision conditions, obtain the current area after the transformation, and perform secondary planting planning;

[0142] The second initial decision condition is: the difficulty scores of all crops are greater than or equal to , but there are crops with simulated growth scores greater than kind of.

[0143] The secondary decision-making and discrimination module is used to evaluate the difficulty of transforming the current area and perform regional transformation discrimination;

[0144] The secondary decision-making and discrimination module includes a transformation evaluation unit, a first transformation decision unit, a second transformation decision unit and a third transformation decision unit;

[0145] The transformation evaluation unit is used to set a transformation difficulty threshold based on the user's planting needs and estimated cost data, and evaluate the difficulty of environmental transformation corresponding to the planting of the corresponding crop in the current area that meets the second initial decision condition, to obtain the environmental transformation difficulty of the corresponding crop;

[0146] The first transformation decision unit is used to select crops that meet the conditions for planting according to the first transformation decision conditions, and perform a secondary transformation on the current area based on the number of crop types that meet the conditions, the corresponding crop planting attributes, and the current area environmental attributes, and repeat the S3 process to perform a secondary planting plan on the transformed current area;

[0147] The first transformation decision condition is: there is at least one crop that meets the user's planting needs and the environmental transformation difficulty is less than the transformation difficulty threshold;

[0148] The second transformation decision unit is used to select the crop with the lowest estimated cost as the corresponding planting crop in the current area according to the second transformation decision condition, and repeat the S6 process to perform environmental transformation and planting planning in the current area;

[0149] The second transformation decision condition is: all crops do not meet the user's planting requirements, but there are crops that meet the environmental transformation difficulty lower than the transformation difficulty threshold;

[0150] The third transformation decision unit is used to select, according to the third transformation decision condition, a crop with the smallest planting difficulty score, the largest simulated growth score, a corresponding crop type that meets the user's planting requirements, and a transformation cost that is less than or equal to the estimated cost as a crop to be planted in the current area, and to transform and plan planting for the current area based on the selected crop; otherwise, no planting plan is performed for the current area;

[0151] The third transformation decision condition is: the planting difficulty scores of all crops are greater than or equal to the planting difficulty score threshold, and the simulated growth scores are less than or equal to the simulated growth score threshold.

[0152] The embodiments of the present invention are described above in conjunction with the accompanying drawings, but the present invention is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of the present invention, ordinary technicians in this field can also change, modify, replace and modify the above-mentioned embodiments without departing from the scope of protection of the purpose of the present invention and the claims, and all of these are protected by the present invention.

Claims

1. A terrain analysis method based on UAV remote sensing monitoring decision-making, characterized by the following steps: include: S1. Obtain multi-angle terrain images, laser measurement data, meteorological data, hydrological data, and user planting requirements and estimated cost data for the current area. Build a regional 3D terrain planting simulation model with hydrological and meteorological data by combining simulation algorithms with 3D generation algorithms. S2. Configure a crop planting attribute library. Based on the planting requirement parameters of each crop in the library, perform planting simulation and evaluation using a regional three-dimensional terrain planting simulation model with hydrological and meteorological data and a crop growth factor function for each growth period. This results in a planting difficulty score and simulated growth score for each crop in the current region. S3. Set the planting difficulty score threshold and simulated growth score threshold , when there is a crop planting difficulty score less than And the simulated growth score is greater than the simulated growth score threshold When the number of types is greater than or equal to 1, the crops that meet the conditions are selected as the crops planted in the current area, and according to the user's planting needs, the current area is secondary planted according to the number of crop types that meet the conditions and the crop planting attributes; S4. When the planting difficulty scores of all crops are greater than or equal to the planting difficulty score threshold , but there are crops with simulated growth scores greater than the simulated growth score threshold If the species is greater than the simulated growth score threshold, The crop planting attributes of the conditions are evaluated and judged for transformation of the current area, the current area after transformation is obtained, and secondary planting planning is carried out.

2. The terrain analysis method based on UAV remote sensing monitoring decision-making according to claim 1 is characterized in that: The steps of evaluating and judging the transformation of the current area include: S5. Set a transformation difficulty threshold based on the user's planting needs and estimated cost data, and evaluate the difficulty of environmental transformation for the crops that meet the conditions in S4 when planted in the current area, to obtain the environmental transformation difficulty for the crops. S6. When there is at least one crop that meets the user's planting requirements and the environmental transformation difficulty is less than the transformation difficulty threshold, the crop that meets the conditions is selected for planting, and the current area is secondary transformed based on the number of crop types that meet the conditions, the corresponding crop planting attributes, and the current area's environmental attributes. At the same time, the S3 process is repeated to perform secondary planting planning on the transformed current area. S7. When all crops do not meet the user's planting requirements, but there are crops that meet the corresponding environmental modification difficulty and are less than the modification difficulty threshold, the crop with the lowest estimated cost is selected as the corresponding crop for the current area, and the S6 process is repeated to perform environmental modification and planting planning for the current area based on the planting requirement parameters corresponding to the crops that meet the user's planting requirements. S8. When the planting difficulty scores of all crops are greater than or equal to the planting difficulty score threshold , and the simulated growth score is less than or equal to the simulated growth score threshold When analyzing all crop attributes, if there is a crop with the lowest planting difficulty score and the highest simulated growth score, the corresponding crop type is the user's planting requirement and the transformation cost is less than or equal to the estimated cost, the current area will be transformed and planted. Otherwise, no planting plan will be made for the current area.

3. The terrain analysis method based on UAV remote sensing monitoring decision-making according to claim 2 is characterized in that: The steps of constructing the regional three-dimensional terrain planting simulation model with hydrological and meteorological features include: S101, generating a three-dimensional terrain sub-model of the current area using a three-dimensional generation algorithm based on the acquired multi-angle terrain image of the current area and the slope and terrain height data in the laser measurement data; S102: Pre-train the configured weather forecast model based on the acquired current regional weather data to obtain a regional weather simulation sub-model, and output the simulated weather status for each time period in the current region; S103. Based on the current regional hydrological data, the configured hydrological model is trained to obtain a regional hydrological simulation sub-model, and the distance between the water source and the current region, the water flow path, and the soil moisture are simulated. S104, integrating the current regional three-dimensional terrain sub-model, the regional meteorological simulation sub-model, and the regional hydrological simulation sub-model through an integration algorithm to obtain a regional three-dimensional terrain model with hydrological and meteorological features; S105: constructing a segmented growth period growth factor function based on the planting environment conditions and simulated meteorological conditions of each crop growth period, the distance between the water source and the current area, the water flow path, and the soil moisture. The segmented growth period growth factor function is embedded in the constructed crop segmented growth simulation sub-model for training to obtain the corresponding growth state of each crop under different growth conditions. S106. Integrate the pre-trained crop segmented growth simulation sub-model into the regional three-dimensional terrain model with hydrometeorology to obtain a regional three-dimensional terrain planting simulation model with hydrometeorology. When the current region changes, repeat the S101-S106 process to obtain a fine-tuned regional three-dimensional terrain planting simulation model with hydrometeorology for each region, and dynamically annotate the simulated meteorological, hydrological and crop growth status according to the time dimension.

4. The terrain analysis method based on UAV remote sensing monitoring decision-making according to claim 3 is characterized in that: The steps for constructing the segmented growth factor function are as follows: S201. Set the crop growth stages to include the sowing stage, seedling stage, vegetative growth stage, reproductive growth stage, and maturity stage. Based on the relationship between each crop's parameters at different growth stages and meteorological and hydrological parameters, use a factor analysis algorithm to analyze and obtain meteorological and hydrological parameter spaces whose contribution is greater than a contribution threshold. S202. Based on the meteorological and hydrological parameter spaces whose contribution is greater than the contribution threshold, a segmented growth period growth factor function corresponding to each crop is constructed by combining a multivariate regression algorithm with a genetic optimization algorithm.

5. The terrain analysis method based on UAV remote sensing monitoring decision-making according to claim 4 is characterized in that: The steps of obtaining the crop planting difficulty score include: S211. Obtain a comprehensive terrain transformation difficulty score based on the current area slope, terrain height difference, and terrain complexity obtained through the evaluation algorithm; S212. Obtain a meteorological adaptability score for the current region using an evaluation algorithm based on the current average temperature, precipitation, and wind speed of the region. S213, obtaining a hydrological adaptability score for the current region through an evaluation algorithm based on the distance between the water source and the current region, the water flow path, and soil moisture information; S214. Obtain the crop planting difficulty score for the current region by weighted average method based on the comprehensive terrain transformation difficulty score, meteorological adaptability score, and hydrological adaptability score for the current region.

6. The terrain analysis method based on UAV remote sensing monitoring decision-making according to claim 5 is characterized in that: The steps of performing secondary planting planning on the current area in S3 include: S301. When only one type of crop meets the conditions, directly plant the corresponding type of crop in the current area; S302. When more than one crop meets the requirements, the shade tolerance, drought tolerance, and lodging resistance of each crop type are obtained based on the number of crop types that meet the requirements, as well as the terrain height, slope, soil moisture, distance to water source, water flow path direction, historical wind speed and direction, and planned planting ratio of each crop type that meets the user's planting requirements. The current area is divided using a particle swarm optimization grid algorithm to obtain crop planting sub-areas of the corresponding crop type. S303: Plant the corresponding crop in the crop planting sub-area, and obtain the crop growth score and corresponding yield during the corresponding planting period; S304: Feedback the growth score and yield of the corresponding type of crops to the particle swarm optimization grid algorithm to make real-time adjustments to the crop planting area.

7. A terrain analysis system based on UAV remote sensing monitoring and decision-making, which is used to implement the terrain analysis method based on UAV remote sensing monitoring and decision-making according to any one of claims 1 to 6, characterized in that: include: Simulation module, initial decision-making and discrimination module and secondary decision-making and discrimination module; The simulation module includes a data acquisition unit, a model building unit and a simulation evaluation unit; The data acquisition unit is used to acquire multi-angle terrain images of the current area, laser measurement data, meteorological data and hydrological data, and user planting requirements and estimated cost data; The model building unit is used to build a regional three-dimensional terrain planting simulation model with hydrological and meteorological data through a simulation algorithm and a three-dimensional generation algorithm based on the data acquired by the data acquisition unit; The simulation evaluation unit is used to perform planting simulation and evaluation based on the corresponding crops in the crop planting attribute library through a regional three-dimensional terrain planting simulation model with hydrological and meteorological conditions and a crop segmented growth period growth factor function, to obtain the planting difficulty score and simulated growth score of each crop in the current area.

8. The terrain analysis system based on UAV remote sensing monitoring and decision-making according to claim 7, characterized in that: The initial decision-making module includes a first initial decision-making unit and a second initial decision-making unit; The first initial decision unit is used to determine the planting difficulty score threshold according to the set planting difficulty score threshold. and simulated growth score threshold , select the crops that meet the first initial decision condition as the crops to be planted in the current area, and according to the user's planting needs, perform secondary planting planning for the current area based on the number of crop types that meet the conditions and the crop planting attributes; the first initial decision condition is: there are crops with a planting difficulty score less than And the simulated growth score is greater than types, and the number of types is greater than or equal to 1; The second initial decision unit is used to select crops that meet the second initial decision conditions as crops to be planted in the current area, and to evaluate and determine the current area based on the planting attributes of the crops that meet the second initial decision conditions, obtain the current area after the transformation, and perform secondary planting planning; The second initial decision condition is: the difficulty scores of all crops are greater than or equal to , but there are crops with simulated growth scores greater than kind of.

9. The terrain analysis system based on UAV remote sensing monitoring and decision-making according to claim 8, characterized in that: The secondary decision-making and discrimination module includes a transformation evaluation unit, a first transformation decision unit, a second transformation decision unit and a third transformation decision unit; The transformation evaluation unit is used to set a transformation difficulty threshold based on the user's planting needs and estimated cost data, and evaluate the difficulty of environmental transformation corresponding to the planting of the corresponding crop in the current area that meets the second initial decision condition, to obtain the environmental transformation difficulty of the corresponding crop; The first transformation decision unit is used to select crops that meet the conditions for planting according to the first transformation decision conditions, and perform a secondary transformation on the current area based on the number of crop types that meet the conditions, the corresponding crop planting attributes, and the current area environmental attributes, and repeat the S3 process to perform a secondary planting plan on the transformed current area; The first transformation decision condition is: there is at least one crop that meets the user's planting needs and the environmental transformation difficulty is less than the transformation difficulty threshold; The second transformation decision unit is used to select the crop with the lowest estimated cost as the corresponding planting crop in the current area according to the second transformation decision condition, and repeat the S6 process to perform environmental transformation and planting planning in the current area; The second transformation decision condition is: all crops do not meet the user's planting requirements, but there are crops that meet the environmental transformation difficulty lower than the transformation difficulty threshold; The third transformation decision unit is used to select, according to the third transformation decision condition, a crop with the smallest planting difficulty score, the largest simulated growth score, a corresponding crop type that meets the user's planting requirements, and a transformation cost that is less than or equal to the estimated cost as a crop to be planted in the current area, and to perform transformation and planting planning for the current area based on the selected crop; The third transformation decision condition is: the planting difficulty scores of all crops are greater than or equal to the planting difficulty score threshold, and the simulated growth scores are less than or equal to the simulated growth score threshold.

Images