Crop assisting method, system and equipment based on multi-source remote sensing data and medium

Abstract

The invention discloses a crop assistance method, system and equipment based on multi-source remote sensing data and a medium, mainly relates to the technical field of crop assistance, and is used for solving the problem that water resources are wasted due to the fact that a traditional irrigation system cannot distinguish water demand characteristics of different crops. The manual field crop type investigation efficiency is low, and the large-range dynamic monitoring requirement is difficult to meet. Comprising the steps of determining crop varieties and growth stages of each preset area based on an unmanned aerial vehicle, a satellite image and a user terminal; obtaining a reference evapotranspiration value and a soil moisture variation, and obtaining the actual water demand of crops in the preset area according to the crop variety, the growth stage, the soil moisture variation and the reference evapotranspiration value; based on the initial irrigation sequence of the preset area and the fitness function, utilizing a genetic algorithm to obtain an optimized irrigation sequence; and based on the optimized irrigation sequence and the actual water demand of the crops, controlling the operation of the irrigation devices in the preset areas.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of crop auxiliary technology, and in particular to a crop auxiliary method and system based on multi-source remote sensing data, a device, and a medium. BACKGROUND

[0002] Current agricultural irrigation mainly relies on two types of technical systems: one is an automatic irrigation system based on time sequence control, which starts and stops the water pump at a preset time through a preset program, such as a center pivot sprinkler; the other is a soil moisture feedback system, which triggers irrigation with the help of a buried humidity sensor (such as TDR or FDR type), and the threshold is usually set at 60%-70% of the field water holding capacity. Crop type identification mainly relies on artificial field investigation, using visual discrimination combined with phenological period recording, and some pilot areas use near-ground spectrometers (such as ASD FieldSpec) for auxiliary identification. The data of the Food and Agriculture Organization of the United Nations shows that about 78% of the large-scale farms in the world use the above technical combination, and the average irrigation water use coefficient is 0.45-0.55.

[0003] The existing technology has three key defects: first, the crop water demand model parameters are fixed, and the water demand characteristics of different varieties are not considered, and the traditional irrigation system cannot distinguish the water demand characteristics of different crops, resulting in waste of water resources. Second, artificial field investigation of crop types is low in efficiency and difficult to meet the needs of large-scale dynamic monitoring SUMMARY The present application provides a crop auxiliary method and system based on multi-source remote sensing data, a device, and a medium to solve the problem that the traditional irrigation system cannot distinguish the water demand characteristics of different crops, resulting in waste of water resources. Artificial field investigation of crop types is low in efficiency and difficult to meet the needs of large-scale dynamic monitoring.

[0004] In a first aspect, the present application provides a crop auxiliary method based on multi-source remote sensing data, the method comprising: determining the crop varieties and growth stages of each preset area based on unmanned aerial vehicles, satellite images, and user terminals; obtaining the actual water demand of crops in the preset area according to the crop varieties, growth stages, soil moisture changes, and reference evapotranspiration values; obtaining the optimized irrigation sequence by using a genetic algorithm based on the initial irrigation sequence of the preset area and the fitness function; controlling the operation of the irrigation devices in each preset area based on the optimized irrigation sequence and the actual water demand of crops.

[0005] In an implementation manner of the present application, the crop varieties and growth stages of each preset area are determined based on unmanned aerial vehicles, satellite images, and user terminals, specifically comprising: Upload regional crop images collected by drones to the trained deep learning recognition model to obtain the first predicted crop variety, first predicted growth stage, and prediction accuracy; Upload regional crop images collected by satellite imagery to the trained deep learning recognition model to obtain the second predicted crop variety, second predicted growth stage, and prediction accuracy; Determine the crop variety and growth stage corresponding to the highest prediction accuracy, which is the first recognition result; Determining whether the first recognition result is consistent with the crop variety and growth stage of the current region uploaded by the user terminal; When they are consistent, determining the crop variety and growth stage of the current region in the first recognition result as the final crop variety and growth stage of the current region; When there is inconsistency and the highest prediction accuracy is greater than a preset accuracy threshold, the crop variety and growth stage in the first recognition result are determined as the final crop variety and growth stage in the current area; When the highest prediction accuracy is less than or equal to a preset accuracy threshold, the crop variety and growth stage in the user terminal are determined to be the final crop variety and growth stage in the current area.

[0006] In one implementation of the present application, the actual water requirement of crops in a preset area is obtained based on the crop variety, growth stage, soil moisture change, and reference evapotranspiration value, specifically including: Determine the crop coefficient corresponding to the current crop variety and growth stage from a preset mapping database between crop varieties, growth stages and crop coefficients; By formula: ET_c= (K_c×ET_0+ΔSM×γ, calculate the actual water requirement of crops; Among them, ET_c represents the actual water requirement of the crop, K_c represents the preset crop coefficient, ΔSM represents the change in soil moisture, and γ represents the correction coefficient.

[0007] In one implementation of the present application, based on the initial irrigation sequence and fitness function of the preset area, a genetic algorithm is used to obtain an optimized irrigation sequence, specifically including: The initial irrigation order of the preset area is encoded into chromosomes, and each gene represents an irrigation preset area. Randomly generate N groups of irrigation sequences as the initial population; Get the fitness calculation function; The fitness calculation function and the initial population are input into the genetic algorithm, and the irrigation sequence with the highest fitness in the iterative population is output.

[0008] In an implementation manner of the present application, based on the optimized irrigation sequence and the actual water requirement of crops, the operation of the irrigation device in each preset area is controlled, specifically including: Based on the irrigation sequence, the preset area currently being irrigated is determined; Based on the PID algorithm, the rotation speed of the water pump in the irrigation device corresponding to the preset area is dynamically controlled until the actual water requirement of the crops is completed.

[0009] In a second aspect, the present application provides a crop auxiliary system based on multi-source remote sensing data, which comprises: The acquisition module is configured to determine the crop variety and growth stage of each preset area based on the unmanned aerial vehicle, satellite image and user terminal; The water amount obtaining module is configured to obtain the reference evapotranspiration value and the soil moisture variation, and obtain the actual water requirement of crops in the preset area according to the crop variety, growth stage, soil moisture variation and reference evapotranspiration value; The sequence obtaining module is configured to obtain the optimized irrigation sequence based on the initial irrigation sequence of the preset area and the fitness function by using the genetic algorithm; The control module is configured to control the operation of the irrigation device in each preset area based on the optimized irrigation sequence and the actual water requirement of crops.

[0010] In an implementation manner of the present application, the acquisition module comprises an acquisition unit, configured to upload the region crop image collected by the unmanned aerial vehicle to the trained deep learning recognition model to obtain the first predicted crop variety, the first predicted growth stage and the prediction accuracy; configured to upload the region crop image collected by the satellite image to the trained deep learning recognition model to obtain the second predicted crop variety, the second predicted growth stage and the prediction accuracy; configured to determine the crop variety and growth stage corresponding to the highest prediction accuracy as the first identification result; configured to determine whether the first identification result is consistent with the current region crop variety and growth stage uploaded by the user terminal; configured to determine that the current region crop variety and growth stage in the first identification result are the final current region crop variety and growth stage when they are consistent; configured to determine that the crop variety and growth stage in the first identification result are the final current region crop variety and growth stage when they are inconsistent and the highest prediction accuracy is greater than a preset accuracy threshold; configured to determine that the crop variety and growth stage in the user terminal are the final current region crop variety and growth stage when the highest prediction accuracy is less than or equal to the preset accuracy threshold.

[0011] In an implementation manner of the present application, the water amount obtaining module comprises a water amount obtaining unit, Used to determine the crop coefficient corresponding to the current crop variety and growth stage from a preset mapping database between crop varieties, growth stages and crop coefficients; By formula: ET_c= (K_c×ET_0+ΔSM×γ, calculate the actual water requirement of crops; Among them, ET_c represents the actual water requirement of the crop, K_c represents the preset crop coefficient, ΔSM represents the change in soil moisture, and γ represents the correction coefficient.

[0012] In a third aspect, the present application provides a crop auxiliary device based on multi-source remote sensing data, the device comprising: processor; and a memory storing executable codes, which, when executed, enable the processor to execute any one of the above-mentioned crop assistance methods based on multi-source remote sensing data.

[0013] In a fourth aspect, the present application provides a non-volatile computer storage medium having computer instructions stored thereon, which, when executed, implement any of the above-mentioned crop assistance methods based on multi-source remote sensing data.

[0014] It can be seen from the above technical solutions that this application has the following advantages: Multi-source data fusion enables precise crop identification and dynamic monitoring. By integrating drone-based near-ground remote sensing, satellite imagery for macroscopic monitoring, and user-terminal field data, this three-dimensional data collection system overcomes the temporal and spatial limitations of traditional manual surveys. This technical solution achieves three key benefits: First, drone-mounted hyperspectral sensors can identify crop variety-specific spectral signatures, addressing the "blind spot" in crop type identification in traditional irrigation systems; second, satellite imagery provides periodic, large-scale data coverage, expanding monitoring from single plots to regional scales (e.g., 10,000 mu of farmland); and third, user terminals supplement manual verification data from crop growth stages, forming an "air-space-ground" collaborative verification mechanism. This multi-scale data fusion improves the accuracy of crop type prediction and increases monitoring frequency from the traditional manual monthly level to daily levels.

[0015] The dual benefits of the intelligent water demand calculation and irrigation strategy optimization based on the evapotranspiration model and the dynamic calculation architecture of soil moisture feedback are: at the data level, by coupling the reference evapotranspiration value (ET0) calculated by the Penman-Monteith equation and the measured data of the soil moisture sensor, a real-time calculation model of the actual water demand of crops (ETc) is established, which is more water-saving than the traditional fixed irrigation quota mode; at the control strategy level, the genetic algorithm driven irrigation sequence optimization realizes three-order improvement: first, the irrigation priority is matched with the critical water demand period of crops (such as the priority of the wheat heading stage), second, the adaptability function balances the water resource allocation efficiency and the equipment energy consumption, and finally, a dynamic irrigation scheme that can adapt to climate change (such as drought warning) is formed. Through testing, the algorithm can reduce the energy consumption of the pumping station while improving the timeliness of irrigation.

[0016] The engineering practice advantages brought by the closed-loop control system directly couple the optimization algorithm with the irrigation equipment control system to produce significant engineering benefits: on the one hand, through industrial protocols such as Modbus / 4G, the "calculation-execution" second-level response is realized, eliminating the time lag of traditional manual scheduling; on the other hand, it supports differentiated control of multiple irrigation devices (such as center pivot, drip irrigation tape), and a single controller can manage the independent start and stop of multiple partition valves. This closed-loop control enables the system to automatically compensate for the uneven irrigation caused by pipe pressure fluctuations. BRIEF DESCRIPTION OF DRAWINGS

[0017] In order to more clearly illustrate the technical solutions of the present application, the drawings needed to be used in the description will be briefly introduced as follows. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can also be obtained by those skilled in the art without creative labor.

[0018] Figure 1 is a flow chart of a crop auxiliary method based on multi-source remote sensing data provided by an embodiment of the present application.

[0019] Figure 2 is a schematic diagram of the internal structure of a crop auxiliary system based on multi-source remote sensing data provided by an embodiment of the present application.

[0020] Figure 3 is a schematic diagram of the internal structure of a crop auxiliary device based on multi-source remote sensing data provided by an embodiment of the present application. DETAILED DESCRIPTION

[0021] Clearly, the embodiments described are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments of the present application, all other embodiments obtained by a person of ordinary skill in the art without creative effort should fall within the protection scope of the present application.

[0022] It should be understood by those skilled in the art that the embodiments described below are only preferred embodiments of the present disclosure, and do not represent that the present disclosure can only be implemented by the preferred embodiments. The preferred embodiments are only used to explain the technical principles of the present disclosure, and are not used to limit the protection scope of the present disclosure. Based on the preferred embodiments provided by the present disclosure, all other embodiments obtained by a person of ordinary skill in the art without creative effort should fall within the protection scope of the present disclosure.

[0023] It should be further noted that the terms "comprising", "containing" or any other variants thereof are intended to cover non-exclusive inclusion, so that the process, method, article or equipment including a series of elements not only includes those elements, but also includes other elements not explicitly listed or inherent to such process, method, article or equipment. Without more limitations, the element defined by the statement "comprising a" does not exclude the presence of another identical element in the process, method, article or equipment including the element.

[0024] The technical solutions of the embodiments of the present application will be described in detail below with reference to the drawings.

[0025] The embodiments provide an auxiliary method for crops based on multi-source remote sensing data, as shown in Figure 1 The method provided by the embodiments of the present application mainly includes the following steps: Step 110, determining the crop variety and growth stage of each preset area based on the unmanned aerial vehicle, satellite image and user terminal.

[0026] In some embodiments, determining the crop variety and growth stage of each preset area based on the unmanned aerial vehicle, satellite image and user terminal specifically includes: uploading the region crop image collected by the unmanned aerial vehicle to the trained deep learning identification model to obtain the first predicted crop variety, the first predicted growth stage and the prediction accuracy; uploading the region crop image collected by the satellite image to the trained deep learning identification model to obtain the second predicted crop variety, the second predicted growth stage and the prediction accuracy; determining the crop variety and growth stage corresponding to the highest prediction accuracy as the first identification result; determine whether the first recognition result is consistent with the current regional crop variety and growth stage uploaded by the user terminal; when consistent, determine that the current regional crop variety and growth stage in the first recognition result are the final current regional crop variety and growth stage; when inconsistent and the highest prediction accuracy is greater than the preset accuracy threshold, determine that the crop variety and growth stage in the first recognition result are the final current regional crop variety and growth stage; when the highest prediction accuracy is less than or equal to the preset accuracy threshold, determine that the crop variety and growth stage in the user terminal are the final current regional crop variety and growth stage.

[0027] Based on the above description, this step improves the reliability of crop recognition through a multi-source data collaborative verification mechanism. The core benefits are reflected in three aspects: in the data acquisition dimension, the high-resolution images (0.5-2 cm / pixel) of the unmanned aerial vehicle can capture the texture details of the crop canopy, satellite images (such as 10 m resolution of Sentinel-2) provide wide coverage, and user terminal data serve as ground truth reference, forming a data acquisition system with complementary spatial scales. In the algorithm decision-making layer, a hybrid decision-making mode of "multi-model voting + artificial verification" is adopted: the deep learning model performs confidence evaluation on the recognition results of the unmanned aerial vehicle / satellite images (outputs an accuracy quantization index), when there is a disagreement among the models, the automatic recognition result with a confidence higher than a preset threshold (such as 85%) is preferred, otherwise, it falls back to manual input data. This mechanism can avoid the misjudgment risk of a single data source. In terms of business adaptability, the dynamic adjustment function of the preset accuracy threshold allows different standards to be set according to crop type differences (such as setting the rice recognition threshold to 90% and the corn threshold to 80%), which is more in line with the actual needs of agricultural scenarios compared to the fixed threshold scheme. The entire process realizes seamless connection between the unmanned aerial vehicle, satellite data platform and agricultural management system through standardized API interface, meeting the rapid monitoring needs of key nodes (such as the heading stage) in the growth period. The biggest feature of this step is to establish an elastic cooperation rule between machine recognition and artificial judgment, which not only retains the efficiency advantage of automated processing, but also ensures the reliability of high-risk decisions through the artificial verification link.

[0028] In step 120, a reference evapotranspiration value and a soil moisture change amount are obtained, and according to the crop variety, the growth stage, the soil moisture change amount and the reference evapotranspiration value, the actual water requirement of crops in a preset area is obtained.

[0029] Among them, according to the crop variety, the growth stage, the soil moisture change amount and the reference evapotranspiration value, the actual water requirement of crops in a preset area is obtained, specifically including: determining the crop coefficient corresponding to the current crop variety and growth stage from the preset mapping database between the crop variety, the growth stage and the crop coefficient; By formula: ET_c= (K_c×ET_0+ΔSM×γ, calculate the actual water requirement of crops; Wherein, ET_c represents the actual water requirement of crops, K_c represents the preset crop coefficient, ΔSM represents the soil moisture variation, and γ represents the correction coefficient.

[0030] Based on the above description, this step realizes the accurate calculation of crop water requirement by fusing meteorological data and soil moisture monitoring. The technical benefits mainly reflect in three aspects: in the aspect of data integration, the Penman-Monteith equation recommended by FAO is used to calculate the reference evapotranspiration value (ET0), combined with the ΔSM data measured by the soil humidity sensor, to form a meteorological-soil double-factor driven water requirement calculation model. Compared with the traditional method which depends on meteorological data alone, it is more in line with the actual water absorption law of crops. In the aspect of calculation model construction, the difference between varieties is realized by the preset crop coefficient database, for example, the Kc value of rice is 1.05 during the tillering period and decreases to 0.9 during the mature period. This dynamic adjustment mechanism during the growth stage can accurately reflect the physiological water requirement change of crops. The γ correction coefficient (usually 0.6-0.8) introduced in the formula effectively compensates for the change of water transport rate caused by soil texture difference, so that the deviation of water requirement calculation results of sandy soil and clay soil is reduced. In the aspect of engineering application, this model supports automatic update calculation every hour. When the soil moisture sensor detects a rainfall event, the system can immediately reduce the irrigation amount to avoid water resource waste caused by superimposed natural precipitation. The entire calculation process uses standardized data interface, which can be compatible with the soil moisture data format of most agricultural Internet of Things systems, and has strong equipment adaptability.

[0031] Step 130, based on the initial irrigation sequence of the preset area and the fitness function, the optimized irrigation sequence is obtained by using genetic algorithm.

[0032] This step can be specifically: Encode the initial irrigation sequence of the preset area into a chromosome, and each gene represents an irrigation preset area Randomly generate N groups of irrigation sequences as the initial population; Obtain the fitness calculation function; Input the fitness calculation function and the initial population into the genetic algorithm, and output the irrigation sequence with the highest fitness in the iteration population.

[0033] Based on the above description, the genetic algorithm is used to optimize the irrigation sequence, and the core advantages are reflected in three aspects: at the algorithm design level, the irrigation area is coded as a chromosome gene representation method, so that each irrigation sequence scheme can be expressed as a computable individual. This coding method preserves the spatial correlation between regions. The random generation mechanism of the initial population (usually set N = 50-100) ensures extensive coverage of the search space, avoiding local optimal solutions. The construction of the fitness function considers three key parameters: the crop water stress index (0-1 standardization), the start-stop energy consumption coefficient of the irrigation equipment, and the pipe network water pressure stability index. By weighted summation, a quantitative evaluation standard is formed (such as setting the weight 0.5:0.3:0.2), so that the optimization direction matches the actual agronomic demand. In terms of computational efficiency, the selection-crossover-variation operation of the genetic algorithm (typical parameters: crossover probability 0.7, mutation probability 0.01) can converge within about 50 iterations, meeting the adjustment frequency requirement of the daily irrigation plan. The entire optimization process outputs the irrigation sequence in JSON format through the REST API interface, which can directly drive the execution module of most intelligent irrigation control systems.

[0034] Step 140, based on the optimized irrigation sequence and the actual water requirement of crops, control the operation of the irrigation device in each preset area.

[0035] In some embodiments, based on the optimized irrigation sequence and the actual water requirement of crops, control the operation of the irrigation device in each preset area, specifically including: Based on the irrigation sequence, determine the preset area currently being irrigated; Based on the PID algorithm, dynamically control the rotation speed of the water pump in the irrigation device corresponding to the preset area until the actual water requirement of the irrigated crops is completed.

[0036] In addition, the present application Figure 2 A crop auxiliary system based on multi-source remote sensing data is provided for the embodiments of the present application. As Figure 2 shown, the system provided by the embodiments of the present application mainly includes: The acquisition module 210 is configured to determine the crop variety and growth stage of each preset area based on the unmanned aerial vehicle, satellite image, and user terminal.

[0037] The acquisition module 210 includes an acquisition unit, configured to upload the region crop image collected by the unmanned aerial vehicle to the trained deep learning identification model to obtain the first predicted crop variety, the first predicted growth stage, and the prediction accuracy; upload the region crop image collected by the satellite image to the trained deep learning identification model to obtain the second predicted crop variety, the second predicted growth stage, and the prediction accuracy; Determine the crop variety and growth stage corresponding to the highest prediction accuracy, which is the first recognition result; Determining whether the first recognition result is consistent with the crop variety and growth stage of the current region uploaded by the user terminal; When they are consistent, determining the crop variety and growth stage of the current region in the first recognition result as the final crop variety and growth stage of the current region; When there is inconsistency and the highest prediction accuracy is greater than a preset accuracy threshold, the crop variety and growth stage in the first recognition result are determined as the final crop variety and growth stage in the current area; When the highest prediction accuracy is less than or equal to a preset accuracy threshold, the crop variety and growth stage in the user terminal are determined to be the final crop variety and growth stage in the current area.

[0038] The water acquisition module 220 is used to obtain reference evapotranspiration values ​​and soil moisture changes, and obtain the actual water requirements of crops in a preset area based on crop varieties, growth stages, soil moisture changes and reference evapotranspiration values.

[0039] The water quantity acquisition module 220 includes a water quantity acquisition unit, Used to determine the crop coefficient corresponding to the current crop variety and growth stage from a preset mapping database between crop varieties, growth stages and crop coefficients; By formula: ET_c= (K_c×ET_0+ΔSM×γ, calculate the actual water requirement of crops; Among them, ET_c represents the actual water requirement of the crop, K_c represents the preset crop coefficient, ΔSM represents the change in soil moisture, and γ represents the correction coefficient.

[0040] The sequence obtaining module 230 is used to obtain an optimized irrigation sequence based on the initial irrigation sequence of the preset area and the fitness function using a genetic algorithm.

[0041] The control module 240 is used to control the operation of the irrigation devices in each preset area based on the optimized irrigation sequence and the actual water demand of the crops.

[0042] The above is a method embodiment of the present application. Based on the same inventive concept, the present application embodiment also provides a crop auxiliary device based on multi-source remote sensing data. Figure 3 As shown, the device includes: a processor; and a memory on which executable codes are stored. When the executable codes are executed, the processor executes a crop assistance method based on multi-source remote sensing data as described in the above embodiment.

[0043] Specifically, the server determines crop varieties and growth stages of each preset area based on a UAV, satellite images and a user terminal; acquires a reference evapotranspiration value and a soil moisture change amount; obtains actual crop water requirements of the preset area according to the crop varieties, the growth stages, the soil moisture change amount and the reference evapotranspiration value; obtains an optimized irrigation sequence by using a genetic algorithm based on an initial irrigation sequence of the preset area and a fitness function; and controls operation of irrigation devices in each preset area based on the optimized irrigation sequence and the actual crop water requirements.

[0044] Based on the foregoing description, the embodiment realizes accurate crop identification and dynamic monitoring through multi-source data fusion by integrating a UAV near-earth remote sensing, satellite image macro monitoring and a three-dimensional data acquisition system of user terminal field data, breaking through the time and space limitations of traditional manual investigation. The technical solution first realizes three core benefits: first, the UAV carrying a hyperspectral sensor can identify crop variety specific spectral characteristics, solving the "crop type identification blind area" problem of traditional irrigation systems; second, satellite images provide periodic large-scale coverage data, expanding the monitoring range from a single field to a regional level (such as tens of thousands of mu of farmland); and third, the user terminal supplements manual verification data of crop growth stages, forming a "space-air-ground" collaborative verification mechanism. This multi-scale data fusion improves the accuracy of crop type prediction and increases the monitoring frequency from the traditional monthly level to the daily level.

[0045] Intelligent water requirement calculation and irrigation strategy optimization produce double benefits based on a dynamic calculation architecture of an evapotranspiration model and soil moisture feedback: at the data level, a real-time calculation model of actual crop water requirements (ETc) is established by coupling the reference evapotranspiration value (ET0) calculated by the Penman-Monteith equation and the measured data of the soil moisture sensor, which is more water-saving than the traditional fixed irrigation quota mode; and at the control strategy level, the genetic algorithm driven irrigation sequence optimization realizes three-order improvement: first, the irrigation priority is matched with the critical water requirement period of crops (such as priority for wheat at the booting stage), second, the fitness function balances water resource allocation efficiency and equipment energy consumption, and finally, a dynamic irrigation scheme that can adapt to climate change (such as drought warning) is formed. Tests show that this algorithm can reduce the energy consumption of the pumping station while improving the timeliness of irrigation.

[0046] The engineering practice advantages brought by the closed-loop control system directly couple the optimization algorithm with the irrigation equipment control system to produce significant engineering benefits: on the one hand, the Modbus / 4G industrial protocol realizes "calculation-execution" second-level response, eliminating the time lag of traditional manual scheduling; on the other hand, it supports differentiated control of multiple irrigation devices (such as center pivot and drip irrigation tape), and a single controller can manage the independent start and stop of multiple sub-zone valves. This closed-loop control enables the system to automatically compensate for the uneven irrigation caused by pipe pressure fluctuations.

[0047] In addition, the embodiment of the present application further provides a nonvolatile computer storage medium, which has executable instructions stored thereon, and the executable instructions, when executed, realize the crop auxiliary method based on multi-source remote sensing data.

[0048] The above description of disclosed embodiments enables one of ordinary skill in the art to make or use the application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein can be applied to other embodiments without departing from the spirit or scope of the application. Thus, the present application is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A crop auxiliary method based on multi-source remote sensing data, characterized in that: The method comprises: Determine the crop varieties and growth stages in each pre-set area based on drone, satellite imagery, and user terminals; Obtain reference evapotranspiration values ​​and soil moisture changes, and obtain the actual water requirements of crops in a preset area based on crop variety, growth stage, soil moisture changes, and reference evapotranspiration values; Based on the initial irrigation sequence and fitness function of the preset area, the optimized irrigation sequence is obtained using genetic algorithm; Based on the optimized irrigation sequence and the actual water demand of the crops, the operation of the irrigation devices in each preset area is controlled.

2. The crop auxiliary method based on multi-source remote sensing data according to claim 1, characterized in that: Based on drones, satellite images, and user terminals, determine the crop varieties and growth stages in each pre-set area, including: Upload regional crop images collected by drones to the trained deep learning recognition model to obtain the first predicted crop variety, first predicted growth stage, and prediction accuracy; Upload regional crop images collected by satellite imagery to the trained deep learning recognition model to obtain the second predicted crop variety, second predicted growth stage, and prediction accuracy; Determine the crop variety and growth stage corresponding to the highest prediction accuracy, which is the first recognition result; Determining whether the first recognition result is consistent with the crop variety and growth stage of the current region uploaded by the user terminal; When they are consistent, determining the crop variety and growth stage of the current region in the first recognition result as the final crop variety and growth stage of the current region; When there is inconsistency and the highest prediction accuracy is greater than a preset accuracy threshold, the crop variety and growth stage in the first recognition result are determined as the final crop variety and growth stage in the current area; When the highest prediction accuracy is less than or equal to a preset accuracy threshold, the crop variety and growth stage in the user terminal are determined to be the final crop variety and growth stage in the current area.

3. The crop auxiliary method based on multi-source remote sensing data according to claim 1, characterized in that: According to the crop variety, growth stage, soil moisture change and reference evapotranspiration value, the actual water requirement of crops in the preset area is obtained, including: Determine the crop coefficient corresponding to the current crop variety and growth stage from a preset mapping database between crop varieties, growth stages and crop coefficients; By formula: ET_c= (K_c×ET_0+ΔSM×γ, calculate the actual water requirement of crops; Among them, ET_c represents the actual water requirement of the crop, K_c represents the preset crop coefficient, ΔSM represents the change in soil moisture, and γ represents the correction coefficient.

4. The crop assisting method based on multi-source remote sensing data according to claim 1, characterized in that: Based on the initial irrigation sequence and fitness function of the preset area, the optimized irrigation sequence is obtained using a genetic algorithm, which includes: The initial irrigation order of the preset area is encoded into chromosomes, and each gene represents an irrigation preset area. Randomly generate N groups of irrigation sequences as the initial population; Get the fitness calculation function; The fitness calculation function and the initial population are input into the genetic algorithm, and the irrigation sequence with the highest fitness in the iterative population is output.

5. The crop auxiliary method based on multi-source remote sensing data according to claim 1, characterized in that: Based on the optimized irrigation sequence and the actual water demand of the crops, the operation of the irrigation devices in each preset area is controlled, including: Based on the irrigation sequence, determine the preset area currently being irrigated; Based on the PID algorithm, the water pump speed in the irrigation device corresponding to the preset area is dynamically controlled until the actual water demand of the crops is met.

6. A crop support system based on multi-source remote sensing data, characterized in that: The system comprises: The collection module is used to determine the crop varieties and growth stages of each preset area based on drones, satellite images and user terminals; The water acquisition module is used to obtain reference evapotranspiration values ​​and soil moisture changes, and obtain the actual water requirements of crops in a preset area based on crop varieties, growth stages, soil moisture changes, and reference evapotranspiration values; A sequence acquisition module is used to obtain an optimized irrigation sequence based on the initial irrigation sequence and fitness function of the preset area using a genetic algorithm; The control module is used to control the operation of the irrigation devices in each preset area based on the optimized irrigation sequence and the actual water demand of the crops.

7. The crop support system based on multi-source remote sensing data according to claim 6, characterized in that: The acquisition module includes an acquisition unit, It is used to upload regional crop images collected by drones to the trained deep learning recognition model to obtain the first predicted crop variety, the first predicted growth stage and the prediction accuracy; Upload regional crop images collected by satellite imagery to the trained deep learning recognition model to obtain the second predicted crop variety, second predicted growth stage, and prediction accuracy; Determine the crop variety and growth stage corresponding to the highest prediction accuracy, which is the first recognition result; Determining whether the first recognition result is consistent with the crop variety and growth stage of the current region uploaded by the user terminal; When they are consistent, determining the crop variety and growth stage of the current region in the first recognition result as the final crop variety and growth stage of the current region; When there is inconsistency and the highest prediction accuracy is greater than a preset accuracy threshold, the crop variety and growth stage in the first recognition result are determined as the final crop variety and growth stage in the current area; When the highest prediction accuracy is less than or equal to a preset accuracy threshold, the crop variety and growth stage in the user terminal are determined to be the final crop variety and growth stage in the current area.

8. The crop support system based on multi-source remote sensing data according to claim 6, characterized in that: The water acquisition module includes a water acquisition unit, Used to determine the crop coefficient corresponding to the current crop variety and growth stage from a preset mapping database between crop varieties, growth stages and crop coefficients; By formula: ET_c= (K_c×ET_0+ΔSM×γ, calculate the actual water requirement of crops; Among them, ET_c represents the actual water requirement of the crop, K_c represents the preset crop coefficient, ΔSM represents the change in soil moisture, and γ represents the correction coefficient.

9. A crop auxiliary device based on multi-source remote sensing data, characterized in that: The device comprises: processor; and a memory storing executable codes, which, when executed, enable the processor to execute the crop assistance method based on multi-source remote sensing data as claimed in any one of claims 1 to 5.

10. A non-volatile computer storage medium, characterized in that Computer instructions are stored thereon, and when the computer instructions are executed, the crop assistance method based on multi-source remote sensing data as described in any one of claims 1 to 5 is implemented.

Images