Consult an Expert
Trademark
Design Registration
Consult an Expert
Trademark
Copyright
Patent
Infringement
Design Registration
More
Consult an Expert
Consult an Expert
Trademark
Design Registration
Login
INTELLIGENT HYBRIDIZATION SYSTEM FOR YIELD OPTIMIZATION USING MACHINE LEARNING
Extensive patent search conducted by a registered patent agent
Patent search done by experts in under 48hrs
₹999
₹399
Abstract
Information
Inventors
Applicants
Specification
Documents
ORDINARY APPLICATION
Published
Filed on 14 November 2024
Abstract
This invention introduces a machine learning-based system to enhance hybrid plant breeding by identifying optimal parent plant combinations for increased yield. The system utilizes real-time and historical environmental data, including temperature, humidity, soil moisture, and light levels, alongside plant growth metrics like chlorophyll content and growth rate. These inputs are analyzed by an embedded machine learning algorithm that recommends parent combinations with high heterosis potential, improving yield predictability and minimizing the number of crosses. The system integrates a cloud-based data center, communication module, and sensor array for data collection and storage, accessible to breeders for offline data input as needed. This approach optimizes selection processes, reduces labor and cost, and shortens breeding cycles, allowing breeders to focus on cross combinations with the highest yield potential.
Patent Information
Application ID | 202411087975 |
Invention Field | BIOTECHNOLOGY |
Date of Application | 14/11/2024 |
Publication Number | 48/2024 |
Inventors
Name | Address | Country | Nationality |
---|---|---|---|
Jagdeep Singh | Ward no-7, Pant Nagar, Kaithal - 136207, Haryana, India | India | India |
Neha Goyal | 1130, U.E. Sector 5, Kurukshetra- 136118, Haryana, India | India | India |
Rajiv Bansal | 1130, U.E. Sector 5, Kurukshetra- 136118, Haryana, India | India | India |
Applicants
Name | Address | Country | Nationality |
---|---|---|---|
Jagdeep Singh | Ward no-7, Pant Nagar, Kaithal - 136207, Haryana, India | India | India |
Neha Goyal | 1130, U.E. Sector 5, Kurukshetra- 136118, Haryana, India | India | India |
Rajiv Bansal | 1130, U.E. Sector 5, Kurukshetra- 136118, Haryana, India | India | India |
Specification
Description:The following specification particularly describes the invention and the manner in which it is to be performed:
TECHNICAL FIELD
The present invention relates the method of increasing crop or plant yield more specifically to a machine learning-embedded system that aids breeders in selecting optimal parent plants for hybridization.
BACKGROUND
[001] Hybridization has long been a core method in agriculture to enhance crop characteristics like yield, resistance, and adaptability. In traditional plant breeding, hybridization requires selecting suitable parent plants based on phenotypic traits and environmental compatibility, followed by extensive testing to identify crosses with high heterosis. This process is complex, labor-intensive, and time-consuming, often yielding unpredictable results.
[002] Besides this Environmental factors including temperature, humidity, light, and soil conditions significantly influence hybrid performance, complicating the selection of optimal parent combinations.
SUMMARY
[003] This invention presents a machine learning (ML)-embedded system that assists breeders in selecting optimal parent plants for hybridization to maximize yield. The system incorporates an array of sensors to collect real-time data on environmental factors, such as temperature, humidity, soil moisture, and light levels, as well as plant growth indicators like chlorophyll content and growth rate. These inputs are processed by a central controller embedded with a machine learning based algorithm, which uses both historical and real-time data to recommend parent combinations with high heterosis potential.
[004] The invention is describes a ML based system that leverages real-time environmental data and historical performance records to determine optimal parent pairs.
[005] In one of the implementations, by simplifying parent selection, this system not only reduces labor and cost but also enhances predictability, enabling breeders to focus on the most promising combinations.
[006] In another implementation, the developed system enables adaptability across different environmental conditions, ensuring that selected hybrids perform well both in controlled and field environments.
[007] In another implementation, the developed system accelerates breeding cycles and improves precision, supporting agricultural goals like increased yield and stress resilience.
[008] The object of the invention is to determining parent plant species based on real-time data and historical data pertaining to plant growth under specified environmental conditions and soil conditions thus optimizing hybridization for increased yield.
[009] Another object of invention is to simplify and expedite the hybrid breeding process by identifying parent combinations with high heterosis potential, reducing labor, time, and resources.
[0010] Another object of invention is to predict hybrid yield potential (low, moderate, or high) based on the selected parent combinations and environmental factors, enhancing accuracy in breeding.
[0011] Another object of invention is to enable selection of parent combinations optimized for specific environmental conditions, whether rich or poor, improving hybrid adaptability.
[0012] Another object of invention is to reduce expenses associated with unsuccessful cross attempts by focusing breeding efforts on promising combinations pre-screened by ML.
BRIEF DESCRIPTION OF THE DRAWINGS
[0013] The foregoing detailed description of embodiments is better understood when read in conjunction with the attached drawing.
[0014] Figure 1 illustrates the structural block diagram about the implementation of the proposed invention .
DETAILED DESCRIPTION
[0015] The present invention introduces an intelligent system for selecting optimal parent plants for hybridization to enhance crop yield. Using environmental and growth sensors, along with an embedded machine learning algorithm, the system analyzes data on temperature, humidity, light, soil moisture, and plant growth. It enables precise identification of parent plants with high-yield potential, streamlining hybrid selection and improving agricultural productivity.
[0016] In another embodiment, the proposed system comprises of a sensor module (101), a central controller (102) with embedded machine learning algorithm (103), a communication module (104), a cloud server (105), a data centre (106) and a display device (108).
[0017] In another embodiment, the sensor module (101) comprises sensors for measuring environmental conditions (temperature, humidity, light, soil moisture) and plant growth metrics (optical sensor for chlorophyll detection, growth rate sensor using linear displacement). This module collects real-time data, which is crucial for determining optimal hybrid combinations.
[0018] In another embodiment, wherein the sensor module (101) includes temperature sensors capable of accurately measuring and logging ambient temperatures across a specified range relevant to plant growth, humidity sensors calibrated for precision under diverse atmospheric conditions, soil moisture sensors for monitoring water availability at root zones critical for plant health, light sensors for detecting variations in light intensity impacting photosynthesis rates, and optical sensors for measuring chlorophyll content, providing a proxy for plant health and photosynthetic efficiency, all of which collectively offer a holistic understanding of the growth environment.
[0019] In another embodiment, the sensor module (101) includes temperature, humidity, soil moisture sensors, optical imaging devices, tilt sensors, linear displacement sensors, and thermal imaging devices. It measures temperature, humidity, stem thickness, and plant height, and detects leaf rolling as an indicator of stress, especially under drought conditions. These devices help monitor plant health and productivity.
[0020] In another embodiment, plant growth metrics such as chlorophyll content, relative water content, leaf rolling, flag leaf area, flag leaf angle, number of tillers, stem thickness and plant height are measured using combination of sensors and optical devices which are included in the sensor module (101)
[0021] In another embodiment, using weight sensors and optical imaging, the system calculates yield metrics in grams to evaluate productive output.
[0022] In another embodiment, recorded time-lapse imaging is employed for indicating the plant's reproductive timing in terms of recording days to 50% flowering.
[0023] In another embodiment, panicle length, total panicle count, grain yield per panicle, and grain volume are measured with optical length tools and imaging software to analyze reproductive traits.
[0024] In another embodiment, laser sneors or ultrsonics sensors can be employed to measure plant height and growth without making physical contact with the plant.
[0025] In another embodiment, the central controller (102) with embedded machine learning algorithm (103) is the core processing unit of the system which is responsible for analyzing incoming data from the sensor module (101) and historical datasets available at the data centre (106) through the cloud server (105). The embedded machine learning algorithm (103) here evaluates which plant species combinations yield the best hybrid results based on current environmental inputs and previous performance records.
[0026] In another embodiment, the communication module (104) interfaces the sensor module (101), data centre (106), and cloud server (105), allowing seamless data transfer. In addition, this communication module (104) facilitates real-time updates from the sensor module (101)to the central controller (102) and also allows breeders to input data offline (107), if needed.
[0027] In another embodiment, support both online and offline data input by breeders, enabling data entry in disconnected environments, and synchronize data between the sensor module (101), the central controller (102), and the cloud server (105) based data centre 106), ensuring continuous availability of the most recent data for analysis, thus enhancing system resilience and functionality in diverse operational settings.
[0028] In another embodiment, the breeder inputs the grain yield per panicle, grain volume total panicle number per plant, grain mass (1000 grain weight), grain number per panicle to cloud based the data centre (106)
[0029] In another embodiment, the communication module (104) provides a range of connectivity including primary connectivity comprising Wi-Fi or LAN in greenhouse or controlled environments for high-speed, reliable communication, field communication including Zigbee or LoRa in open-field settings for wide-area, low-power connectivity, with cellular fallback in regions with cellular coverage and data synching, For remote locations without real-time connectivity, Bluetooth or a local storage option could allow for periodic data syncs when a connection becomes available.
[0030] In another embodiment, the cloud server (105) and the data centre (106) provides storage for large datasets, including historical environmental data and past hybridization results, accessible by the central controller (102) for improved prediction accuracy.
[0031] In another embodiment, the display device (108) displays the recommendations and analytics for breeder access, indicating parent combinations with predicted yield levels as established by the the central controller (102) with embedded machine learning algorithm (103) based on real time sensor input available from the sensor module (101) and the cloud (105) based data centre (106).
[0032] In another embodiment, the machine learning component embedded in the central controller employs a Support Vector Machine (SVM) model for classifying and ranking parent plant combinations based on predicted hybrid yield potential. The SVM model functions by identifying a hyperplane that optimally separates parent combinations into categories of low, moderate, or high yield. To handle complex, non-linear relationships within the input data, the SVM utilizes a radial basis function (RBF), which transforms the data into a higher-dimensional feature space. This transformation allows the model to maximize the separation margin between classes, enhancing the accuracy and robustness of predictions. Hyperparameters, such as regularization and kernel-specific parameters, are optimized to ensure the model generalizes well to unseen data, thereby improving yield prediction for various environmental conditions
[0033] In another embodiment, the method of selecting the parent plan/ crop species and predicating the plant/ crop yield is as follows:
a. Data Collection and Transmission: The sensor module (101) gathers real-time data on environmental conditions such as temperature, humidity, luminescence, soil moisture, wind speed and plant growth parameters based on chlorophyll detection and height rate method based on linear displacement. This real time data is transmitted to the central controller (102) via the communication module (104).
b. Processing and analysis: The central controller (102), equipped with the machine learning algorithm (103), integrates real-time sensor data (101) with historical information stored in the data centre (106) and cloud server (105). This analysis identifies patterns in parent plant performance under various conditions and predicts the most promising hybrid combinations.
c. Prediction and Recommendation: Based on data analysis performed by the central controller (102) on the real time data and the historical data, the central controller (102) provides breeders with recommendations, categorizing yield potential as low, moderate, or high. This process narrows down cross combinations, reducing labor and resources associated with unsuccessful breeding attempts.
d. Offline Data Input: Breeders can manually input environmental and growth data through the communication module (104) when offline, ensuring continuous functionality even without real-time connectivity. Along with this the breeders, can also input the pst hybridization results.
e. Adaptability and Continuous Improvement: The machine learning model (103) is designed to adapt to varying environmental conditions, continually improving its prediction accuracy as new data is collected, thereby refining hybrid recommendations over time.
[0034] In another embodiment, the proposed system provide higher accuracy in selecting parent plants, enhancing yield predictability.
[0035] In another embodiment, the proposed system significantly reduces the labor and resources required for cross attempts by focusing on the most promising hybrids.
[0036] In another embodiment, the proposed system adapts to different environmental conditions, allowing for hybrid recommendations that are robust in diverse settings.
[0037] In another embodiment, the proposed machine learning driven system insights shorten breeding timelines, enabling quicker adaptation to changing agricultural needs.
, Claims:1. A system for selecting optimal parent plants for hybridization aimed at increasing crop yield, comprising:
a. a sensor module (101), configured to continuously monitor and collect real time environmental, field conditions and plant growth data, including but not limited to temperature, humidity, soil moisture, light intensity, and plant growth metrics such as chlorophyll content, relative water content, leaf rolling, flag leaf area, flag leaf angle, number of tillers, stem thickness and plant height using a combination of sensors and optical devices.
b. a central controller (102), operatively connected to the sensor module (101), embedded with a machine learning algorithm (103) designed to process the real time incoming data from the sensor module (101), alongside historical data stored in a data center (106), to analyze the potential of various parent plant combinations for producing high-yield hybrids;
c. a communication module (104), interfaced with both the central controller (102) and a cloud server (105) based data center (106), configured to facilitate seamless bidirectional data transfer between the sensor module (101), the central controller (102), and the data centre (106), enabling real-time updates and offline data input functionality for uninterrupted operation in remote or disconnected environments;
d. the data center (106)and the cloud server (105), configured to store a large volume of historical environmental, genetic, and hybrid performance data; these data repositories provide a comprehensive data source that enhances the predictive accuracy of the machine learning algorithm (103) by integrating real-time sensor module (101) data and historical data inputs available at the data centre (106);
and
e. a display device (108) interfaced with the central controller (102), configured to visually present hybridization recommendations, analytics, and yield potential predictions, categorized into low, moderate, and high yield levels based on the embedded machine learnig algorithm(103) analysis, to guide breeders in selecting the most suitable parent plants for optimal hybridization outcomes.
2. The system as claimed in claim 1, wherein the sensor module (101) includes
a. Temperature sensor for measuring the ambient temperature , humidity sensor for measuring the ambient humidity , soil moisture sensor to determine the relative water content,
b. optical imaging devices for capturing measurements such as, leaf rolling, flag leaf area, flag leaf angle, number of tillers, stem thickness and plant height;
c. tilt sensors or inclination sensors for monitoring flag leaf angle;
d. linear displacement sensors for non-contact measurements of stem thickness and plant height;
and
e. thermal imaging devices for detecting leaf rolling as an indicator of stress, especially under drought conditions.
3. The machine learning algorithm as claimed in Claim 1, wherein, the machine learning algorithm (103) embedded in the central controller (102) is configured to:
a. integrate real-time sensor module (101) data with historical datasets in the cloud (105) based data centre (106) to identify patterns and correlations between specific parent plant combinations and observed yield outcomes under similar environmental conditions;
b. continuously learn and update its predictive model based on new data inputs, thereby enhancing its accuracy and adaptability to evolving environmental conditions and plant responses, and
c. classify potential hybrid yield levels into predefined categories of low, moderate, high based on an analysis of plant growth metrics and environmental compatibility, allowing breeders to prioritize cross combinations with the highest predicted yield potential.
4. The communication module as claimed in claim1, wherein, the communication module (104) is designed to
a. support both online and offline data input by breeders, enabling data entry in disconnected environments, and
b. synchronize data between the sensor module (101), the central controller (102), and the cloud server (105) based data centre 106), ensuring continuous availability of the most recent data for analysis, thus enhancing system resilience and functionality in diverse operational settings.
5. A method of selecting optimal parent plants for hybridization aimed at increasing crop yield comprising the steps of:
a. Data Collection and Transmission: The sensor module (101) gathers real-time data on environmental conditions such as temperature, humidity, luminescence, soil moisture, wind speed and plant growth parameters based on chlorophyll detection and height rate method based on linear displacement. This real time data is transmitted to the central controller (102) via the communication module (104).
b. Processing and analysis: The central controller (102), equipped with the machine learning algorithm (103), integrates real-time sensor data (101) with historical information stored in the data centre (106) and cloud server (105). This analysis identifies patterns in parent plant performance under various conditions and predicts the most promising hybrid combinations.
c. Prediction and Recommendation: Based on data analysis performed by the central controller (102) on the real time data and the historical data, the central controller (102) provides breeders with recommendations, categorizing yield potential as low, moderate, or high. This process narrows down cross combinations, reducing labor and resources associated with unsuccessful breeding attempts.
d. Offline Data Input: Breeders can manually input environmental and growth data through the communication module (104) when offline, ensuring continuous functionality even without real-time connectivity. Along with this the breeders, can also input the pst hybridization results.
e. Adaptability and Continuous Improvement: The machine learning model (103) is designed to adapt to varying environmental conditions, continually improving its prediction accuracy as new data is collected, thereby refining hybrid recommendations over time.
Documents
Name | Date |
---|---|
202411087975-COMPLETE SPECIFICATION [14-11-2024(online)].pdf | 14/11/2024 |
202411087975-DECLARATION OF INVENTORSHIP (FORM 5) [14-11-2024(online)].pdf | 14/11/2024 |
202411087975-DRAWINGS [14-11-2024(online)].pdf | 14/11/2024 |
202411087975-FIGURE OF ABSTRACT [14-11-2024(online)].pdf | 14/11/2024 |
202411087975-FORM 1 [14-11-2024(online)].pdf | 14/11/2024 |
202411087975-POWER OF AUTHORITY [14-11-2024(online)].pdf | 14/11/2024 |
202411087975-PROOF OF RIGHT [14-11-2024(online)].pdf | 14/11/2024 |
202411087975-REQUEST FOR EARLY PUBLICATION(FORM-9) [14-11-2024(online)].pdf | 14/11/2024 |
Talk To Experts
Calculators
Downloads
By continuing past this page, you agree to our Terms of Service,, Cookie Policy, Privacy Policy and Refund Policy © - Uber9 Business Process Services Private Limited. All rights reserved.
Uber9 Business Process Services Private Limited, CIN - U74900TN2014PTC098414, GSTIN - 33AABCU7650C1ZM, Registered Office Address - F-97, Newry Shreya Apartments Anna Nagar East, Chennai, Tamil Nadu 600102, India.
Please note that we are a facilitating platform enabling access to reliable professionals. We are not a law firm and do not provide legal services ourselves. The information on this website is for the purpose of knowledge only and should not be relied upon as legal advice or opinion.