Method and system for providing site-specific fertilizer recommendations

Abstract

A system and method for providing fertilizer recommendations to farmland based on the determination of crop nutrient status, wherein baseline values ​​and field variability are determined for said crop nutrient status. The fertilizer recommendations are then adjusted based on the calculated baseline and field variability of said crop nutrient status.

Description

Technical Field

[0001] This disclosure relates to a system and method for providing location-specific fertilizer recommendations to crops in farmland. Background Technology

[0002] Determining the appropriate amount of fertilizer required for a crop is one of the most important decisions farmers will face. Nitrogen deficiency will reduce yields, while excess will result in economic losses and environmental damage. Deficiencies in other nutrients will lead to crop defects and reduced crop quality. However, crops exhibit field variability due to differences in parameters affecting crop growth and development (water, soil properties), making it challenging to establish site-specific fertilizer recommendations that may be optimal for the entire crop.

[0003] Among the various methods used to estimate crop nitrogen levels, remote imagery generated by satellites or other equivalent drones is becoming increasingly important in practice due to the availability of readily available solutions from different providers. Remote sensing can determine crop nitrogen levels in remote fields without requiring field inspections. However, using remote imagery introduces well-known drawbacks such as noise, weather, and atmospheric conditions. Furthermore, remote methods cannot assess other parameters that depend on soil variation characteristics. Therefore, an integrated remote sensing-based solution is intended to further account for field variations.

[0004] Existing technology

[0005] Known methods for reducing noise in remote sensing for agricultural purposes aim to analyze the received signal using an iterative model fitted with a support vector machine, along with noise identification and masking methods. For example, US2021027429 A1 discloses one such method. However, these methods require significant computational effort and time.

[0006] Regardless of the objective of calculating vegetation indices, the smoothing techniques used for denoising in US2021201024 A1 are well-known. Previously, to identify crops based on NDVI processing, vegetation indices were obtained through Gaussian smoothing and model denoising. This required calibrating multiple model parameters and filtering out some dynamics, but could again result in the loss of some information from the original processed indices at the cost of high computational cost. Summary of the Invention

[0007] The purpose of this disclosure is to establish a method for providing variable fertilizer recommendations based on deterministic methods, which overcomes the aforementioned problems while taking into account the field variability of fertilizer recommendations.

[0008] According to a first aspect of this disclosure, the objective and other objectives are achieved by a computer-implemented method for providing variable fertilizer recommendations for crops, the method comprising the steps of: identifying at least one farmland including at least one crop; determining the crop nutrient status for at least one crop in the at least one farmland, wherein determining the crop nutrient status for at least one crop in the at least one farmland comprises: determining a baseline value for the crop nutrient status of the at least one farmland; determining field variability of the crop nutrient status of the at least one farmland; determining the crop nutrient status based on the baseline value and based on the field variability; and determining a variable fertilizer recommendation for the at least one farmland based on the determined crop nutrient status.

[0009] This method enables precise fertilizer recommendations that provide a true representation of absolute nitrogen uptake and field variability.

[0010] According to another embodiment, determining a baseline value for crop nutrient status includes: receiving remote data, which includes image data of at least one farmland; generating at least one first vegetation index indicating crop nutrient status based on the image data; and determining the baseline value based on the vegetation index.

[0011] This method calculates baseline values ​​using remote image data.

[0012] According to another embodiment, determining the field variability of crop nutrient status includes: receiving remote data, the remote data including image data indicating the current crop nutrient status of at least one field; generating at least one second vegetation index based on the image data; and determining the current field variability of variable fertilizer recommendations based on the vegetation index.

[0013] This method is used to calculate field variability terms using remote image data.

[0014] According to another embodiment, at least one vegetation index used to determine the baseline value is defined as SX = f(RXXX(1), RXXX(2), ..., RXXX(i), ..., RXXX(n)), where RXXX(i) for i = 1, ..., n represents multiple reflectance data related to a given wavelength included in the remote data, and f is a predetermined mathematical relationship.

[0015] Using this method, the baseline values ​​are obtained based on appropriate vegetation indices.

[0016] According to another embodiment, generating at least one vegetation index based on image data to determine field variability further includes determining at least one vegetation index that is different from at least one vegetation index used to determine a baseline value for a variable fertilizer recommendation.

[0017] Using this method, the inherent instrument noise of the first vegetation index can be avoided when generating field variability terms.

[0018] According to another embodiment, the at least one vegetation index used to determine field variability is different from the at least one vegetation index used to determine the baseline value. This further includes defining the at least one vegetation index used to determine field variability as SX′=f′(RXXX(1), RXXX(2), ...RXXX(i′), ..., RXXX(n)), where f′ is the same mathematical relation as that present in the at least one vegetation index used to determine the baseline value, wherein at least one RXXX(i′) included in the at least one vegetation index used to determine field variability replaces at least one of the reflectance data included in the at least one vegetation index used to determine the baseline value, and at least one RXXX(i′) represents reflectance data related to a wavelength different from the replaced at least one reflectance data.

[0019] Using this method, a pair of suitable vegetation indices can be generated to improve the determination of crop nutrient status.

[0020] According to another embodiment, RXXX(i′) is a reflectance value selected in the bands associated with green, red, vegetation red edge and near-infrared bands.

[0021] Using this method, the wavelengths selected for other vegetation indices belong to the spectrum that indicates agricultural characteristics.

[0022] According to another embodiment, RXXX(i′) is selected by detecting at least one region near the identified farmland that has a spectral flatness value independent of a given wavelength, and selecting at least one band from the green, red, vegetation red edge and near-infrared bands that represents the lowest standard deviation of the detected at least one region.

[0023] Using this method, the selected channels have a lower standard deviation, reducing noise in other vegetation indices.

[0024] According to another embodiment, the variable fertilizer recommendations are further adjusted based on field data or local analysis of soil nutrient content, soil organic matter and / or mineralization data.

[0025] This approach allows for the use of local data and other agronomic parameters to improve remote fertilizer recommendations.

[0026] According to another embodiment, the method further includes generating a machine-readable script file for agricultural equipment based on determined variable fertilizer recommendations for fertilizer application.

[0027] Using this method, agricultural equipment can be automatically configured to make the fertilizer recommendations of this disclosure.

[0028] According to another embodiment, the method further includes implementing variable fertilizer recommendations via agricultural equipment configured to apply fertilizer.

[0029] Using this method, fertilizer recommendations can be given directly.

[0030] According to another aspect, systems, data processing apparatuses, computer-readable storage media, and computer program products configured to perform the methods discussed above are contemplated within this disclosure. Attached Figure Description

[0031] The accompanying drawings, which are included to provide a further understanding of this disclosure and are incorporated in and constitute a part of this specification, illustrate embodiments of this disclosure and, together with the description, serve to explain the principles of this disclosure.

[0032] Figure 1 The field of application of this disclosure is shown.

[0033] Figure 2 A schematic representation of a system according to an embodiment of the present disclosure is shown.

[0034] Figure 3 The vegetation index SX (top row) and the combination of SX and SX' (bottom row) are shown for the designated fields at different crop stages in spring.

[0035] Figure 4 The vegetation index SX (top row) and the combination of SX and SX' (bottom row) are shown for the designated fields at different crop stages in autumn.

[0036] Figure 5 A flowchart of a method according to a main embodiment of the present disclosure is shown.

[0037] The accompanying drawings are provided to aid in the easy understanding of the technical concepts of this disclosure, and it should be understood that the concepts of this disclosure are not limited to the drawings. The concepts of this disclosure should be interpreted as extending to any modifications, equivalents, and substitutions other than those shown in the drawings. Several embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. Detailed Implementation

[0038] As used herein, the singular forms “a,” “an,” and “the” include both the singular and plural unless the context clearly indicates otherwise. The terms “comprise,” “comprises,” and “including,” “include,” “contain,” and “contains” as used below are synonymous and include or exclude other unmentioned portions, elements, or method steps. When this specification refers to a product or process that “includes” a particular feature, portion, or step, it means that other features, portions, or steps may also be present, but it may also refer to an embodiment that includes only the listed features, portions, or steps.

[0039] The numerical values ​​listed with the aid of the accompanying drawings include all values ​​and fractions within these ranges, as well as the referenced endpoint values. As used, the term "approximately" means, when referring to a measurable value such as a parameter, quantity, time period, etc., that includes a specified value and a variation of + / - 10% or less, preferably + / - 5% or less, more preferably + / - 1% or less, and even more preferably + / - 0.1% or less, provided that such variations apply to the disclosure herein. It should be understood that the values ​​referred to by the term "approximately" themselves have also been disclosed.

[0040] Unless otherwise defined, all terms appearing in this disclosure, including technical and scientific terms, have the meanings that are commonly given to them by those skilled in the art. For further guidance, definitions are included to further explain the terms used in the description of this disclosure.

[0041] Figure 1 The text describes farmland including crops within an agricultural zone, as well as other systems and equipment with which system 100 can interoperate. Figure 2 The image shows an example of a system 100 according to the present disclosure. According to the present disclosure, system 100 includes multiple components such as a memory unit 110, a processor 120, a wired / wireless communication unit 130, and an input / output unit 140. System 100 can also be operatively connected to a personal or mobile device 200 via the communication unit 130.

[0042] System 100 includes an agricultural advice engine 220, which, if remote in nature, can be remotely connected to via communication unit 130. In this case, the agricultural advice engine 220 can be represented by a computer, a remotely accessible server, other client-server architecture, or any other electronic device typically included under the term data processing equipment. System 100 does not need to be located near the farmland where advice should be given.

[0043] System 100 may also be represented by: a laptop computer or handheld device with an integrated agricultural suggestion engine 150 that can operate entirely at the farm location and may include a GPS unit 180 or any other suitable positioning device, and a fully remote computer or server configured to communicate with another personal or mobile device 200 from which the user operates system 100.

[0044] It should be understood that the existence of the remote suggestion engine and the integrated suggestion engine is not mutually exclusive. The integrated agricultural suggestion engine 150 can be a local copy of the remote agricultural suggestion engine 220 or a lightweight version thereof to support periods of low network connectivity and offline operation. Furthermore, the mobile or personal device 200 is considered to allow users to input and output data and includes any state-of-the-art mobile computing device as commonly understood.

[0045] System 100 and the remote or integrated agricultural recommendation engine may include field and farm data as well as external data and / or be configured to receive said data, wherein external data includes weather data, satellite imagery data, and other data provided by weather forecast providers or other third parties. Among other things, field data may include current and past data on at least one of the following: field and geographic identifiers regarding the geometry of farmland boundaries, including the presence of unmanaged areas within the farmland; topographic data; crop identifiers for current and past crops (crop variety and type, growth status, planting data and date, plant nutrients and health status); harvest data (yield, value, product quality, estimated or recorded historical values); and soil data (type, pH, soil organic matter (SOM), and / or cation exchange capacity (CEC)). Farm data may include other data regarding planned and past tasks, such as field maintenance practices and agricultural practices, fertilizer application data, pesticide application data, irrigation data, and other field reports, as well as historical series data, thereby allowing comparison of data with past data and processing of other administrative data, such as shifts, logs, and other organizational data. Planned and past tasks may include other activities such as monitoring of plants and pests, application of pesticides, fungicides or crop nutrient products, measurement of at least one farm or field parameter, maintenance and repair of ground hardware and other similar activities.

[0046] System 100 can be configured to receive and / or retrieve soil data from available online soil databases (such as SoilGrids from World Soil Information, SSURGO (from the USDA Soil Survey Geographic Database), or any similar soil data repository) and via user input.

[0047] System 100 may be further configured to receive any of the aforementioned data and other field data from a predetermined number of locations within or near the analysis area, manually input by the user / farmer via input / output unit 140 or received from dedicated sensors 270 via communication unit 130. Further, system 100 and agricultural suggestion engine 220 may be configured to receive weather data from nearby weather stations 260 and / or external crop / farm sensors or sensing devices 270, as well as via input unit 140. The nearby weather stations 260 and / or external crop / farm sensors 270 are configured to communicate via one or more networks. In another embodiment, weather data is provided by an external weather forecasting company. The weather data may further include a range of current and past data for at least one of the following: temperature, cumulative precipitation, relative humidity, wind speed, solar irradiance, cumulative sunshine hours, and forecasts, etc.

[0048] System 100 may be further operatively connected to agricultural equipment 300. Examples of agricultural equipment 300 include tractors, combine harvesters, harvesters, seeders, trucks, fertilizer equipment, and any other physical machinery or hardware items, typically mobile machinery, that can be used in agriculture-related tasks. In one embodiment, system 100 may be configured to communicate with agricultural equipment 300 via a wireless network to set a variable-rate application prescription for a determined crop, or alternatively, to determine a measurement area as an indication of destination. System 100 may be further configured to generate machine-readable script files for agricultural equipment 300 for fertilizer application.

[0049] Figure 5 A flowchart illustrating a method according to a main embodiment of this application is shown, which will be disclosed below.

[0050] In one embodiment, the computer-implemented method of this disclosure is configured to provide 1400 variable, location-specific fertilizer recommendations for crops, the method including the step of determining 1000 at least one farmland comprising at least one crop. An example of how the current method can determine at least one farmland could be a user-provided predetermined field, but the currently disclosed method can be configured to automatically retrieve the field to which the recommendation is intended based on farm and / or field data. The farmland can also be determined based on the user's location, which may be provided by system 100, mobile device 200, or agricultural equipment 300. In another embodiment, system 100 is configured to determine the boundaries of the determined farmland if the determined farmland does not have data regarding boundary locations in existing farm data.

[0051] The currently disclosed method is further configured to determine the crop nutrient status of at least 1300 farmlands. This application can be configured to use suitable remote data for remote determination of crop nutrient content. Remote data may refer to data provided by an imaging satellite 250 or a suitable manned or unmanned imaging vehicle 240. These satellite or vehicle systems are configured to communicate via a dedicated network and in a manner not required by the commonly used methods disclosed herein. Among the various remote data available for use, satellite data is now widely available from numerous public (NASA's LANDSAT, ESA's SENTINEL) and / or private providers. However, this method is not limited to satellite data platforms, as the spectral bands available for this method are provided across a wide range of standard satellite data available both publicly and privately. Due to the differences between different satellite and optical sensor platforms, it is not intended to limit the support of the currently disclosed method to precise and specific wavelengths, nor to provide a given wavelength for orientation. While different factors and corrections can be introduced to account for these variations, it should be understood that wavelengths close to those mentioned below are used, as the specifications of the platforms vary accordingly.

[0052] In one embodiment, the remote data is acquired from the Sentinel-2 satellite. The Sentinel-2 mission includes an MSI (Multispectral Instrument) that uses high spatial resolution data to monitor the Earth's surface. The MSI operates passively by collecting sunlight reflected from the Earth, and is therefore a more efficient and energy-saving detection method. Sentinel-2 consists of 13 bands with different spatial resolutions (10m, 20m, or 60m), located in the visible, near-infrared, and short-wave infrared portions of the spectrum. In this embodiment, the current method uses image data associated with spectral bands having at least a plurality of wavelengths including approximately 700nm and 850nm. In another embodiment, the method uses data associated with spectral bands having wavelengths of approximately 740nm and 780nm. Using the Sentinel-2 spectral bands from the MSI produces high-resolution (approximately 20m) measurements, and is therefore preferred for the currently disclosed implementation.

[0053] In embodiments, remote data may include data associated with different spectral bands identified for improving the method. Depending on the nature and source of the remote data, other compensation and calibration algorithms are considered in this application.

[0054] Once remote data is received, the method is configured to generate at least one coefficient derived from the remote data. Different coefficients (or indices) are used in the literature to obtain different agricultural information, such as the Differential Vegetation Index (NDVI) and the Normalized Difference Vegetation Index (NDVI). However, NDVI is sensitive to the effects of soil brightness, soil color, atmosphere, clouds, cloud shadows, and canopy shadows, and requires remote sensing calibration. In this sense, other coefficients to consider could include: the Atmospheric Impedance Vegetation Index (ARVI) to reduce dependence on atmospheric influences; the Soil Adjusted Vegetation Index (SAVI); or the Transformed Soil Atmospheric Impedance Vegetation Index (TSARVI) that considers the differences in vegetation under different soil types.

[0055] In one embodiment, in order to provide a reliable vegetation index or coefficient indicating aboveground nitrogen absorption present in vegetation, at least one index of this application may include different wavelengths at the so-called red edge of vegetation between 670 nm and 800 nm.

[0056] For example, the formula for the vegetation index under consideration can be expressed as: SX = f(R760, R730), where R760 and R730 represent the reflectance values ​​of the image data associated with each wavelength, or the values ​​closest to them on each satellite platform as described above, and f represents a predetermined mathematical relationship for different wavelengths according to the expected final result, as known in the field of agronomic remote sensing, where different vegetation indices are used by appropriately combining reflectance data from different suitable wavelengths. The mathematical relationship f is defined as a set of operations that define the corresponding vegetation index, as will be shown below. Therefore, this index is sensitive to the chlorophyll content present in the vegetation and can be directly derived from a direct relationship with the total nitrogen uptake of the ground within the canopy. Although the data associated with the corresponding wavelengths have been indicated as R760 and R730 for this example, this is only a non-limiting example indicating two suitable wavelengths for determining chlorophyll. Although wavelengths will be named with only 3 digits for the purpose of symbol simplicity, the use of corresponding wavelengths with values ​​exceeding 1000 is not excluded in the method of the present application. Therefore, a general vegetation index can be indicated as any arbitrary mathematical function based on an appropriate relationship between data related to a specific wavelength, and can be represented by any appropriate mathematical function f that establishes a specific relationship between wavelengths, and can then be generally expressed as follows:

[0057] SX=f(RXXX(1),RXXX(2),....,RXXX(n))

[0058] Where RXXX(i) represents each of the multiple wavelengths of image data used to determine the various indices. Typically, at least two different wavelengths derived from remote image data are used, but the present disclosure is not limited to a fixed number.

[0059] In another embodiment, in order to provide a reliable vegetation index or coefficient indicating aboveground fresh or dry matter in vegetation, the remote data may include other remote data, which may include other wavelengths at or near the water absorption zone, such as around 970 nm, 1100 nm, 1450 nm or 1950 nm, in order to improve fertilizer recommendations by adjusting other parameters present in the field (e.g., biomass).

[0060] Biomass determination can be independent of chlorophyll. In another embodiment, fresh biomass can be calculated, for example, by SW = g(R900, R970), where, as described above, R900 and R970 represent reflectance values ​​of image data associated with each wavelength, or the values ​​closest to them across various satellite platforms, and g can represent similar or different suitable mathematical relationships for different wavelengths. Determining chlorophyll content and fresh biomass, which are directly related to crop nitrogen content, allows for more accurate fertilizer recommendations that include other elements such as potassium or phosphorus.

[0061] However, the list of coefficients (or indices) does not imply limitations. Different coefficients vary considerably, and these can also be used in this method, which aims to obtain the best advantages of each vegetation index while mitigating their disadvantages by finding a suitable pair of vegetation indices, thereby improving the determination of crop nutrient status at different crop stages. For example, NDVI is highly sensitive to soil and atmospheric influences, making it noisy and exhibiting high saturation levels in the early stages of crop growth. While NDVI may provide a good, noise-free indicator of relative field variability in later stages of crop growth, it does not provide a good absolute estimate of aboveground nitrogen uptake. Given the current disclosure, this and other benefits of the current method will become clear.

[0062] The method of the present application is further configured to determine a baseline value for the crop nutrient status of at least one determined farmland. In an embodiment, the baseline value can be determined by receiving remote data, wherein the remote data includes image data of at least one farmland. Based on the received image data, the method of the present application generates at least one vegetation index as disclosed above, indicating the crop nutrient status, wherein the method of the present application is configured to determine the crop nutrient status baseline value based on the at least one vegetation index. Therefore, the baseline value is defined as a constant value for the entire farmland, however, the overall crop nutrient status of the entire farmland is taken into account when determining it.

[0063] The method of the current application is further configured to determine the field variability of crop nutrient status in at least 1200 farmlands, and to determine 1400 site-specific variable fertilizer recommendations 4000 based on the baseline value of crop nutrient status in at least one farmland and based on the field variability.

[0064] In another embodiment, field variability can be determined by remote sensing. In this embodiment, the method of the present application is configured to receive remote data, wherein the remote data includes image data, the image data including reflectance values ​​of at least one field, and generate at least one vegetation index indicating the current crop nutrient status based on the image data, and determine the current field variability of the crop nutrient status based on the vegetation index.

[0065] In another embodiment, at least one additional vegetation index is determined to determine the current field variability based on the vegetation index used to determine the baseline value for crop nutrient status. As mentioned above, a suitable vegetation index can be determined as SX = f(R760, R730). In this case, another vegetation index can be determined such that SX' = f'(R760, RXXX), as explained below, is determined, where f' represents the same mathematical relationship as SX and RXXX. Although two indices with specific wavelengths have been given above, this is not intended to be a limiting example, as any pair of defined vegetation indices following the method explained below is expected to demonstrate the required advantages.

[0066] In the above case, if the vegetation index to be used is NDVI, its formula is as follows:

[0067]

[0068] Where NIR represents the reflectance value associated with near-infrared wavelengths, and RED represents the reflectance measurement acquired in the red (visible spectrum), at least one vegetation index can be interpreted by replacing either of the two channels with other channels associated with different wavelengths, but maintaining the same specific mathematical relationship, as described below. In this case, as seen, the mathematical relationship defining NDVI is a set of additions / subtractions and / or quotients / multiplications between individual reflectance data at different wavelengths, and this mathematical relationship should be maintained to take advantage of the currently disclosed benefits.

[0069] Therefore, following the SX nomenclature used above, other vegetation indices with field variability can be summarized as follows:

[0070] SX′=f'(RXXX(1),RXXX(2),..RXXX(i'),…,RXXX(n))

[0071] Wherein, as mentioned above, f' will be the same mathematical relation that exists with at least one vegetation index SX used to determine the baseline value. RXXX(i') represents at least one other reflectance data for a given wavelength, including at least one RXXX(i') representing reflectance data in at least one vegetation index used to determine field variability replacing each of at least one of a plurality of reflectance data used to determine the baseline value, and at least one RXXX(i') represents reflectance data associated with a wavelength different from the at least one wavelength replaced.

[0072] Because of this alternative and appropriately defined pair of vegetation indices, one of which is determined to be the true absolute value representing crop nutrient status and the other is determined to address inherent instrument noise in the specific band used, and because of the mathematical relationships between the definitions of the vegetation indices, which are related to the calculation of at least one other vegetation index, a more robust determination of crop nutrient status can be achieved.

[0073] To understand the advantages of this vegetation index Figure 3 and Figure 4 This shows examples of suitable vegetation indices for SX (upward) and combinations of SX and SX' (downward), as well as different field variations for each index. This can be understood in both plots, representing different times (represented in different columns in each plot) and seasons (…). Figure 3 Display the corresponding spring season for a given year, and Figure 4 (Showing the autumn of a given year), the upward (SX) graph, while a good indicator of the absolute value of crop nutrient status as tests show, exhibits very noisy behavior, indicated by the granularity of the image. On the other hand, the downward (SX') graph shows a smoother nature of field variability and is better representative of the changes present in the field, as can be understood, and still represents a good indicator of absolute values ​​due to the favorable choice of baseline values.

[0074] Therefore, when determining appropriate vegetation indices, selection must be based on the sensitivity of the indices to different crop stages and other phenomena (bare soil, linear correlation, and signal-to-noise ratio, SNR) while maintaining good correlation with crop nutrient status. According to the following embodiments, a true representation of crop nutrient status, illustrating both the absolute level and field variability of crop nutrient status, can be achieved. While some vegetation indices may well represent the absolute level of crop nutrient status or any other order of magnitude to be measured, this may come at the cost of high noise levels in the identified farmland, compromising the reliable determination of field variability. This method achieves a robust baseline value for crop nutrient status, representing good absolute nutrient values, while appropriately selecting other vegetation indices with lower noise components, thus enabling more robust field variability.

[0075] In another embodiment, due to the advantage of establishing the agricultural characteristics of crops, spectral bands including corresponding wavelength bands associated with green, red, vegetation red edges and near-infrared bands are used (e.g., within the Sentinel-2 platform, this corresponds to bands 3-8A).

[0076] In another embodiment, the method of the present application is configured to determine which band among those commonly used for agricultural purposes exhibits the smallest standard deviation. In the case of remote satellite sensing, since the instrument characteristics of different satellite platforms are not readily available, this can be achieved by detecting at least one area of ​​a predetermined size near a identified field, which has a flat spectral response independent of a given wavelength. For example, an airport landing runway or the upper surface of dense clouds. Therefore, the method of the present application is configured to determine at least one area in the received image data that satisfies the above conditions. Once at least one area has been determined, the present method is configured to determine the spectral band among the suitable bands mentioned above that exhibits the lowest standard deviation within the at least one area. In the case of a manned or unmanned imaging aircraft 240, the method is simplified by accessing optical sensing devices and their possible analysis.

[0077] Following this determination, other vegetation indices are determined by replacing at least one of the spectral bands used in the original vegetation indices with the spectral band determined to have the lowest standard deviation near at least one farmland. While simply replacing at least one of the wavelengths with one of the appropriate wavelengths for determining agricultural characteristics reduces the noise present in the vegetation indices from the corresponding wavelengths by eliminating the existing correlations of the wavelengths defined by the mathematical relationships present in SX and decoupling the other vegetation indices SX′ from this noise, further improvements can be achieved by selecting spectral bands with lower standard deviations.

[0078] Therefore, given a pair of defined vegetation indices SX and SX′, the complete variable fertilizer recommendation, taking into account baseline values ​​and current field variability, can be determined as follows:

[0079] N REC (x, y)=F(Avg(SX)+m[SX′(x, y)-Avg(SX′)]),

[0080] Here, F represents a properly calibrated agronomic function that translates crop nutrient status into fertilizer application. F may include other terms that take into account soil nutrient content, soil organic matter, and / or mineralization data available through local analysis or in field data.

[0081] Therefore, the variable proposal N REC(x, y) can be based on the baseline value of the vegetation index SX (represented as the mean of the index in the equation above, although further measures of central tendency, such as the median, can be used), which best shows the absolute crop nutrient status and current field variability, representing the variation in at least one field determined based on the vegetation index (SX'). While it may lack a true representation of the absolute crop nutrient status, it improves the signal quality (SNR, signal-to-noise ratio) of the field variability. The calibration constant m represents an optional scaling factor between SX and SX' (which can be configured to account for different amplitudes present for different bands).

[0082] Once accurate fertilizer recommendations have been determined by following the above embodiments, the currently disclosed method is configured in another embodiment to generate machine-readable script files for agricultural equipment to apply fertilizer.

[0083] In another embodiment, the currently disclosed method further includes implementing fertilizer recommendations via agricultural implementer 300, which may receive machine-readable scripts via physical storage device or alternatively via communication unit 130.

[0084] While process steps, method steps, algorithms, etc., may be described sequentially, such processes, methods, and algorithms may be configured to operate in an alternating order. In other words, any sequence or order of steps that can be described does not necessarily indicate a requirement to perform these steps in that order. The steps of the process described herein may be performed in any actual order. Furthermore, some steps may be performed simultaneously, in parallel, or concurrently. The various methods described herein can be implemented by combining one or more machine-readable storage media containing code according to this disclosure with suitable standard computer hardware to execute the code contained therein. Apparatus for implementing the various embodiments of this disclosure may involve one or more computers (or one or more processors within a single computer) and a storage system that includes or has network access to a computer program encoded according to the various methods described herein, and the method steps of this disclosure may be performed by modules, routines, subroutines, or sub-parts of a computer program product. While various embodiments of this disclosure have been described for the foregoing, other and further embodiments of this disclosure may be devised without departing from the basic scope of this disclosure. The scope of this disclosure is defined by the appended claims. This disclosure is not limited to the described embodiments, versions, or examples, but is included in enabling a person of ordinary skill in the art to make and use this disclosure when combined with information and knowledge available to a person of ordinary skill in the art.

[0085] While plural nouns are preferred throughout the publication to allow for gender neutrality in the drafting of the text when referring to people (users, farmers), there is no limitation on the number of people that should be considered relevant to the current publication. This is done in accordance with the gender-neutral guidelines amendment that came into effect on March 1, 2021, and as an example for others.

[0086] While this disclosure has been illustrated by way of description of various embodiments, and these embodiments have been described in considerable detail, the applicant does not intend to limit or in any way restrict the scope of the appended claims to such details. Additional advantages and modifications will readily emerge for those skilled in the art. Therefore, this disclosure, in its broader aspects, is not limited to the specific details shown and described, representative apparatuses and methods, and illustrative examples.

[0087] Therefore, its detailed description should not be construed as limiting in all respects, but rather as illustrative. The scope of this disclosure should be determined by a reasonable interpretation of the appended claims, and all variations within the scope of equivalence are included within the scope of the present disclosure.

Claims

1. A computer-implemented method for providing variable fertilizer recommendations to crops, the method comprising the steps of: a. Identify at least one field containing at least one crop; b. Determining the crop nutrient status for the at least one crop in the at least one farmland, wherein determining the crop nutrient status for the at least one crop in the at least one farmland includes: i. Determine baseline values ​​and field variability of the crop nutrient status for the at least one farmland; ii. Determining baseline values ​​and field variability includes: 1) Receive remote data, said remote data including image data of the at least one farmland; 2) Generate at least one first vegetation index indicating crop nutrient status based on the image data, and determine the baseline value based on the at least one first vegetation index; 3) Generate at least one second vegetation index indicating crop nutrient status based on the image data, and determine the field variability based on the at least one second vegetation index; 4) wherein the at least one first vegetation index used to determine the baseline value is defined as wherein denotes the plurality of reflectance data comprised in the remote data related to a given wavelength for i = 1,..., n, and f is a predetermined mathematical relationship, 5) wherein the at least one second vegetation index used to determine field variability is defined as wherein f' is the same mathematical relationship as the mathematical relationship present for the at least one first vegetation index used to determine the baseline value, wherein RXXX(i') included in the at least one second vegetation index used to determine field variability replaces at least one of the plurality of reflectance data related to the given wavelength represented by RXXX(i) included in the at least one first vegetation index used to determine the baseline value, and RXXX(i') represents reflectance data related to a wavelength different from the given wavelength; iii. Determine the crop nutrient status based on the baseline values ​​and the field variability; c. Based on the determined crop nutrient status, determine variable fertilizer recommendations for the at least one farmland. Where RXXX(i') is the reflectance value selected from the bands associated with green, red, vegetation red edges, and near-infrared bands, and RXXX(i') is selected by: detecting at least one region near the identified farmland that has a spectral flatness value independent of the given wavelength, and selecting at least one band from the green, red, vegetation red edge and near-infrared bands that exhibits the lowest standard deviation for the detected at least one region.

2. The method of claim 1, wherein the variable fertilizer recommendation is further adjusted based on field data or local analysis of soil nutrient content, soil organic matter and / or mineralization data.

3. The method of claim 1, wherein the method further comprises generating a machine-readable script file for agricultural equipment based on determined variable fertilizer recommendations to perform corresponding fertilizer application.

4. The method of claim 1, wherein the method further comprises implementing the variable fertilizer recommendation by means of an agricultural device configured to perform fertilizer application.

5. A system for providing fertilizer recommendations, the system comprising means configured to perform the method according to any one of claims 1 to 4.

6. A data processing apparatus comprising means for performing the method according to any one of claims 1 to 4.

7. A computer-readable storage medium comprising instructions that, when executed by a computer system, cause the computer system to perform the method according to any one of claims 1 to 4.

8. A computer program product comprising instructions that, when executed by a computer, cause the computer to perform the method according to any one of claims 1 to 4.

Images