Description:BACKGROUND
Field of Invention
[001] Embodiments of the present invention generally relate to a drought forecasting system and particularly to a drought anomaly detection system for large-scale agricultural farms.
Description of Related Art
[002] Agricultural drought is a primary factor behind crop failure, food insecurity, and economic losses in farming communities. Large-scale farms often face challenges in detecting drought conditions early enough to act before yield losses become irreversible. A lack of timely and precise information prevents farmers, agronomists, and policymakers from making informed decisions about irrigation, crop planning, and resource allocation.
[003] Current solutions rely on satellite-based systems or Internet of Things (IoT) sensor networks. Satellite-based systems provide wide-area coverage and identify vegetation stress or reduced soil moisture, while IoT networks measure soil moisture, temperature, and humidity at the ground level. These approaches provide useful insights and enable partial monitoring of agricultural drought, with each focusing on either broad spatial coverage or localized precision.
[004] Despite their utility, existing solutions show limitations that reduce effectiveness at scale. Satellite systems often provide low-resolution data, experience delays of days or weeks, and remain vulnerable to cloud cover or weather interference. IoT sensor networks, while precise, require high deployment costs, provide limited coverage, and fail to integrate broader climatic trends.
[005] There is thus a need for an improved and advanced drought anomaly detection system for large-scale agricultural farms that can administer the aforementioned limitations in a more efficient manner.
SUMMARY
[006] Embodiments in accordance with the present invention provide a drought anomaly detection system for large-scale agricultural farms. The system comprising a data acquisition unit adapted to acquire remote sensing data from a satellite. The remote sensing data is selected from vegetation indices, a land surface temperature, soil moisture values, or a combination thereof. The system further comprising sensors adapted to measure real-time field-level data selected from a soil moisture, an ambient temperature, a humidity, a rainfall, or a combination thereof. The sensors comprises an Internet of Things (IoT) enabled moisture sensors, temperature sensors, humidity sensors, rainfall gauges, leaf wetness sensors, or a combination thereof. The system further comprising a processing unit communicatively connected to the data acquisition unit and to the sensors. The processing unit is configured to receive the remote sensing data and the real-time field-level data from the data acquisition unit and the sensors, respectively; pre-process the received remote sensing data and the received real-time field-level data by filtering out noise due to cloud interference, cross-validating with inputs of the sensors, or a combination thereof; synchronize the processed remote sensing data and the processed real-time field-level data on a temporal and spatial basis. The synchronization of data comprises geospatial alignment of satellite imagery with Global Positioning System coordinates of the sensors and temporal alignment of sensor readings with satellite acquisition time; process the synchronized data using an artificial intelligence-based data fusion engine to calculate a risk score indicative of early drought stress; compare the risk score with a threshold score; and generate a drought alert, when the risk score is greater than the threshold score.
[007] Embodiments in accordance with the present invention further provide a method for detecting drought anomalies across large-scale agricultural farms. The method comprising steps of receiving remote sensing data and real-time field-level data from a data acquisition unit and sensors, respectively; pre-processing the received remote sensing data and the received real-time field-level data by filtering out noise due to cloud interference, cross-validating with inputs of the sensors, or a combination thereof; synchronizing the processed remote sensing data and the processed real-time field-level data on a temporal and spatial basis. The synchronization of data comprises geospatial alignment of satellite imagery with Global Positioning System coordinates of the sensors and temporal alignment of sensor readings with satellite acquisition time; processing the synchronized data using an artificial intelligence-based data fusion engine to calculate a risk score indicative of early drought stress; comparing the risk score with a threshold score; and generating a drought alert, when the risk score is greater than the threshold score.
[008] Embodiments of the present invention may provide a number of advantages depending on their particular configuration. First, embodiments of the present application may provide a drought anomaly detection system for large-scale agricultural farms.
[009] Next, embodiments of the present application may provide a drought anomaly detection system that provides an integrated platform that combines wide-area satellite remote sensing data with real-time IoT sensor inputs for both macro-level coverage and micro-level precision in drought monitoring.
[0010] Next, embodiments of the present application may provide a drought anomaly detection system that enables early detection of drought anomalies by analyzing subtle changes in vegetation indices, soil moisture, and climatic variables before visible crop stress occurs, thus supporting proactive intervention.
[0011] Next, embodiments of the present application may provide a drought anomaly detection system that offers zone-specific drought risk assessment within large-scale farms, allowing targeted irrigation planning, efficient resource allocation, and localized decision-making.
[0012] Next, embodiments of the present application may provide a drought anomaly detection system that incorporates artificial intelligence and machine learning algorithms that enhance accuracy by cross-validating satellite-derived anomalies with sensor data, resulting in improved prediction of drought onset and severity.
[0013] Next, embodiments of the present application may provide a drought anomaly detection system that delivers real-time visualization and automated alerts through a user-friendly interface, thereby enabling timely decision-making and seamless integration with existing farm management or irrigation systems.
[0014] These and other advantages will be apparent from the present application of the embodiments described herein.
[0015] The preceding is a simplified summary to provide an understanding of some embodiments of the present invention. This summary is neither an extensive nor exhaustive overview of the present invention and its various embodiments. The summary presents selected concepts of the embodiments of the present invention in a simplified form as an introduction to the more detailed description presented below. As will be appreciated, other embodiments of the present invention are possible utilizing, alone or in combination, one or more of the features set forth above or described in detail below.
BRIEF DESCRIPTION OF THE DRAWINGS
[0016] The above and still further features and advantages of embodiments of the present invention will become apparent upon consideration of the following detailed description of embodiments thereof, especially when taken in conjunction with the accompanying drawings, and wherein:
[0017] FIG. 1 illustrates a block diagram of a drought anomaly detection system for large-scale agricultural farms, according to an embodiment of the present invention;
[0018] FIG. 2 illustrates a connectivity diagram of a drought anomaly detection system for large-scale agricultural farms, according to an embodiment of the present invention; and
[0019] FIG. 3 depicts a flowchart of a method for detecting drought anomalies across large-scale agricultural farms, according to an embodiment of the present invention.
[0020] The headings used herein are for organizational purposes only and are not meant to be used to limit the scope of the description or the claims. As used throughout this application, the word "may" is used in a permissive sense (i.e., meaning having the potential to), rather than the mandatory sense (i.e., meaning must). Similarly, the words “include”, “including”, and “includes” mean including but not limited to. To facilitate understanding, like reference numerals have been used, where possible, to designate like elements common to the figures. Optional portions of the figures may be illustrated using dashed or dotted lines, unless the context of usage indicates otherwise.
DETAILED DESCRIPTION
[0021] The following description includes the preferred best mode of one embodiment of the present invention. It will be clear from this description of the invention that the invention is not limited to these illustrated embodiments but that the invention also includes a variety of modifications and embodiments thereto. Therefore, the present description should be seen as illustrative and not limiting. While the invention is susceptible to various modifications and alternative constructions, it should be understood, that there is no intention to limit the invention to the specific form disclosed, but, on the contrary, the invention is to cover all modifications, alternative constructions, and equivalents falling within the scope of the invention as defined in the claims.
[0022] In any embodiment described herein, the open-ended terms "comprising", "comprises”, and the like (which are synonymous with "including", "having” and "characterized by") may be replaced by the respective partially closed phrases "consisting essentially of", “consists essentially of", and the like or the respective closed phrases "consisting of", "consists of”, the like.
[0023] As used herein, the singular forms “a”, “an”, and “the” designate both the singular and the plural, unless expressly stated to designate the singular only.
[0024] FIG. 1 illustrates a block diagram of a drought anomaly detection system 100 (hereinafter referred to as the system 100) for large-scale agricultural farms, according to an embodiment of the present invention. In an embodiment of the present invention, the system 100 may be adapted to acquire environmental and satellite-based data, process the acquired data using intelligent algorithms, and generate outputs indicative of potential drought conditions.
[0025] The system 100 may be adapted to integrate multiple functional units that collectively enable acquisition of remote data, collection of field-level measurements, execution of pre-processing and data fusion tasks, and generation of reliable drought anomaly information. The integration within the system 100 ensures that the data may be synchronized, validated, and analyzed in a manner that improves accuracy and reliability of drought detection. The system 100 may operate as a unified platform that combines sensing, processing, communication, and power management capabilities to provide early warning of drought anomalies. By functioning as a modular and scalable architecture, the system 100 supports deployment across diverse agricultural environments and enables effective monitoring of large-scale farms.
[0026] In an embodiment of the present invention, the system 100 may support remote firmware updates for Internet of Things sensors through over-the-air protocols, and a dashboard (not shown) may include system health monitoring of battery levels and connectivity status. In an embodiment of the present invention, initial deployment of the system 100 may include a site survey to determine ideal sensor locations using satellite-based vegetation and soil moisture variability. Further, each sensor node may be calibrated for local soil type and climatic conditions.
[0027] In an embodiment of the present invention, the system 100 may provide application programming interfaces that allow integration with automated irrigation systems, farm management platforms, or government monitoring systems. In an embodiment of the present invention, the system 100 may support horizontal scalability by adding new clusters of sensors for additional farms, and the analytics engine may be modular and adaptable to different crop types and agro-climatic regions.
[0028] According to the embodiments of the present invention, the system 100 may incorporate non-limiting hardware components to enhance a processing speed and an efficiency, such as the system 100 may comprise a data acquisition unit 102, sensors 104, a processing unit 106, an edge computing unit 108, a communication interface 110, and a power supply unit 112. In an embodiment of the present invention, the hardware components of the system 100 may be integrated with computer-executable instructions for overcoming the challenges and the limitations of the existing systems.
[0029] In an embodiment of the present invention, the data acquisition unit 102 may be adapted to acquire remote sensing data from a satellite. In an embodiment of the present invention, the data acquisition unit 102 may acquire satellite imagery and data through platforms and application programming interfaces such as Copernicus Open Access Hub, NASA Earth data, or Google Earth Engine, with data resolution ranging between 10 to 30 meters. The remote sensing data may be selected from vegetation indices, land surface temperature, soil moisture values, and so forth. The vegetation indices may include a Normalized Difference Vegetation Index and an Enhanced Vegetation Index. The satellite supplying the remote sensing data to the data acquisition unit 102 may be, but not limited to, a Sentinel-2, Moderate Resolution Imaging Spectroradiometer (MODIS), a Landsat, and so forth. In an embodiment of the present invention, the data acquisition unit 102 may be adapted to extract rainfall estimates and evapotranspiration values in addition to vegetation indices, land surface temperature, and soil moisture values, thereby enriching the dataset for drought anomaly detection.
[0030] In an embodiment of the present invention, the sensors 104 may be adapted to measure real-time field-level data selected from a soil moisture, an ambient temperature, a humidity, a rainfall, and so forth. The sensors 104 comprise an Internet of Things (IoT) enabled moisture sensors, temperature sensors (DHT22), humidity sensors (SHT31), rainfall gauges, leaf wetness sensors, a tipping bucket sensor, an ultrasonic gauge, and so forth. The sensors 104 may be deployed in stratified zones of the farm based on soil type, crop variety, and topography, thereby ensuring representative data collection across diverse micro-environments. The sensors 104 may communicate with a gateway device using protocols such as Long Range Wide Area Network (LoRaWAN), Narrowband Internet of Things (NB-IoT), Zigbee, or Wi-Fi, depending on the infrastructure available in the farm. The sensors 104 may be adapted to dispatch the real-time field-level data to a gateway device (not shown) using protocols such as Long Range Wide Area Network (LoRaWAN), Narrowband Internet of Things (NB-IoT), Zigbee, or Wi-Fi, depending on the infrastructure available in the farm. The gateway device may be adapted to collect data from a plurality of Internet of Things sensors and forward the data to a cloud server using cellular (4G/5G), Ethernet, or Wi-Fi.
[0031] In an embodiment of the present invention, the processing unit 106 may be communicatively connected to the data acquisition unit 102 and to the sensors 104. In an embodiment of the present invention, initial deployment of the system may include a site survey supported by the data acquisition unit 102 to determine optimal locations for the sensors 104, wherein each sensor node may be calibrated for local soil types and climatic conditions, and the processing unit 106 may tune satellite algorithms using baseline Normalized Difference Vegetation Index and Land Surface Temperature trends. The processing unit 106 may be, an Arduino, an Espressif Systems 32-bit Microcontroller (ESP32), a STMicroelectronics 32-bit Microcontroller (STM32), and so forth.
[0032] In an embodiment of the present invention, the processing unit 106 may be configured to receive the remote sensing data from the data acquisition unit 102 and the real-time field-level data from the sensors 104 through a combination of wired and wireless modems. In an embodiment of the present invention, the sensors 104 may transmit the real-time field-level data to the processing unit 106 directly or through an edge computing unit 108, while the data acquisition unit 102 may acquire satellite imagery from external services via application programming interfaces and automatically geo-reference the imagery to match sensor coordinates, wherein all data streams may be stored in a secure cloud database. The processing unit 106 may comprise input ports, data buses, and communication protocols that enable seamless transfer of heterogeneous data streams.
[0033] The data acquisition unit 102 may supply satellite-derived remote sensing data in a structured digital format such as a Geographic Tagged Image File Format (GeoTIFF), a Network Common Data Form (NetCDF), a JavaScript Object Notation (JSON), and so forth, that may be transmitted to the processing unit 106 through a cloud-based application programming interface or through a dedicated communication channel. The sensors 104 may transmit real-time field-level data such as soil moisture, temperature, humidity, and rainfall in the form of time-stamped packets via wireless protocols such as Long Range Wide Area Network (LoRaWAN), Narrowband Internet of Things (NB-IoT), Zigbee (Zigbee), or Wireless Fidelity (Wi-Fi). The processing unit 106 may include a data ingestion layer that authenticates, decodes, and aligns the incoming data streams. This ingestion layer may use temporal synchronization to match sensor readings with satellite acquisition times, and spatial synchronization to map sensor coordinates with satellite imagery. The processing unit 106 may be configured to buffer the incoming data to handle differences in latency, and may use error-checking mechanisms such as cyclic redundancy check (CRC) to ensure integrity of the transmitted information. Through this configuration, the processing unit 106 may be enabled to reliably receive, interpret, and store both satellite-derived remote sensing data and sensor-based field data, thereby ensuring a unified and accurate input dataset for further analysis and drought anomaly detection.
[0034] The processing unit 106 may be configured to pre-process the received remote sensing data and the received real-time field-level data by filtering out noise due to cloud interference, cross-validating with inputs of the sensors 104, and so forth. The pre-processing may include filtering out noise, validating inconsistencies, and normalizing the data into a unified format. The processing unit 106 may be configured to apply cloud masking algorithms to the satellite-derived remote sensing data to filter out distortions caused by cloud interference, haze, or atmospheric scattering. For example, spectral thresholding and machine learning classifiers may be used to identify cloud-covered pixels in Normalized Difference Vegetation Index or Land Surface Temperature images, and such pixels may be flagged or excluded from subsequent computations.
[0035] The processing unit 106 may be configured to further perform cross-validation of the satellite-derived parameters with the sensor-based measurements. For instance, soil moisture values inferred from remote sensing imagery may be compared with ground-truth values measured by soil moisture sensors of the sensors 104. When discrepancies exceed a predefined threshold, the processing unit 106 may be configured to adjust the satellite-derived values using regression models or bias-correction algorithms based on the sensor data. Additionally, the processing unit 106 may be configured to perform temporal smoothing and spatial interpolation to fill missing data points and reduce random fluctuations in the incoming streams. This may include applying moving averages, spline interpolation, or kriging techniques to create a continuous and consistent dataset across time and space. Through such pre-processing, the processing unit 106 may be configured to ensure that the satellite-derived data and the sensor data are accurate, noise-reduced, and harmonized, thereby providing a reliable basis for anomaly detection and drought risk assessment.
[0036] The processing unit 106 may be configured to synchronize the processed remote sensing data and the processed real-time field-level data on a temporal and spatial basis. The synchronization of data comprises geospatial alignment of satellite imagery with Global Positioning System coordinates of the sensors and temporal alignment of sensor readings with satellite acquisition time. The processing unit 106 may be configured to perform geospatial alignment of the satellite imagery with the Global Positioning System coordinates of the sensors so that each field measurement may be mapped to the correct pixel or georeferenced unit in the satellite dataset. The processing unit 106 may be configured to use geographic information system techniques such as coordinate transformation, resampling, and reprojection to ensure that the satellite imagery and the sensor coordinates correspond to the same spatial grid. The processing unit 106 may be configured to perform temporal alignment of the sensor readings with the satellite acquisition time. Since the sensors 104 may generate continuous time-series data while the satellite may provide imagery at periodic intervals, the processing unit 106 may be configured to interpolate, average, or resample the sensor readings to match the temporal resolution of the satellite data. For example, soil moisture measurements taken every hour may be aggregated to coincide with the timestamp of a satellite pass.
[0037] The processing unit 106 may be configured to manage latency differences by buffering incoming data streams and employing time-synchronization protocols to standardize timestamps. The processing unit 106 may be configured to apply metadata tagging so that every data element contains both temporal and spatial identifiers, enabling accurate merging of datasets. Through this configuration, the processing unit 106 may be configured to generate a harmonized dataset such that each remote sensing parameter may be accurately matched with corresponding ground-truth measurements in both space and time, thereby enabling precise anomaly detection and risk assessment.
[0038] The processing unit 106 may be configured to process the synchronized data using an artificial intelligence-based data fusion engine to calculate a risk score indicative of early drought stress. The artificial intelligence-based data fusion engine employs machine learning algorithms selected from a Random Forest, a Long Short-Term Memory network, a Support Vector Machine, or a combination thereof for anomaly detection. The processing unit 106 may be configured to integrate satellite-derived vegetation indices, land surface temperature, and soil moisture values with sensor-derived parameters such as soil moisture, ambient temperature, humidity, and rainfall into a unified feature set. The processing unit 106 may be configured to apply feature normalization and dimensionality reduction techniques to prepare the synchronized data for machine learning analysis. The processing unit 106 may be configured to input the unified feature set into the artificial intelligence-based data fusion engine, that may be adapted to employ machine learning algorithms selected from the Random Forest, the Long Short-Term Memory network, the Support Vector Machine, or a combination thereof for anomaly detection.
[0039] The processing unit 106 may be configured to train the artificial intelligence-based data fusion engine using historical datasets of climate variables, crop growth patterns, and known drought events, enabling the model to learn associations between input features and drought outcomes. The processing unit 106 may be configured to continuously update and retrain the engine as new data becomes available, thereby improving accuracy over time. The processing unit 106 may be configured to calculate the risk score by quantifying the probability or severity of drought stress in a specific zone. The risk score may be expressed as a numerical value or percentage, and may be compared against a threshold to classify the region, such as low risk, medium risk, or high risk. The processing unit 106 may be configured to generate this risk score dynamically, allowing near real-time assessment of drought anomalies. Through this configuration, the processing unit 106 may be configured to deliver a reliable, adaptive, and intelligent mechanism for early detection of drought stress before visible crop damage occurs.
[0040] The processing unit 106 may be configured to compare the risk score with a threshold score. The processing unit 106 may be configured to store one or more threshold scores in a memory, the threshold scores being predefined based on historical drought patterns, crop-specific vulnerability indices, or regulatory guidelines. The processing unit 106 may be configured to retrieve the calculated risk score from the artificial intelligence-based data fusion engine and evaluate the score against the stored threshold. The processing unit 106 may be configured to implement a simple comparison logic in which a drought condition may be indicated when the risk score exceeds the threshold score. The processing unit 106 may further be configured to employ multiple threshold levels to classify severity of drought stress, for example, into categories such as “low”, “moderate”, and “high”.
[0041] The processing unit 106 may be configured to dynamically update the threshold score based on seasonal variations, crop type, or external inputs from agronomists. The processing unit 106 may be configured to adjust thresholds adaptively using statistical models or rolling averages of recent data, thereby maintaining relevance under changing climatic conditions. The processing unit 106 may be configured to record the comparison results and tag them with corresponding temporal and spatial metadata, ensuring that each evaluation may be traceable and may be visualized on the dashboard. Through this configuration, the processing unit 106 may be configured to perform reliable threshold-based classification of the risk score, enabling timely generation of drought alerts and effective decision support for farm management.
[0042] The processing unit 106 may be configured to generate a drought alert, when the risk score may be greater than the threshold score. The processing unit 106 may be configured to produce the alert in association with the geolocation of the detected anomaly, the time of occurrence, and the severity level derived from the magnitude of the risk score. The processing unit 106 may be configured to generate the drought alert in text-based, numeric, or graphical formats suitable for interpretation by a user device 200 (as shown in FIG. 2) or an external system. The processing unit 106 may be configured to encode the drought alert with information that allows identification of affected zones within the agricultural farm.
[0043] The processing unit 106 may be configured to archive the generated drought alert in a storage medium for historical tracking and validation purposes. The processing unit 106 may further be configured to prepare the drought alert for transmission through the communication interface 110 so that the alert may be delivered to external devices. The processing unit 106 may be configured to ensure that the drought alert reflects the analytical outcome of the comparison between the risk score and the threshold score, thereby providing a reliable and actionable indication of drought stress conditions.
[0044] The processing unit 106 may be configured to transmit the generated drought alert to the user device 200, for visualization and decision support, via the communication interface 110. The processing unit 106 may be configured to enable the communication interface 110 to establish a communication link with the user device 200 through wireless or wired connections, including cellular networks, Wi-Fi, Ethernet, or low-power wide-area technologies. The processing unit 106 may be configured to encode the drought alert in one or more transmission formats, such as short message service, email, push notification, or structured data packets suitable for application programming interfaces. The processing unit 106 may be configured to embed metadata into the alert, including geographic coordinates of the affected area, a timestamp of detection, and a severity classification based on the magnitude of the risk score.
[0045] The processing unit 106 may be configured to implement data integrity and authentication mechanisms such as checksum validation and encryption protocols to ensure that the drought alert may be delivered securely and without error. The processing unit 106 may be configured to retransmit the alert in case of communication failure, thereby guaranteeing reliable delivery to the user device 200. The processing unit 106 may be configured to support simultaneous transmission of the drought alert to multiple user devices, enabling farm managers, agronomists, and other stakeholders to receive the information in real time. The processing unit 106 may further be configured to generate acknowledgments when the user device 200 successfully receives the alert, thereby confirming completion of the transmission process. Through this configuration, the processing unit 106 may be configured to directly deliver actionable drought alerts to user device 200, ensuring timely access to critical information for decision-making in agricultural management.
[0046] In an embodiment of the present invention, the processing unit 106 106 may be connected to a cloud backend hosted on a platform such as Amazon Web Services, Microsoft Azure, or Google Cloud Platform, the backend being configured to manage data ingestion, storage, time-series processing, and satellite image pre-processing, including cloud masking and vegetation index calculation. In an embodiment of the present invention, the processing unit 106 may operate in conjunction with a cloud backend hosted on platforms such as Amazon Web Services, Microsoft Azure, or Google Cloud Platform, employing services including Amazon Simple Storage Service, Influx DB, or Timescale DB for object storage and time-series handling, and applying pre-processing functions such as cloud masking and vegetation index calculation. In an embodiment of the present invention, the processing unit 106 may be configured to deliver remote firmware updates to the sensors 104 through over-the-air protocols, to perform redundant cloud backups of collected data, and to provide dashboard-based monitoring of system health, including battery status and connectivity.
[0047] In an embodiment of the present invention, the edge computing unit 108 may be adapted to perform preliminary data cleaning, compression, and threshold-based alerts in offline mode. The edge computing unit 108 may be adapted to filter out redundant, incomplete, or erroneous entries from the incoming remote sensing data and sensor data streams. The edge computing unit 108 may be adapted to normalize the filtered data into standardized formats to ensure compatibility with subsequent processing stages. The edge computing unit 108 may be adapted to apply compression techniques, including lossless data encoding or lightweight binary serialization, to reduce the data volume while maintaining integrity of critical parameters. Such compression allows efficient storage and transmission, particularly in resource-constrained agricultural environments.
[0048] The edge computing unit 108 may further be adapted to implement threshold-based evaluation of key parameters such as soil moisture, temperature, or humidity. When the measured values exceed or fall below predefined static thresholds, the edge computing unit 108 may be adapted to generate a local drought alert without invoking the complete artificial intelligence-based data fusion engine. The edge computing unit 108 may be adapted to execute these functions autonomously in offline mode when connectivity with cloud or central processing resources may be unavailable. Upon restoration of connectivity, the edge computing unit 108 may be adapted to synchronize the locally processed datasets and generated alerts with the main system to maintain consistency and continuity. Through this configuration, the edge computing unit 108 may be adapted to enhance resilience of the system 100 by ensuring uninterrupted operation, localized decision support, and reduced dependency on continuous network availability.
[0049] In an embodiment of the present invention, the communication interface 110 may be adapted to transmit the generated drought alert to the user device 200. The communication interface 110 may be adapted to transmit the generated drought alert in a format such as, but not limited to, a short message service (SMS), an email, an application programming interface of an irrigator, and so forth. In an embodiment of the present invention, the communication interface 110 may be adapted to employ lightweight protocols such as Message Queuing Telemetry Transport (MQTT) or Hypertext Transfer Protocol (HTTP) to transmit data from the sensors 104 to the processing unit 106 with low latency and high reliability.
[0050] In an embodiment of the present invention, the power supply unit 112 may be adapted to supply operational power to the sensors 104. The power supply unit 112 may be, but not limited to, solar panels, lithium-ion batteries, and so forth. The solar panels, in combination with the lithium-ion batteries, may support off-grid operation to ensure uninterrupted data collection.
[0051] FIG. 2 illustrates a connectivity diagram of the system 100, according to an embodiment of the present invention. In an embodiment of the present invention, the user device 200 may be adapted to receive the generated drought alert from the system 100 via the communication interface 110. The generated drought alert may be provided, on the user device 200, in the form of risk scores, moisture balance maps, predictive drought forecasts, and so forth. The generated drought alert may be decoded and executed on the user device 200 for visualization and decision support. The visualization may include the dashboard displaying sensor analytics, satellite overlays, and historical drought patterns. The dashboard may generate and project weekly drought summaries, crop vulnerability assessments, and water demand forecasts, that may be provided to the user through the dashboard or by automated reports. In an embodiment of the present invention, the communication interface 110 may be adapted to provide the dashboard that may enable configuration of custom alert thresholds, interactive zoom and filter functions, and real-time overlays of satellite data from the data acquisition unit 102 and sensor data from the sensors 104. In an embodiment of the present invention, the dashboard may be implemented using web technologies such as React, Leaflet.js, or Mapbox, and may support interactive features including zoom, filter, and customizable alert thresholds.
[0052] FIG. 3 depicts a flowchart of a method for detecting the drought anomalies across the large-scale agricultural farms using the system 100, according to an embodiment of the present invention.
[0053] At step 302, the system 100 may receive the remote sensing data and the real-time field-level data from the data acquisition unit 102 and the sensors 104.
[0054] At step 304, the system 100 may pre-process the received remote sensing data and the received real-time field-level data by filtering out noise due to cloud interference, cross-validating with inputs of the sensors 104, and so forth.
[0055] At step 306, the system 100 may synchronize the processed remote sensing data and the processed real-time field-level data on the temporal and spatial basis. The synchronization of data may comprise geospatial alignment of satellite imagery with Global Positioning System coordinates of the sensors and temporal alignment of sensor readings with satellite acquisition time.
[0056] At step 308, the system 100 may process the synchronized data using the artificial intelligence-based data fusion engine to calculate the risk score indicative of the early drought stress.
[0057] At step 310, the system 100 may compare the risk score with the threshold score. Upon comparison, if the risk score may be greater than the threshold score, then the method 300 may proceed to a step 312. Else, the method 300 may revert to the step 302.
[0058] At step 312, the system 100 may generate the drought alert.
[0059] At step 314, the system 100 may transmit the generated drought alert to the user device 200, for visualization and decision support, via the communication interface 110.
[0060] While the invention has been described in connection with what is presently considered to be the most practical and various embodiments, it is to be understood that the invention is not to be limited to the disclosed embodiments, but on the contrary, is intended to cover various modifications and equivalent arrangements included within the scope of the appended claims.
[0061] This written description uses examples to disclose the invention, including the best mode, and also to enable any person skilled in the art to practice the invention, including making and using any devices or systems and performing any incorporated methods. The patentable scope of the invention is defined in the claims, and may include other examples that occur to those skilled in the art. Such other examples are intended to be within the scope of the claims if they have structural elements that do not differ from the literal language of the claims, or if they include equivalent structural elements within substantial differences from the literal language of the claims. , Claims:CLAIMS
I/We Claim:
1. A drought anomaly detection system (100) for large-scale agricultural farms, the system (100) comprising:
a data acquisition unit (102) adapted to acquire remote sensing data from a satellite, wherein the remote sensing data is selected from vegetation indices, a land surface temperature, soil moisture values, or a combination thereof;
sensors (104) adapted to measure real-time field-level data selected from a soil moisture, an ambient temperature, a humidity, a rainfall, or a combination thereof, wherein the sensors (104) comprise an Internet of Things (IoT) enabled moisture sensors, temperature sensors, humidity sensors, rainfall gauges, leaf wetness sensors, or a combination thereof;
a processing unit (106) communicatively connected to the data acquisition unit (102) and to the sensors (104), characterized in that the processing unit (106) is configured to:
receive the remote sensing data and the real-time field-level data from the data acquisition unit (102) and the sensors (104), respectively;
pre-process the received remote sensing data and the received real-time field-level data by filtering out noise due to cloud interference, cross-validating with inputs of the sensors (104), or a combination thereof;
synchronize the processed remote sensing data and the processed real-time field-level data on a temporal and spatial basis, wherein the synchronization of data comprises geospatial alignment of satellite imagery with Global Positioning System coordinates of the sensors and temporal alignment of sensor readings with satellite acquisition time;
process the synchronized data using an artificial intelligence-based data fusion engine to calculate a risk score indicative of early drought stress;
compare the risk score with a threshold score; and
generate a drought alert, when the risk score is greater than the threshold score.
2. The system (100) as claimed in claim 1, wherein the processing unit (106) is configured to transmit the generated drought alert to a user device (200), for visualization and decision support, via a communication interface (110).
3. The system (100) as claimed in claim 1, wherein the artificial intelligence-based data fusion engine employs machine learning algorithms selected from a Random Forest, a Long Short-Term Memory network, a Support Vector Machine, or a combination thereof for anomaly detection.
4. The system (100) as claimed in claim 1, wherein the vegetation indices include a Normalized Difference Vegetation Index and an Enhanced Vegetation Index.
5. The system (100) as claimed in claim 1, wherein the generated drought alert is provided, on a user device (200), in the form of risk scores, moisture balance maps, predictive drought forecasts, or a combination thereof.
6. The system (100) as claimed in claim 1, wherein the satellite supplying the remote sensing data to the data acquisition unit (102) is selected from a Sentinel-2, Moderate Resolution Imaging Spectroradiometer (MODIS), a Landsat, or a combination thereof.
7. The system (100) as claimed in claim 1, comprising an edge computing unit (108) adapted to perform preliminary data cleaning, compression, and threshold-based alerts in offline mode.
8. The system (100) as claimed in claim 1, comprising a power supply unit (112) adapted to supply operational power to the sensors (104).
9. A method (300) for detecting drought anomalies across large-scale agricultural farms, the method (300) is characterized by steps of:
receiving remote sensing data and real-time field-level data from a data acquisition unit (102) and sensors (104), respectively;
pre-processing the received remote sensing data and the received real-time field-level data by filtering out noise due to cloud interference, cross-validating with inputs of the sensors (104), or a combination thereof;
synchronizing the processed remote sensing data and the processed real-time field-level data on a temporal and spatial basis, wherein the synchronization of data comprises geospatial alignment of satellite imagery with Global Positioning System coordinates of the sensors and temporal alignment of sensor readings with satellite acquisition time;
processing the synchronized data using an artificial intelligence-based data fusion engine to calculate a risk score indicative of early drought stress;
comparing the risk score with a threshold score; and
generating a drought alert when the risk score is greater than the threshold score.
10. The method (300) as claimed in claim 9, comprising a step of transmitting the generated drought alert to a user device (200), for visualization and decision support, via a communication interface (110).
Date: October 10, 2025
Place: Noida

Nainsi Rastogi
Patent Agent (IN/PA-2372)
Agent for the Applicant