Method, device and equipment for estimating irrigation volume, storage medium and program product

Abstract

The embodiment of the invention provides a method and device for estimating irrigation volume, equipment, a storage medium and a program product. The method comprises the following steps: determining a growth period of a crop; fusing the plurality of initial soil moisture time sequences based on a space-time fusion strategy corresponding to the growth period to obtain a historical soil moisture time sequence; estimating a future moisture rising time period, calculating the moisture rising amount of the future moisture rising time period, and sending an irrigation signal indicating that the crops enter the irrigation time period when the moisture rising amount is greater than a preset moisture rising amount threshold value; in response to the irrigation signal, adjusting the corresponding soil hierarchy according to the soil hierarchy weight corresponding to the growth period and summing to obtain the soil depth of the crop; estimating the soil moisture consumption in the future moisture rising period, correcting the soil moisture consumption according to the soil depth, and outputting the irrigation amount. According to the method for estimating the irrigation volume of the crops, the soil depth factor of the crops is considered during calculation, and the irrigation volume of the crops is estimated more accurately.

Description

Technical Field

[0001] This disclosure pertains to the field of smart agriculture and precision irrigation technology, and particularly relates to a method, apparatus, equipment, storage medium, and program product for estimating irrigation volume. Background Technology

[0002] Due to limitations in the collection environment and testing equipment, it is quite difficult to accurately determine the irrigation amount of crops.

[0003] Currently, to address the aforementioned issues, deep learning-based irrigation models are commonly used to estimate irrigation amounts. However, these models suffer from high computational and energy consumption, and their black-box nature leads to opaque decision-making logic, making it difficult to gain user trust and limiting their widespread application in resource-constrained marginal agriculture scenarios. Additionally, remote sensing inversion for irrigation estimation is also frequently used, but its data is easily affected by weather conditions and satellite revisit cycles. Furthermore, it considers fewer factors during modeling, resulting in significant deviations in irrigation estimates during critical growth periods, leading to insufficient accuracy and practicality. Summary of the Invention

[0004] This disclosure provides a method, apparatus, device, storage medium, and program product for estimating irrigation amounts, which can accurately estimate crop irrigation amounts.

[0005] In a first aspect, embodiments of this disclosure provide a method for estimating crop irrigation amounts, the method comprising: Determine the growth period of the crop; Multiple initial soil moisture time series of crops are acquired using multi-source sensors. Based on a spatiotemporal fusion strategy corresponding to the growth period, multiple initial soil moisture time series are fused to obtain historical soil moisture time series. Estimate future soil moisture and periods of future moisture increase based on historical soil moisture time series; Calculate the amount of water increase during the future water increase period, and when the amount of water increase exceeds the preset water increase threshold, issue an irrigation signal for the crop to enter the irrigation period. In response to irrigation signals, the corresponding soil layers are adjusted according to the soil layer weights corresponding to the growth period and summed to obtain the soil depth of the crop. Estimate the soil moisture consumption during the future period of rising water levels, adjust the soil moisture consumption based on soil depth, and output the irrigation amount.

[0006] In one feasible implementation, the method further includes obtaining the moisture rise threshold through the following process: Determine the baseline threshold for water increase based on the reproductive period; The basic threshold for water rise is obtained by correcting the soil moisture, evaporation, and rainfall in the crop area during the water rise period.

[0007] In one feasible implementation, the method further includes: fusing multiple initial soil moisture time series based on a spatiotemporal fusion strategy corresponding to the growth period to obtain a historical soil moisture time series, including: Multiple initial soil moisture time series are time-aligned. For each sensor's initial soil moisture time series, existing time points and missing time points are determined. A first fusion variable is set based on the growth period. The mean of the existing time points is calculated using the first fusion variable to fill in the missing time points, and the time fusion result is output. The spatiotemporal fusion results of multiple time fusion results are spatially aligned. For the time fusion results of each sensor, the existing spatial segments and the missing spatial segments are determined. A second fusion variable is set based on the reproductive period. The existing spatial segments are interpolated using the second fusion variable to fill the missing spatial segments, and the spatiotemporal fusion results are output. Sensor weights are set based on the reproductive period. For each sensor, the spatiotemporal fusion result of the corresponding sensor is adjusted according to the sensor weight. The spatiotemporal fusion results of multiple spatiotemporal fusion results are summed to obtain the historical soil moisture time series.

[0008] In one feasible implementation, the method further includes: setting a first fusion variable based on the reproductive period, and calculating the mean of existing time points using the first fusion variable to fill in missing time points, including: Based on the reproductive period, a corresponding time sliding window is set, and based on the reproductive period, the contribution weights of multiple existing time points within the time sliding window to filling in the missing time points are set. Soil moisture at existing time points is adjusted to fill contribution weights, and the soil moisture at multiple existing time points is summed to obtain the soil moisture at missing time points.

[0009] In one feasible implementation, the method further includes: setting a second fusion variable based on the reproductive period, and using the second fusion variable to interpolate existing spatial segments to fill in missing spatial segments, including: Based on the reproductive period, corresponding main sensors, interpolation strategies, and auxiliary sensors are set; Interpolation calculations are performed on the temporal fusion results of the main sensor for the missing spatial segment, and the interpolation results are calibrated using the temporal fusion results of the auxiliary sensor.

[0010] In one feasible implementation, the method further includes: determining the crop's growth period, including: The normalized differential vegetation index is calculated based on the near-infrared and red light band reflectance characteristics of crops. Each sub-growth period was set based on the numerical changes, first derivative changes, second derivative changes, standard deviation changes, duration of changes, and time to reach the maximum value of the normalized differential vegetation index.

[0011] In one feasible implementation, the method further includes: setting each sub-growth period based on the numerical change, first derivative change, second derivative change, standard deviation change, duration of change, and time to reach the maximum value of the normalized differential vegetation index, including: The emergence period was set based on the numerical changes of the normalized differential vegetation index and the changes of its first derivative. The jointing period was set based on the change in the first derivative of the normalized differential vegetation index and the duration of the change; The heading period was set based on the change of the second derivative of the normalized differential vegetation index and the time when it reached the maximum value. The grouting period was set based on the first derivative change of the normalized differential vegetation index and the duration of the change; Maturity was determined based on the numerical and standard deviation changes of the normalized differential vegetation index.

[0012] In one feasible implementation, the method further includes: estimating soil moisture consumption during future periods of rising water levels, correcting the soil moisture consumption for soil depth, and outputting irrigation amounts, including: Estimate the average rainfall during the future period of rising water levels, and estimate the soil moisture consumption during the future period of rising water levels based on the area of ​​the future period of rising water levels. Soil moisture consumption is corrected based on soil depth, and the difference between the corrected soil moisture consumption and the average rainfall is used as the irrigation amount and output.

[0013] Secondly, embodiments of this disclosure provide an apparatus for estimating crop irrigation amounts, the apparatus comprising: The growth period determination module is used to determine the growth period of crops; The multi-source data fusion module is used to acquire multiple initial soil moisture time series of crops from multiple source sensors, and fuse multiple initial soil moisture time series based on a spatiotemporal fusion strategy corresponding to the growth period to obtain historical soil moisture time series. The future soil moisture estimation module is used to estimate future soil moisture and the period of future moisture increase based on historical soil moisture time series. The irrigation period identification module is used to calculate the amount of water increase during future water increase periods, and when the amount of water increase is greater than the preset water increase threshold, it sends an irrigation signal to indicate that the crop has entered the irrigation period. The soil depth calculation module is used to respond to irrigation signals by adjusting the corresponding soil layers according to the soil layer weights corresponding to the growth period and summing them to obtain the soil depth of the crop. The irrigation output module is used to estimate the soil moisture consumption during the future period of rising water levels, correct the soil moisture consumption with soil depth, and output the irrigation amount.

[0014] Thirdly, embodiments of this disclosure provide an apparatus for estimating crop irrigation amounts. The apparatus includes a processor and a memory storing computer program instructions. The processor reads and executes the computer program instructions to implement the above-described method for estimating crop irrigation amounts.

[0015] Fourthly, embodiments of this disclosure provide a computer storage medium storing computer program instructions, which, when executed by a processor, implement the aforementioned method for estimating crop irrigation amounts.

[0016] Fifthly, embodiments of this disclosure provide a computer program product, including a computer program that, when executed by a processor, implements the above-described method for estimating crop irrigation amounts.

[0017] The method, apparatus, device, storage medium, and program product for estimating irrigation volume according to embodiments of this disclosure calculate irrigation volume based on soil moisture data from multiple sources. Simultaneously, the calculation considers the soil depth factor of the crop, making the data sources more accurate and able to truly represent relevant crop variables, thus more accurately estimating crop irrigation volume. Furthermore, in specific calculations, different spatiotemporal fusion strategies are used to fuse multi-source data based on different crop growth stages. Soil depth is divided into multiple soil layers, and weights are assigned to each soil layer according to different growth stages to obtain the final crop soil depth. This enables dynamic correction of growth stages and root system stratification modeling for farmland irrigation water volume calculation, resulting in higher accuracy. In addition, when the crop type or the environment of the same crop changes, the growth stage setting can be adjusted in a timely manner, thereby adjusting the parameters for calculating irrigation volume at any time, significantly improving the accuracy and adaptability of irrigation volume estimation. Attached Figure Description

[0018] To more clearly illustrate the technical solutions of the embodiments of this disclosure, the accompanying drawings used in the embodiments of this disclosure will be briefly introduced below. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 This is a schematic flowchart of a method for estimating crop irrigation amount provided in an embodiment of this disclosure; Figure 2 This is a schematic flowchart of a spatiotemporal fusion method provided in an embodiment of this disclosure; Figure 3 This is a schematic diagram of a historical soil moisture time series curve provided in an embodiment of this disclosure; Figure 4 This is a schematic diagram of a device for estimating crop irrigation amount provided in an embodiment of this disclosure; Figure 5This is a schematic diagram of the structure of a device for estimating crop irrigation amount provided in an embodiment of this disclosure. Detailed Implementation

[0020] The features and exemplary embodiments of various aspects of this disclosure will now be described in detail. To make the objectives, technical solutions, and advantages of this disclosure clearer, the following detailed description, in conjunction with the accompanying drawings and specific embodiments, will provide a further detailed description. It should be understood that the specific embodiments described herein are intended only to explain this disclosure and not to limit it. For those skilled in the art, this disclosure can be implemented without some of these specific details. The following description of the embodiments is merely to provide a better understanding of this disclosure by illustrating examples.

[0021] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising..." does not exclude the presence of additional identical elements in the process, method, article, or apparatus that includes the element.

[0022] To achieve more efficient water-based agriculture, accurate estimation of irrigation water consumption for farmland crops is crucial. Currently, there are two main approaches to estimating agricultural irrigation volume. One approach involves building a prediction model for agricultural irrigation water consumption in irrigation districts based on deep learning algorithms. This model combines soil, meteorological, and crop growth data reported by sensors and detectors to generate predictions and obtain irrigation water consumption figures. However, this approach suffers from complex models, requiring significant computing power for training and inference. Deploying it to edge nodes can lead to high energy consumption and slow response times, making it difficult to promote in resource-constrained agricultural scenarios. Furthermore, the black-box nature of deep learning models makes it difficult for users (such as farmers and agricultural technicians) to understand the prediction logic, reducing their trust in them. For example, it cannot explain why increased irrigation is needed at a certain time, potentially leading to skepticism in practical applications. The other approach involves extracting soil moisture time series data for each pixel from remote sensing images, determining the irrigation period, and then calculating the final total irrigation amount based on rainfall. However, this scheme relies on a single remote sensing data source to obtain soil moisture time series, which is easily affected by cloud cover and insufficient spatiotemporal resolution. In addition, it does not consider the differences in irrigation requirements at different growth stages of crops (such as seedling stage, jointing stage, and heading stage), resulting in a large deviation between the estimated results and the actual water demand. At the same time, the soil depth parameter is fixed in the calculation of total irrigation volume, without taking into account the dynamic changes in crop root depth, which leads to errors in water conversion.

[0023] To address the problems of existing technologies, embodiments of this disclosure provide a method, apparatus, device, storage medium, and program product for estimating irrigation volume. The method of this disclosure first divides the crop into growth stages, then performs different fusion processes on soil moisture data acquired by various types of sensors according to different growth stages. Irrigation periods are distinguished based on the fused historical soil moisture time series, and finally, the total irrigation volume is estimated within each irrigation period. This disclosure uses multi-source data collected by multiple sensors as a foundation, combined with soil depth data from root system stratification modeling, to accurately estimate the irrigation volume required by the crop. Furthermore, in the specific calculation processes such as multi-source data fusion, irrigation period identification, and soil depth calculation, corresponding parameters and strategies are dynamically adjusted according to different growth stages, significantly improving the accuracy and adaptability of irrigation volume estimation.

[0024] The method for estimating crop irrigation amounts provided in the embodiments of this disclosure will be described below.

[0025] Figure 1 A flowchart illustrating a method for estimating crop irrigation amounts according to an embodiment of this disclosure is shown. Figure 1 As shown, the method may include the following steps: S10: Determine the growth period of the crop.

[0026] The growth period of a crop refers to the time from sowing to seed maturity or harvest of the main product, measured in days; it can also be considered as the duration of morphological changes. The length of the growth period is determined by factors such as genetic characteristics, climatic conditions, and cultivation techniques.

[0027] In one feasible implementation, this disclosure determines the crop growth stage based on the Normalized Difference Vegetation Index (NDVI). First, the NDVI is calculated based on the near-infrared and red-light reflectance characteristics of the crop. Specifically, obtaining the NDVI can be achieved by collecting raw data of the crop in the red and near-infrared bands using satellites, drones, or ground equipment, performing necessary radiometric calibration and atmospheric correction on the raw data to obtain the true surface reflectance. The surface reflectance in the red and near-infrared bands is then substituted into the NDVI calculation formula to obtain the final NDVI. Alternatively, the NDVI can be processed into an NDVI time series curve, which clearly shows the approximate growth stage of the crop. Then, sub-growth stages are set based on the numerical changes, first derivative changes, second derivative changes, standard deviation changes, duration of changes, and the time of reaching the maximum value of the NDVI. Sub-growth stages do not overlap, and multiple sub-growth stages constitute the crop's growth stage.

[0028] NDVI, based on the unique spectral reflectance characteristics of crops to visible and near-infrared wavelengths, can accurately reflect the vitality status of crops on a large spatial scale. This disclosure uses NDVI to delineate crop growth stages, providing a spatialized, quantifiable, and efficient perspective that facilitates understanding crop growth processes at both a fine and macroscopic scale.

[0029] The sub-growth stages can be set as follows: The growth stages include emergence, jointing, heading, grain filling, and maturity. Emergence is set based on the numerical and first derivative changes of the normalized difference vegetation index (NDVI); jointing is set based on the first derivative changes and duration of these changes; heading is set based on the second derivative changes and the time it reaches its maximum value; grain filling is set based on the first derivative changes and duration of these changes; and maturity is set based on the numerical and standard deviation changes of the NDVI. Alternatively, other similar methods can be used to define sub-growth stages, and the number of sub-growth stage types can be increased or decreased according to the characteristics of the crop.

[0030] In one feasible implementation, the specific reproductive period is set as follows: Seedling stage When NDVI first exceeds the threshold, the emergence period is determined according to the following formula 1: in, This represents the minimum t value under specific conditions; t represents a time point. NDVI represents the first derivative of NDVI with respect to time; NDVI(t) represents the NDVI value at time t.

[0031] Propagation period NDVI continues to rise, the rapid growth period begins, and the jointing period is determined according to the following formula 2: During the heading stage, NDVI reaches its peak. The heading stage is determined according to the following formula 3: in, This represents the time point t at which NDVI reaches its maximum value; This represents the second derivative of NDVI with time.

[0032] Grouting period NDVI begins to decrease; the grouting period is determined according to the following formula 4: Maturity NDVI is stable at low values, and the maturity period is determined according to the following formula 5: in, It represents the standard deviation within a 7-day window.

[0033] This disclosure not only considers the numerical changes of NDVI, but also utilizes other variables, such as changes in the first derivative, changes in the second derivative, changes in standard deviation, duration of change, and time to reach the maximum value, to comprehensively utilize the crop growth information contained in NDVI and achieve more accurate division and determination of the growth period.

[0034] S20: Multiple initial soil moisture time series of crops are acquired using multi-source sensors, and multiple initial soil moisture time series are fused based on a spatiotemporal fusion strategy corresponding to the growth period to obtain historical soil moisture time series.

[0035] This disclosure estimates irrigation volume during the irrigation phase. Changes in irrigation volume can be fed back to soil moisture in real time. Soil moisture is easy to collect; therefore, by collecting and analyzing soil moisture data, irrigation volume can be obtained. Thus, soil moisture time series data is used as the basis for estimating irrigation volume. However, due to system errors and specific interference effects from a single data source, the information obtained is incomplete and inaccurate. Therefore, this disclosure uses multiple sensors to collect initial soil moisture time series data for crops.

[0036] In one implementation, the multi-source sensors can be ordinary sensors used for ground measurements, sensors mounted on drones, and sensors mounted on satellites, collecting three-dimensional and comprehensive information on soil moisture from multiple dimensions and perspectives. Then, the initial soil moisture time series from multiple different sensors are fused along both temporal and spatial dimensions to obtain a historical soil moisture time series. The fused historical soil moisture time series can leverage the data advantages of each sensor, thus more closely approximating the actual changes in crop soil moisture over time.

[0037] In addition, it should be noted that when performing data fusion, this disclosure sets different spatiotemporal fusion strategies based on the reproductive period, so that the current spatiotemporal fusion strategy is closer to the current reproductive period, thereby making the fusion result more accurate.

[0038] Employing multi-source data can overcome the limitations of single sensors, enabling multiple data sources to complement, verify, and collaborate, thereby enhancing data reliability. When facing complex real-world problems (such as precision agriculture, climate change, and environmental monitoring), multi-source data can minimize uncertainty, facilitating more scientific decision-making. Furthermore, during the data fusion process, determining a more suitable spatiotemporal fusion strategy for the current stage of crop growth based on the crop's growth period, and then fusing data based on this strategy, facilitates the acquisition of more accurate historical soil moisture time series.

[0039] S30: Estimate future soil moisture and the period of future moisture increase based on historical soil moisture time series.

[0040] Since there is a correlation between past irrigation time and amount and future irrigation time and amount for the same crop, future soil moisture conditions can be estimated based on historical soil moisture time series, facilitating timely decision-making. By predicting future soil moisture, it is possible to anticipate whether there is a water deficit or sufficiency. In addition to predicting future soil moisture, since the purpose of this disclosure is to estimate irrigation amount and identify whether the crop is in the irrigation stage, this disclosure also predicts the periods during which future soil moisture will increase, i.e., the time periods during which soil moisture will maintain an upward trend.

[0041] By predicting future soil moisture and the periods when it will rise, more detailed information about future soil moisture can be obtained, facilitating subsequent refined decision-making. Furthermore, identifying irrigation stages based on future soil moisture rather than real-time soil moisture allows for timely responses and accurate prediction of irrigation end times, enabling precision irrigation.

[0042] S40: Calculate the amount of water increase during the future water increase period, and when the amount of water increase is greater than the preset water increase threshold, issue an irrigation signal for the crop to enter the irrigation period.

[0043] Because future soil moisture increases can be caused by various factors, such as rainfall, groundwater recharge, temperature decreases and humidity increases, and human-induced irrigation, it is necessary to analyze the period of future moisture increase to determine whether it is caused by human-induced irrigation. This disclosure introduces a moisture increase threshold. By comparing the moisture increase threshold with the predicted moisture increase during the future moisture increase period, it is determined whether the future moisture increase is caused by human-induced irrigation, thus confirming that irrigation has already occurred and issuing an irrigation signal for the crop to enter the irrigation period. At the start of irrigation, the irrigation amount is estimated.

[0044] It should be noted that since there may be multiple future periods of soil moisture increase, it is necessary to calculate the amount of moisture increase for each period of moisture increase and compare it with the moisture increase threshold. When the moisture increase is greater than the preset moisture increase threshold, it indicates that the irrigation process has begun.

[0045] By identifying the irrigation stage, it's possible to determine in a timely manner whether the irrigation volume estimation process should be initiated. Compared to calculating irrigation volume constantly, this step can significantly save computational resources.

[0046] In one feasible implementation, the water rise threshold is dynamically adjusted through the following process: a base threshold for water rise is determined based on the growth period; the base threshold for water rise is corrected by soil moisture, evaporation, and rainfall in the crop area during the water rise period, thus obtaining the water rise threshold.

[0047] Specifically, firstly, different baseline thresholds for water increase are set according to different growth stages. For example, during the seedling stage, when the root system is shallow and water demand is low, the baseline threshold for water increase is set to 3%; during the jointing stage, when the stem elongates rapidly and water demand increases, it is set to 5%; during the heading stage, when water demand peaks, it is set to 6%; during the grain-filling stage, when water demand remains high, it is set to 5.5%; and during the maturity stage, when physiological maturity occurs and water demand decreases sharply, it is set to 2%. Of course, other values ​​can also be set according to the characteristics of the crop. Then, the baseline threshold for water increase is adjusted using meteorological factors to obtain the threshold for water increase.

[0048] In one implementation, the moisture rise threshold can be set based on the following formula 6: in, This represents the baseline threshold for the increase in water content. Indicates the threshold for the increase in moisture content; This represents an adjustment factor, ranging from 0.1 to 0.3, with higher values ​​for arid regions and lower values ​​for humid regions. P represents the daily reference evapotranspiration; P represents the daily average precipitation.

[0049] This disclosure takes into account that the water requirements of crops vary at different growth stages, and sets a basic threshold for water increase based on this. Then, it is corrected by combining meteorological factors to make the water increase threshold more in line with the actual situation, thereby facilitating the accurate identification of irrigation stages.

[0050] S50: In response to irrigation signals, adjust the corresponding soil layers according to the soil layer weights corresponding to the growth period and sum them to obtain the soil depth of the crop.

[0051] This disclosure incorporates soil depth into irrigation volume estimation to reconstruct the actual irrigation scenario based on soil depth.

[0052] Specifically, this disclosure considers the differences in soil moisture at different soil depths. For example, the topsoil is directly affected by evaporation and precipitation, resulting in large moisture fluctuations. The middle soil layer, being the main water absorption layer for crops, exhibits higher moisture stability. Deep soil moisture migrates slowly and is less affected by root water absorption. Therefore, a soil stratification model was established, dividing the soil into multiple levels according to depth. Simultaneously, the different root and stem depths at different growth stages were considered; crops penetrate different soil layers at different stages. For example, during the seedling stage, roots are shallow, and water absorption is concentrated in the topsoil. During the jointing stage, stems elongate, and roots extend into the middle layers. During the heading stage, flowering and pollination occur, water demand peaks, and roots penetrate deeply. During the grain-filling stage, grains are filling, relying on moisture from the middle and deep layers. During the maturity stage, physiological maturity occurs, and deep moisture sustains metabolism. Therefore, it is necessary to assign weights to soil layers based on different growth stages to allow the differences in soil depth at different layers to participate in the estimation of irrigation amounts, thereby improving the accuracy of irrigation calculations.

[0053] In one implementation, soil depth is divided into three soil layers: topsoil (0-30cm), middle soil (30-60cm), and deep soil (60-100cm). The following weights are assigned to different soil layers at different growth stages: During seedling emergence, the weight for topsoil is 0.6, for middle soil 0.3, and for deep soil 0.1. During jointing, the weight is 0.3, for middle soil 0.5, and for deep soil 0.2. During heading, the weight is 0.1, for middle soil 0.7, and for deep soil 0.2. During grain filling, the weight is 0.1, for middle soil 0.6, and for deep soil 0.3. During the mature stage, the soil layer weight is 0.2 for the topsoil, 0.3 for the middle soil layer, and 0.5 for the deep soil layer. Of course, other values ​​can be set for the soil layer type, the depth range of each layer, and the soil layer weight.

[0054] This disclosure incorporates soil depth into irrigation estimation, refining soil depth into specific strata and assigning weights to each stratum based on the growth stage. This weighting is then used to adjust soil moisture, accurately recreating the irrigation scenario and obtaining precise irrigation amounts. Furthermore, different parameters can be set for different crops without changing the irrigation estimation method, thus improving the practicality and applicability of the irrigation estimation method disclosed herein.

[0055] S60: Estimate soil moisture consumption during the future period of rising water levels, adjust soil moisture consumption based on soil depth, and output irrigation amount.

[0056] Conventional irrigation volume estimation is based on the farmland water balance equation. The farmland water balance equation is: Irrigation volume = Crop water consumption - Effective rainfall - Effective soil water storage in the early stages + Water loss. However, it does not consider soil factors; soil water consumption, water storage, and crop penetration vary at different depths. Therefore, this disclosure incorporates soil depth into irrigation volume estimation, using soil depth to correct for soil water consumption, thus recreating a realistic irrigation scenario and obtaining accurate irrigation volume estimates.

[0057] In one implementation, step S60 can be achieved through the following process: First, the average rainfall during the future water rise period is estimated, and the soil moisture consumption during the future water rise period is estimated based on the area of ​​the future water rise period. Then, the soil moisture consumption is corrected based on soil depth, and the difference between the corrected soil moisture consumption and the average rainfall is used as the irrigation amount and output.

[0058] Specifically, the irrigation amount can be estimated using the following formula 7: Where n represents the total number of irrigation periods; i represents the current time point; m represents the number of pixels; j represents the current pixel; P i (j) h represents the average rainfall during irrigation period i; k ω represents the soil depth in the k-th soil layer; k This represents the soil depth weight corresponding to the k-th soil layer; This represents the estimated irrigation amount; soil moisture at time point i and pixel j is represented as... ; This represents the amount of soil moisture consumed at time point i and pixel j.

[0059] In one implementation, step S20 can be achieved as follows. Figure 2 A schematic flowchart of a spatiotemporal fusion method provided in an embodiment of this disclosure is shown. Figure 2 As shown, the spatiotemporal fusion method includes steps S21 to S23, the specific contents of which are as follows.

[0060] S21. Align multiple initial soil moisture time series. For each sensor's initial soil moisture time series, determine the existing time points and missing time points. Set a first fusion variable based on the growth period. Calculate the mean of the existing time points using the first fusion variable to fill in the missing time points. Output the time fusion result.

[0061] After aligning the initial soil moisture time series obtained from multiple sensors, some time points may be missing. These time points correspond to initial soil moisture values ​​that are either nonexistent or abnormal. Therefore, they need to be corrected and supplemented based on the first fusion variable. Specifically, the mean of the initial soil moisture values ​​corresponding to the existing time points near the missing time point is calculated, and the resulting mean value is used as the initial soil moisture value corresponding to the missing time point.

[0062] In one implementation, a sliding window mean imputation method is used. For each missing time point t, the mean soil moisture of all available data (i.e., existing time points) within the sliding window is taken, thereby calculating the mean of existing time points to imput the missing time point. First, a first fusion variable is set based on the growth period. The first fusion variable includes the time sliding window and the contribution weights of multiple existing time points within the time sliding window to imputing the missing time point. Then, the soil moisture of existing time points is adjusted according to the imputation contribution weights, and the soil moisture of multiple existing time points is summed to obtain the soil moisture of the missing time point.

[0063] In one implementation, the soil moisture value at the missing time point can be calculated according to the following formula 8: in, The time range of the time sliding window is dynamically adjusted according to the reproductive period; N represents the window size; N is the amount of valid data within the time sliding window. For example, if the current time sliding window is 10 days, but there is no data for 3 days, then N=7. In this case, there are 3 missing time points and 7 existing time points. This represents the soil moisture value at the k-th time point within the sliding window; This represents the weight coefficient at the k-th time point, i.e., the contribution weight of the data at that time point to the imputed value; This represents the soil moisture value at the missing time point t.

[0064] The first fusion variable can be set as follows: The window size is 5 days for the seedling emergence stage; 3 days for the jointing stage; 2 days for the heading stage; 2 days for the grain filling stage; and 5 days for the maturity stage. Regarding the fill contribution weight, all fill contribution weights are the same for the seedling emergence, jointing, and heading stages, set to 1. During the grain filling stage, recent data has a more significant weight; for example, the weight for the first two days is 2, and the weight for the last two days is 1. During the maturity stage, only historical data is used for fill, and future data has a weight of 0.

[0065] This disclosure sets different time sliding window sizes for different growth stages. During the jointing, heading, or grain-filling stages, when soil moisture changes rapidly, it's necessary to retain data details and avoid over-smoothing; therefore, smaller windows are used. Conversely, during the seedling and maturity stages, when soil moisture changes more slowly, larger windows are used to better average out random noise and obtain stable infill values. Furthermore, weights are assigned to multiple existing time points within a single time sliding window based on the growth stage. This takes into account the subtle differences in trends at different time points within each growth stage of the crop, thus assigning higher infill contribution weights to existing time points with a stronger impact from missing time points. Setting specific time sliding window sizes and infill contribution weights according to the growth stage facilitates refined control of the time fusion process and achieves precise time fusion.

[0066] S22. Spatially align multiple temporal fusion results. For the temporal fusion results of each sensor, determine the existing spatial segments and missing spatial segments. Set a second fusion variable based on the reproductive period. Use the second fusion variable to interpolate the existing spatial segments to fill in the missing spatial segments and output the spatiotemporal fusion results.

[0067] After temporal fusion of initial soil moisture time series obtained from multiple sensors, spatial fusion is also required. This necessitates spatial alignment of the multiple temporal fusion results. At this stage, some spatial segments may exist, meaning the initial soil moisture values ​​for these time periods are either nonexistent or abnormal. Therefore, correction and supplementation based on a second fusion variable are necessary. Specifically, interpolation is performed using the initial soil moisture values ​​corresponding to existing spatial segments near the missing segment, and the interpolated value is used as the initial soil moisture value corresponding to the missing segment.

[0068] In one implementation, different interpolation methods are used to fill in missing spatial segments based on the reproductive period. First, a second fusion variable is set based on the reproductive period. This second fusion variable consists of the main sensor, the interpolation strategy, and the auxiliary sensor. The interpolation strategy can be inverse distance weighting (IDW) or weighted summation interpolation. Then, interpolation is performed on the temporal fusion results of the main sensor for the missing spatial segments, and the interpolation results are calibrated using the temporal fusion results of the auxiliary sensor.

[0069] Specifically, this can be achieved through the following process.

[0070] During the seedling emergence stage, a standard sensor used for ground monitoring is used as the primary sensor, and the IDW (In-Depth Wave) interpolation strategy is adopted, without setting up auxiliary sensors. Interpolation calculations are performed using the following formula (Formula 9): Where, d i Let (x, y) be the Euclidean distance from the target point (x, y) to the i-th ordinary sensor; The soil moisture value after spatial fusion at the target point (x,y); This represents the soil moisture value of the i-th ordinary sensor.

[0071] During the growth stage, sensors mounted on satellites and drones are used as the primary sensors for data resolution fusion. A weighted summation interpolation strategy is employed, with ordinary sensors used for ground monitoring serving as secondary sensors for calibration. Interpolation calculations are performed using the following formula 10: in, This represents the calibration value of a common sensor; This represents the satellite soil moisture value before space fusion; This indicates the soil moisture value of the drone before spatial fusion; This represents the soil moisture value of the i-th ordinary sensor before spatial fusion; This represents the soil moisture value after spatial fusion at the target point (x,y).

[0072] During the heading stage, sensors mounted on satellites and drones are used as the primary sensors for data resolution fusion. A weighted summation interpolation strategy is adopted, without auxiliary sensors. Interpolation calculations are performed using the following formula 11: in, Indicates the coverage area of ​​the drone. This indicates the area of ​​a satellite pixel.

[0073] During the grouting period, sensors mounted on high-frequency UAVs are used as primary sensors, and sensors mounted on satellites are used as secondary sensors to fill blind spots. A weighted summation interpolation strategy is adopted. Interpolation calculations are performed using the following formula 12: in, This represents the soil moisture value after averaging the soil moisture values ​​from ordinary sensors before spatial fusion.

[0074] During the mature stage, satellite-borne sensors are used as the primary sensors, while ordinary sensors used for ground monitoring are used as secondary sensors to verify moisture residue. A weighted summation interpolation strategy is employed. Interpolation calculations are performed using the following formula 13: in, ; This represents the soil moisture value after attenuation of the soil moisture value from a regular sensor before spatial fusion.

[0075] This disclosure employs multiple interpolation strategies during spatial fusion and sets up main and auxiliary sensors according to different crop growth stages, fully considering the changes in the spatial state of crops at different stages, and thus the variations in sensor measurement results. For example, during the seedling stage, the crop height is very short. At this time, sensors mounted on satellites or UAVs cannot obtain accurate soil moisture values ​​due to the long distance, nor can they provide better calibration results. Therefore, ordinary ground-based sensors are used as the main sensors, and no auxiliary sensors are set up. At the same time, to compensate for the data deficiencies of single sensors, high-precision IDW (Inverse Distance Weighted Interpolation) is used for interpolation calculation. During the maturity stage, inverse distance weighted interpolation or weighted summation interpolation can be used. During the grain-filling stage, continuous rainy weather makes satellite sensors susceptible to the influence of clouds and atmospheric conditions, resulting in inferior measurement results compared to UAV sensors. Ordinary sensors have very low crop growth detection efficiency. Therefore, UAV sensors are used as the main sensors, and satellite sensors are used to fill the blind spots of UAVs. With the support of these high-precision data, only a weighted summation interpolation strategy needs to be set up to simplify the control process.

[0076] S23. Based on the growth period, set sensor weights, adjust the spatiotemporal fusion results of each sensor according to the sensor weight, and sum the multiple spatiotemporal fusion results to obtain the historical soil moisture time series. Figure 3 A schematic diagram of a historical soil moisture time series curve provided for the implementation of this disclosure. For example... Figure 3 As shown, the horizontal axis represents time, and the vertical axis represents soil moisture.

[0077] In one implementation, the historical soil moisture time series is obtained using the following formula 14: in, Indicates the sensor weights of the satellite sensors; Indicates the sensor weights of the drone's sensors; This represents the sensor weights of common sensors used for ground measurements; This represents the satellite soil moisture value after spatiotemporal fusion; This represents the soil moisture value of the drone after spatiotemporal fusion; This represents the soil moisture value from a standard sensor after spatiotemporal fusion. This represents a historical soil moisture time series.

[0078] The sensor weights for satellites, drones, and conventional sensors can be set as follows: During the seedling stage, the sensor weights for satellites, drones, and conventional sensors are 0, 0.3, and 0.7, respectively; during the jointing stage, the weights are 0.4, 0.4, and 0.2, respectively; during the heading stage, the weights are 0.5, 0.4, and 0.1, respectively; during the grain-filling stage, the weights are 0.3, 0.6, and 0.1, respectively; and during the maturity stage, the weights are 0.8, 0, and 0.2, respectively.

[0079] As can be seen from the above, spatiotemporal fusion based on the growth period makes it easy to finely set the specific fusion process according to the crop's growth status and surrounding environment. This makes the entire spatiotemporal fusion process highly controllable and flexible. For different crops or different stages of the same crop, targeted spatiotemporal fusion can be achieved simply by changing the parameters, making it highly applicable.

[0080] In summary, this disclosure constructs a soil moisture time series based on the fusion of multi-source data during the crop's growth period, effectively addressing the differences in irrigation needs across different growth stages and the susceptibility of single data sources to interference. Furthermore, a dynamic threshold adjustment mechanism based on the growth period is proposed for irrigation period identification, resulting in more accurate identification results. Additionally, the inclusion of soil depth stratification and growth stage-specific data in irrigation calculation accurately characterizes the dynamic changes in crop root depth, leading to more precise estimation results and significantly improving the accuracy and adaptability of irrigation quantity estimation.

[0081] Figure 4 This is a schematic diagram of a device for estimating crop irrigation amounts provided in an embodiment of this disclosure. Figure 4 As shown, the device may include a growth period determination module 210, a multi-source data fusion module 220, a future soil moisture estimation module 230, an irrigation period identification module 240, a soil depth calculation module 250, and an irrigation output module 260.

[0082] The growth period determination module 210 is used to determine the growth period of crops; The multi-source data fusion module 220 is used to acquire multiple initial soil moisture time series of crops from multiple source sensors, and fuse multiple initial soil moisture time series based on a spatiotemporal fusion strategy corresponding to the growth period to obtain historical soil moisture time series. The future soil moisture estimation module 230 is used to estimate future soil moisture and future periods of moisture increase based on historical soil moisture time series. The irrigation period identification module 240 is used to calculate the amount of water increase during the future water increase period, and when the amount of water increase is greater than the preset water increase threshold, it sends an irrigation signal that the crop has entered the irrigation period. The soil depth calculation module 250 is used to respond to irrigation signals, adjust the corresponding soil layers according to the soil layer weights corresponding to the growth period, and sum them to obtain the soil depth of the crop. The irrigation output module 260 is used to estimate the soil moisture consumption during the future period of rising water levels, correct the soil moisture consumption with soil depth, and output the irrigation amount.

[0083] In one embodiment, the irrigation period identification module 240 is further configured to: Determine the baseline threshold for water increase based on the reproductive period; The basic threshold for water rise is obtained by correcting the soil moisture, evaporation, and rainfall in the crop area during the water rise period.

[0084] In one embodiment, the multi-source data fusion module 220 is further configured to: Multiple initial soil moisture time series are time-aligned. For each sensor's initial soil moisture time series, existing time points and missing time points are determined. A first fusion variable is set based on the growth period. The mean of the existing time points is calculated using the first fusion variable to fill in the missing time points, and the time fusion result is output. The spatiotemporal fusion results of multiple time fusion results are spatially aligned. For the time fusion results of each sensor, the existing spatial segments and the missing spatial segments are determined. A second fusion variable is set based on the reproductive period. The existing spatial segments are interpolated using the second fusion variable to fill the missing spatial segments, and the spatiotemporal fusion results are output. Sensor weights are set based on the reproductive period. For each sensor, the spatiotemporal fusion result of the corresponding sensor is adjusted according to the sensor weight. The spatiotemporal fusion results of multiple spatiotemporal fusion results are summed to obtain the historical soil moisture time series.

[0085] In one embodiment, the multi-source data fusion module 220 is further configured to: Based on the reproductive period, a corresponding time sliding window is set, and based on the reproductive period, the contribution weights of multiple existing time points within the time sliding window to filling in the missing time points are set. Soil moisture at existing time points is adjusted to fill contribution weights, and the soil moisture at multiple existing time points is summed to obtain the soil moisture at missing time points.

[0086] In one embodiment, the multi-source data fusion module 220 is further configured to: Based on the reproductive period, corresponding main sensors, interpolation strategies, and auxiliary sensors are set; Interpolation calculations are performed on the temporal fusion results of the main sensor for the missing spatial segment, and the interpolation results are calibrated using the temporal fusion results of the auxiliary sensor.

[0087] In one implementation, the reproductive period determination module 210 is further configured to: The normalized differential vegetation index is calculated based on the near-infrared and red light band reflectance characteristics of crops. Each sub-growth period was set based on the numerical changes, first derivative changes, second derivative changes, standard deviation changes, duration of changes, and time to reach the maximum value of the normalized differential vegetation index.

[0088] In one implementation, the reproductive period determination module 210 is further configured to: The emergence period was set based on the numerical changes of the normalized differential vegetation index and the changes of its first derivative. The jointing period was set based on the change in the first derivative of the normalized differential vegetation index and the duration of the change; The heading period was set based on the change of the second derivative of the normalized differential vegetation index and the time when it reached the maximum value. The grouting period was set based on the first derivative change of the normalized differential vegetation index and the duration of the change; Maturity was determined based on the numerical and standard deviation changes of the normalized differential vegetation index.

[0089] In one embodiment, the irrigation output module 260 is further configured to: Estimate the average rainfall during the future period of rising water levels, and estimate the soil moisture consumption during the future period of rising water levels based on the area of ​​the future period of rising water levels. Soil moisture consumption is corrected based on soil depth, and the difference between the corrected soil moisture consumption and the average rainfall is used as the irrigation amount and output.

[0090] Figure 5 A schematic diagram of the hardware structure of a device for estimating crop irrigation amounts provided in an embodiment of this disclosure is shown.

[0091] The device for estimating crop irrigation amounts may include a processor 301 and a memory 302 storing computer program instructions.

[0092] Specifically, the processor 301 may include a central processing unit (CPU), an application specific integrated circuit (ASIC), or one or more integrated circuits that can be configured to implement the embodiments of this disclosure.

[0093] Memory 302 may include mass storage for data or instructions. For example, and not limitingly, memory 302 may include a hard disk drive (HDD), floppy disk drive, flash memory, optical disk, magneto-optical disk, magnetic tape, or Universal Serial Bus (USB) drive, or a combination of two or more of these. In one instance, memory 302 may include removable or non-removable (or fixed) media, or memory 302 may be non-volatile solid-state memory. Memory 302 may be internal or external to the integrated gateway disaster recovery device.

[0094] In one instance, memory 302 may be read-only memory (ROM). In one instance, the ROM may be a mask-programmed ROM, a programmable ROM (PROM), an erasable PROM (EPROM), an electrically erasable PROM (EEPROM), an electrically rewritable ROM (EAROM), or flash memory, or a combination of two or more of these.

[0095] Memory 302 may include read-only memory (ROM), random access memory (RAM), disk storage media device, optical storage media device, flash memory device, electrical, optical, or other physical / tangible memory storage device. Therefore, generally, memory includes one or more tangible (non-transitory) computer-readable storage media (e.g., memory devices) encoded with software including computer-executable instructions, and when the software is executed (e.g., by one or more processors), it is operable to perform the operations described with reference to the method according to one aspect of this disclosure.

[0096] The processor 301 reads and executes computer program instructions stored in the memory 302 to achieve... Figures 1 to 3 The method for estimating crop irrigation amounts in the illustrated embodiment.

[0097] In one example, the device for estimating crop irrigation amounts may also include a communication interface 303 and a bus 304. Wherein, for example... Figure 5 As shown, the processor 301, memory 302, and communication interface 303 are connected through bus 304 and complete communication with each other.

[0098] The communication interface 303 is mainly used to realize communication between various modules, devices, units and / or equipment in the embodiments of this disclosure.

[0099] Bus 304 includes hardware, software, or both, that couples components of an online data traffic metering device together. For example, and not as a limitation, the bus may include an Accelerated Graphics Port (AGP) or other graphics bus, an Extended Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), a Hyper Transport (HT) interconnect, an Industry Standard Architecture (ISA) bus, an Infinite Bandwidth Interconnect, a Low Pin Count (LPC) bus, a memory bus, a Microchannel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a Video Electronics Standards Association Local (VLB) bus, or other suitable buses, or combinations of two or more of these. Where appropriate, bus 304 may include one or more buses. Although specific buses are described and illustrated in embodiments of this disclosure, this disclosure contemplates any suitable bus or interconnect.

[0100] Furthermore, in conjunction with the methods for estimating crop irrigation amounts in the above embodiments, this disclosure can provide a computer storage medium for implementation. The computer storage medium stores computer program instructions; when these computer program instructions are executed by a processor, they implement any of the methods for estimating crop irrigation amounts in the above embodiments.

[0101] This application also provides a computer program product, including a computer program that, when executed by a processor, implements any of the methods for estimating crop irrigation amounts described in the above embodiments.

[0102] It should be clarified that this disclosure is not limited to the specific configurations and processes described above and shown in the figures. For the sake of brevity, detailed descriptions of known methods are omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of this disclosure is not limited to the specific steps described and shown, and those skilled in the art can make various changes, modifications, and additions, or change the order of steps, after understanding the spirit of this disclosure.

[0103] The functional blocks shown in the above block diagram can be implemented as hardware, software, firmware, or a combination thereof. When implemented in hardware, they can be, for example, electronic circuits, application-specific integrated circuits (ASICs), appropriate firmware, plug-ins, function cards, etc. When implemented in software, the elements of this disclosure are programs or code segments used to perform the required tasks. Programs or code segments can be stored on a machine-readable medium or transmitted over a transmission medium or communication link via data signals carried on a carrier wave. "Machine-readable medium" can include any medium capable of storing or transmitting information. Examples of machine-readable media include electronic circuits, semiconductor memory devices, read-only memory (ROM), flash memory, erasable read-only memory (EROM), floppy disks, compact disc read-only memory (CD-ROM), optical disks, hard disks, fiber optic media, radio frequency (RF) links, etc. Code segments can be downloaded via computer networks such as the Internet, intranets, etc.

[0104] It should also be noted that the exemplary embodiments mentioned in this disclosure describe methods or systems based on a series of steps or apparatus. However, this disclosure is not limited to the order of the above steps; that is, the steps can be performed in the order mentioned in the embodiments, or in a different order, or several steps can be performed simultaneously.

[0105] The aspects of this disclosure have been described above with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this disclosure. It should be understood that each block in the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus to produce a machine such that these instructions, executable via the processor of the computer or other programmable data processing apparatus, enable the implementation of the functions / actions specified in one or more blocks of the flowchart illustrations and / or block diagrams. Such a processor can be, but is not limited to, a general-purpose processor, a special-purpose processor, a special application processor, or a field-programmable logic circuit. It is also understood that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can also be implemented by special-purpose hardware performing the specified functions or actions, or can be implemented by a combination of special-purpose hardware and computer instructions.

[0106] The above are merely specific embodiments of this disclosure. Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, modules, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here. It should be understood that the protection scope of this disclosure is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in this disclosure, and these modifications or substitutions should all be covered within the protection scope of this disclosure.

Claims

1. A method for estimating crop irrigation amounts, characterized in that, include: Determine the growth period of the crop; Multiple initial soil moisture time series of the crop are acquired using multi-source sensors, and the multiple initial soil moisture time series are fused based on a spatiotemporal fusion strategy corresponding to the growth period to obtain a historical soil moisture time series. Estimate future soil moisture and periods of future moisture increase based on the historical soil moisture time series; Calculate the amount of water increase during the future water increase period, and when the amount of water increase is greater than a preset water increase threshold, issue an irrigation signal for the crop to enter the irrigation period. In response to the irrigation signal, the corresponding soil layers are adjusted according to the soil layer weights corresponding to the growth period and summed to obtain the soil depth of the crop; Estimate the soil moisture consumption during the future period of rising water levels, correct the soil moisture consumption based on the soil depth, and output the irrigation amount.

2. The method according to claim 1, characterized in that, The threshold for moisture increase is obtained through the following process: Determine the basic threshold for water increase based on the aforementioned reproductive period; The basic threshold for the increase in water content is obtained by correcting the soil moisture, evaporation, and rainfall in the crop area during the period of water increase.

3. The method according to claim 1, characterized in that, The method of fusing multiple initial soil moisture time series based on a spatiotemporal fusion strategy corresponding to the growth period to obtain a historical soil moisture time series includes: The initial soil moisture time series are aligned in time. For each sensor's initial soil moisture time series, the existing time points and missing time points are determined. A first fusion variable is set based on the growth period. The mean of the existing time points is calculated using the first fusion variable to fill in the missing time points, and the time fusion result is output. The multiple time fusion results are spatially aligned. For the time fusion result of each sensor, the existing spatial segment and the missing spatial segment are determined. A second fusion variable is set based on the reproductive period. The existing spatial segment is interpolated using the second fusion variable to fill the missing spatial segment. The spatiotemporal fusion result is then output. Based on the growth period, sensor weights are set, and for each sensor, the spatiotemporal fusion result of the corresponding sensor is adjusted according to the sensor weight. The spatiotemporal fusion results of multiple sensors are summed to obtain the historical soil moisture time series.

4. The method according to claim 3, characterized in that, The step of setting a first fusion variable based on the reproductive period, and using the first fusion variable to calculate the mean of the existing time points to fill in the missing time points, includes: Based on the reproductive period, a corresponding time sliding window is set, and based on the reproductive period, multiple existing time points within the time sliding window are set to contribute weights to filling the missing time points. The soil moisture at the existing time points is adjusted according to the filling contribution weight, and the soil moisture at multiple existing time points is summed to obtain the soil moisture at the missing time points.

5. The method according to claim 3, characterized in that, The step of setting a second fusion variable based on the reproductive period, and using the second fusion variable to interpolate the existing spatial segments to fill in the missing spatial segments, includes: Based on the reproductive period, corresponding main sensors, interpolation strategies, and auxiliary sensors are set; Interpolation calculations are performed on the temporal fusion results of the main sensor for the missing spatial segment, and the interpolation calculation results are calibrated using the temporal fusion results of the auxiliary sensor.

6. The method according to claim 1, characterized in that, Determining the growth period of the crop includes: The normalized differential vegetation index was calculated based on the near-infrared and red light band reflectance characteristics of the crop. Each sub-growth period is set based on the numerical changes, first derivative changes, second derivative changes, standard deviation changes, duration of changes, and time to reach the maximum value of the normalized differential vegetation index.

7. The method according to claim 6, characterized in that, Each sub-growth period is set based on the numerical changes, first derivative changes, second derivative changes, standard deviation changes, duration of changes, and time to reach the maximum value of the normalized differential vegetation index, including: The emergence period is set based on the numerical changes and first derivative changes of the normalized differential vegetation index. The jointing period is set based on the change of the first derivative of the normalized differential vegetation index and the duration of the change. The heading period is set based on the change of the second derivative of the normalized differential vegetation index and the time when it reaches its maximum value. The grouting period is set based on the change of the first derivative of the normalized differential vegetation index and the duration of the change. The maturity period is set based on the numerical changes and standard deviation changes of the normalized difference vegetation index.

8. The method according to claim 1, characterized in that, The process of estimating soil moisture consumption during the future period of rising water levels, correcting the soil moisture consumption based on the soil depth, and outputting the irrigation amount includes: Estimate the average rainfall during the future water rise period, and estimate the soil moisture consumption during the future water rise period based on the area of ​​the future water rise period; The soil moisture consumption is corrected based on the soil depth, and the difference between the corrected soil moisture consumption and the average rainfall is used as the irrigation amount and output.

9. A device for estimating crop irrigation amounts, characterized in that, The device includes: A growth period determination module is used to determine the growth period of the crop; A multi-source data fusion module is used to acquire multiple initial soil moisture time series of the crop using multi-source sensors, and fuse the multiple initial soil moisture time series based on a spatiotemporal fusion strategy corresponding to the growth period to obtain a historical soil moisture time series. The future soil moisture estimation module is used to estimate future soil moisture and future periods of moisture increase based on the historical soil moisture time series. The irrigation period identification module is used to calculate the amount of water increase during the future water increase period, and when the amount of water increase is greater than a preset water increase threshold, it sends an irrigation signal that the crop has entered the irrigation period. A soil depth calculation module is used to respond to the irrigation signal by adjusting the corresponding soil layer according to the soil layer weight corresponding to the growth period and summing the results to obtain the soil depth of the crop. The irrigation output module is used to estimate the soil moisture consumption during the future water rise period, correct the soil moisture consumption with the soil depth, and output the irrigation amount.

10. A device for estimating crop irrigation amounts, characterized in that, The device includes: a processor and a memory storing computer program instructions; the processor reads and executes the computer program instructions to implement the method for estimating crop irrigation as described in any one of claims 1-8.

11. A computer storage medium, characterized in that, The computer storage medium stores computer program instructions, which, when executed by a processor, implement the method for estimating crop irrigation amounts as described in any one of claims 1-8.

12. A computer program product, characterized in that, Includes a computer program that, when executed by a processor, implements the method for estimating crop irrigation amounts as described in any one of claims 1-8.

Images