A method, system, device and storage medium for monitoring the growth condition of tea trees

Abstract

The application discloses a tea tree growth condition monitoring method, system and device and a storage medium, and relates to the technical field of tea tree growth monitoring. The method comprises the following steps: collecting tea tree monitoring data; the tea tree monitoring data comprises video images, spectrum data, soil moisture conditions and meteorological environment data of tea tree growth; the tea tree monitoring data is preprocessed to obtain a monitoring database; data in the monitoring database is input into a tea fresh leaf yield image yield estimation model to predict tea tree yield; the tea fresh leaf yield image yield estimation model is constructed according to a tea leaf bud target recognition network model and a regression model; the tea leaf bud target recognition network model is constructed according to a deep learning network; and automatic fertilization is performed on the tea tree according to the tea tree yield and the monitoring database. The application can improve the monitoring precision of the tea tree growth condition.

Description

Technical Field

[0001] This invention relates to the field of tea tree growth monitoring technology, and in particular to a method, system, device and storage medium for monitoring the growth status of tea trees. Background Technology

[0002] Traditional tea garden management relies mainly on manual labor. The estimation of tea yield, the determination of the picking period, and the monitoring and management of tea tree growth are still based on accumulated experience. This method is time-consuming, labor-intensive, and not very accurate, and cannot accurately monitor the growth of tea trees. Summary of the Invention

[0003] The purpose of this invention is to provide a method, system, device, and storage medium for monitoring the growth status of tea trees, which can improve the accuracy of monitoring the growth status of tea trees.

[0004] To achieve the above objectives, the present invention provides the following solution:

[0005] A method for monitoring the growth status of tea trees, comprising:

[0006] Collect tea tree monitoring data; the tea tree monitoring data includes video images of tea tree growth, spectral data, soil moisture and meteorological environmental data;

[0007] The tea tree monitoring data is preprocessed to obtain a monitoring database;

[0008] The data from the monitoring database is input into the tea leaf yield image estimation model to predict the tea tree yield; the tea leaf yield image estimation model is constructed based on the tea bud target recognition network model and regression model; the tea bud target recognition network model is constructed based on a deep learning network.

[0009] The tea trees are automatically fertilized based on the tea tree yield and the monitoring database.

[0010] Optionally, the tea tree monitoring data is preprocessed to obtain a monitoring database, specifically including:

[0011] The tea tree monitoring data was deduplicated, formatted, and missing values ​​were filled to obtain complete data;

[0012] The complete data is converted to a different format to obtain valid data.

[0013] Using an ETL tool, the valid data is converted into a collection format to obtain a collection database;

[0014] The data in the aforementioned collection database are integrated and correlated to obtain a monitoring database.

[0015] Optionally, the training process of the tea bud target recognition network model is as follows:

[0016] Acquire training data; the training data includes images of tea buds and the corresponding number of buds.

[0017] Construct a YOLOv3 deep learning network model;

[0018] The training data is input into the YOLOv3 deep learning network model, and the model is trained according to the loss function. The trained YOLOv3 deep learning network model is then identified as the tea bud target recognition network model.

[0019] Optionally, the tea leaf yield estimation model is represented as follows:

[0020] Y F =N F ×SLW / 100

[0021] Y g =Y F ×A g ×C / A F

[0022] Among them, Y F N represents the predicted tea tree yield within the video image; F SLW represents the number of buds in the video image; Y represents the weight of 100 buds. g Estimating the yield of tea buds; A g C represents the tea tree planting area; A represents the tea tree coverage. F The area of ​​the captured image.

[0023] Optionally, based on the tea tree yield and the monitoring database, automatic fertilization of the tea trees is performed, specifically including:

[0024] Based on the tea tree yield and fertilization decision algorithm, output fertilization control commands;

[0025] The fertilization plan is determined based on the fertilization control command and the associated data in the monitoring database, and the water and fertilizer integration equipment is controlled to automatically fertilize the tea trees according to the fertilization plan.

[0026] This invention also provides a system for monitoring the growth status of tea trees, comprising:

[0027] The data acquisition module is used to collect tea tree monitoring data, which includes video images of tea tree growth, spectral data, soil moisture and meteorological environmental data.

[0028] The preprocessing module is used to preprocess the tea tree monitoring data to obtain a monitoring database;

[0029] The yield prediction module is used to input the data from the monitoring database into the tea fresh leaf yield image estimation model to predict the tea tree yield; the tea fresh leaf yield image estimation model is constructed based on the tea bud target recognition network model and regression model; the tea bud target recognition network model is constructed based on a deep learning network.

[0030] Based on the tea tree yield and the set fertilization plan, the tea trees are automatically fertilized.

[0031] The present invention also provides an electronic device, including a memory and a processor, wherein the memory is used to store a computer program, and the processor runs the computer program to enable the electronic device to perform the monitoring method for the growth status of tea trees as described above.

[0032] The present invention also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the method for monitoring the growth status of tea trees as described above.

[0033] According to specific embodiments provided by the present invention, the present invention discloses the following technical effects:

[0034] This invention discloses a method, system, device, and storage medium for monitoring the growth status of tea trees. The method includes collecting tea tree monitoring data, which includes video images of tea tree growth, spectral data, soil moisture data, and meteorological data. The monitoring data is preprocessed to obtain a monitoring database. Data from the monitoring database is input into a tea leaf yield estimation model to predict tea tree yield. The tea leaf yield estimation model is constructed based on a tea bud target recognition network model and a regression model. The tea bud target recognition network model is constructed based on a deep learning network. Based on the tea tree yield and the monitoring database, automatic fertilization is applied to the tea trees. This invention improves the accuracy of monitoring tea tree growth. Attached Figure Description

[0035] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0036] Figure 1 This is a flowchart illustrating the method for monitoring the growth status of tea trees according to the present invention. Detailed Implementation

[0037] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0038] The purpose of this invention is to provide a method, system, device, and storage medium for monitoring the growth status of tea trees, which can improve the accuracy of monitoring the growth status of tea trees.

[0039] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0040] like Figure 1 As shown, the present invention provides a method for monitoring the growth status of tea trees, including:

[0041] Step 100: Collect tea tree monitoring data; the tea tree monitoring data includes video images of tea tree growth, spectral data, soil moisture and meteorological environmental data;

[0042] Step 200: Preprocess the tea tree monitoring data to obtain a monitoring database;

[0043] Step 300: Input the data from the monitoring database into the tea leaf yield image estimation model to predict the tea tree yield; the tea leaf yield image estimation model is constructed based on the tea bud target recognition network model and regression model; the tea bud target recognition network model is constructed based on a deep learning network;

[0044] Step 400: Automatically fertilize the tea trees based on the tea tree yield and the monitoring database.

[0045] As one specific implementation of step 200, it includes:

[0046] The tea tree monitoring data is deduplicated, formatted, and missing values ​​are filled to obtain complete data; the complete data is then converted to a new data format to obtain valid data; using an ETL tool, the valid data is converted to a set format to obtain a set database; the data in the set database is then integrated and correlated with relevant data to obtain a monitoring database.

[0047] The training process of the tea bud target recognition network model is as follows:

[0048] Acquire training data; the training data includes images of tea buds and the corresponding number of buds.

[0049] Construct a YOLOv3 deep learning network model;

[0050] The training data is input into the YOLOv3 deep learning network model, and the model is trained according to the loss function. The trained YOLOv3 deep learning network model is then identified as the tea bud target recognition network model.

[0051] The tea leaf yield estimation model based on the image is represented as follows:

[0052] Y F =N F ×SLW / 100

[0053] Y g =Y F ×A g ×C / A F

[0054] Among them, Y F N represents the predicted tea tree yield within the video image; F SLW represents the number of buds in the video image; Y represents the weight of 100 buds. g Estimating the yield of tea buds; A g C represents the tea tree planting area; A represents the tea tree coverage. F The area of ​​the captured image.

[0055] As one specific implementation of step 400, it includes:

[0056] Based on the tea tree yield and fertilization decision algorithm, a fertilization control command is output; based on the fertilization control command and the associated data of the monitoring database, a fertilization plan is determined, and the integrated water and fertilizer equipment is controlled to automatically fertilize the tea trees according to the fertilization plan.

[0057] Based on the above technical solution, the following embodiments are provided.

[0058] The above plan will be implemented through the following steps:

[0059] The first step is to intelligently collect data on the atmospheric and soil environment, soil nutrients, and tea tree growth in the tea garden using a tea tree growth monitoring instrument.

[0060] The second step is to clean, parse, and transform the raw data collected by the equipment, outputting data in a database-standard storage format that meets system requirements, and then constructing the database.

[0061] In this embodiment, various data processing tools and scripting techniques are employed. Specifically: Data cleaning: Scripts are used to deduplicate, format, and fill in missing values ​​in sensor data, ensuring accuracy and completeness. Data parsing: Scripts are used to convert data into an operable format. For example, JSON data is converted to CSV or XML format, forming effective data for later analysis. Data conversion: ETL tools are used for data format conversion and data governance. For example, data is converted from a MySQL database to a MongoDB database. Data integration: Scripts are used to integrate data from multiple data sources. For example, image data, spectral data, soil moisture data, and environmental monitoring data from different sensing devices are integrated. Data analysis: Scripts are used to perform statistical analysis, modeling, and other operations on the data to match correlation data related to tea tree growth and derive conclusions on tea tree growth.

[0062] The third step involves using deep learning technology to preprocess the data, monitor the sensor status, and optimize the frequency of data acquisition.

[0063] In this embodiment, deep learning technology employs a neural network structure containing multiple layers of neurons. Each layer weights and processes the input signal before passing it to the next layer for further processing. Deep learning technology trains the network using a large amount of data and a backpropagation algorithm, enabling the network to gradually learn abstract features and automatically extract useful features, thereby achieving efficient image recognition and other tasks.

[0064] The fourth step involves using the embedded tea leaf yield estimation model, tea leaf quality analysis model, and soil nutrient diagnosis model to predict tea yield, diagnose tea quality, and assess soil nutrients in real time.

[0065] In this embodiment, the tea leaf yield estimation model calculates the tea leaf yield by identifying tea leaf buds (one bud and one leaf), the weight of 100 buds, the yield coefficient, the sample area, and the coverage.

[0066] 1) A network model for identifying tea buds was constructed using deep learning technology.

[0067] 2) Use an untrained test set for sprout identification.

[0068] 3) Accuracy and recall are used for model evaluation. The specific evaluation calculation formulas are as follows:

[0069] P = TP / (TP + FP)

[0070] R = TP / (TP + FN)

[0071] F1 = (2P × R) / (P + R)

[0072] In the formula,

[0073] P – Accuracy of sprout identification;

[0074] TP – Correctly identify the number of tea buds of various types;

[0075] FP – Incorrectly identifies the number of various tea buds;

[0076] FN – Unable to identify the number of various types of buds;

[0077] R – Recall rate;

[0078] F1 – Harmonic mean of precision and recall.

[0079] 4) The model evaluation accuracy should be greater than 85%, and the recall rate should be 0-1.

[0080] 5) Soil nutrient content can be estimated by using the relationship between soil electrical conductivity and soil nutrients. Soil electrical conductivity can reflect the concentration of dissolved salts and ions in the soil, and these dissolved salts and ions are often associated with the nutrient content in the soil.

[0081] 6) The specific model for calculating soil nutrients using electrical conductivity can vary depending on the actual situation. In this embodiment, a simple linear regression model and a multiple linear regression model are used:

[0082] Simple linear regression model: Nutrient content = a × conductivity + b;

[0083] Here, a and b are regression coefficients, which need to be trained and determined based on actual data.

[0084] Multiple linear regression model: Nutrient content = b0 + b1 × conductivity 1 + b2 × conductivity 2 + ...;

[0085] Where b0, b1, b2, ... are regression coefficients, and conductivity 1, conductivity 2, ... are selected conductivity parameters related to nutrient content.

[0086] 7) The yield of fresh tea leaves is estimated by calculating the number of tender buds and the weight of 100 buds obtained through the automatic bud recognition model, using the following formula:

[0087] Y F =N F ×SLW / 100

[0088] Y g =Y F ×A g ×C / A F

[0089] Among them, Y F N represents the predicted tea tree yield within the video image; F SLW represents the number of buds in the video image; Y represents the weight of 100 buds. g Estimating the yield of tea buds; A g C represents the tea tree planting area; A represents the tea tree coverage. F The area of ​​the captured image.

[0090] The fifth step, based on the above diagnostic results, is to establish a precise management plan for tea picking and tea garden fertilization. Part of the plan is implemented manually through real-time push notifications via a mini-program, while the other part is implemented automatically through an automatic control system to achieve automatic fertilization of the tea garden.

[0091] Example of automatic fertilization control:

[0092] 1. Various sensors installed in the tea garden (such as light sensors, temperature and humidity sensors, soil moisture sensors, high-definition cameras, etc.) are used to collect data on the growth status of tea trees, the climate of the tea garden environment and soil temperature and humidity, soil pH value and soil nitrogen, phosphorus and potassium content in real time, and transmit the data to the central control system through a wireless network.

[0093] 2. After receiving data collected by sensors, the central control system uses advanced data processing and analysis algorithms, such as regression models, to perform real-time analysis of tea tree growth, tea garden climate, and soil nutrients. For example, image recognition algorithms are used to analyze tea leaf color, leaf surface temperature, and morphology, while data mining algorithms are used to monitor and evaluate tea garden climate and soil nutrients online.

[0094] 3. Based on the data analysis results, the central control system will automatically trigger the fertilization system to apply fertilizer under suitable conditions, according to the set fertilization rules and the needs of the tea trees. The fertilization decision algorithm will comprehensively consider factors such as the growth status of the tea trees, the climate of the tea garden, and soil nutrients to determine the appropriate fertilization plan and amount.

[0095] 4. The automatic fertilization system includes an integrated water and fertilizer system and a control device. Based on the fertilization decision algorithm, the control device transmits commands to the integrated water and fertilizer system to achieve accurate fertilization. The integrated water and fertilizer system adjusts the water-fertilizer ratio and fertilization time as needed to ensure that the tea trees receive an adequate supply of water and nutrients.

[0096] This embodiment has the following beneficial effects:

[0097] This embodiment provides an intelligent monitoring and diagnosis method for tea tree growth using video, spectral analysis, soil moisture data, and meteorological conditions. This method can monitor the growth of tea trees in the tea garden in real time, promptly detect changes in the tea trees' condition, prevent and resolve problems that arise during the tea tree growth process, improve tea yield and quality, and provide accurate data support for tea production.

[0098] In addition, the present invention also provides a monitoring system for the growth status of tea trees, comprising:

[0099] The data acquisition module is used to collect tea tree monitoring data, which includes video images of tea tree growth, spectral data, soil moisture and meteorological environmental data.

[0100] The preprocessing module is used to preprocess the tea tree monitoring data to obtain a monitoring database;

[0101] The yield prediction module is used to input the data from the monitoring database into the tea fresh leaf yield image estimation model to predict the tea tree yield; the tea fresh leaf yield image estimation model is constructed based on the tea bud target recognition network model and regression model; the tea bud target recognition network model is constructed based on a deep learning network.

[0102] Based on the tea tree yield and the set fertilization plan, the tea trees are automatically fertilized.

[0103] The present invention also provides an electronic device, including a memory and a processor, wherein the memory is used to store a computer program, and the processor runs the computer program to enable the electronic device to perform the monitoring method for the growth status of tea trees as described above.

[0104] The present invention also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the method for monitoring the growth status of tea trees as described above.

[0105] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to each other.

[0106] This document uses specific examples to illustrate the principles and implementation methods of the present invention. The descriptions of the above embodiments are only for the purpose of helping to understand the core ideas of the present invention. Furthermore, those skilled in the art will recognize that, based on the ideas of the present invention, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A method for monitoring the growth status of tea trees, characterized in that, include: Collect tea tree monitoring data; The tea tree monitoring data includes video images of tea tree growth, spectral data, soil moisture, and meteorological environmental data; The tea tree monitoring data is preprocessed to obtain a monitoring database; The data from the monitoring database is input into the tea leaf yield image estimation model to predict the tea tree yield; the tea leaf yield image estimation model is constructed based on the tea bud target recognition network model and regression model; the tea bud target recognition network model is constructed based on a deep learning network. Based on the tea tree yield and the monitoring database, the tea trees are automatically fertilized; The training process of the tea bud target recognition network model is as follows: Acquire training data; the training data includes images of tea buds and the corresponding number of buds. Construct a YOLOv3 deep learning network model; The training data is input into the YOLOv3 deep learning network model and trained according to the loss function. The trained YOLOv3 deep learning network model is then identified as the tea bud target recognition network model. The tea leaf yield estimation model based on the image is represented as follows: AND F =N F ×SLW / 100 AND g And F ×A g ×C / A F Among them, Y F N represents the predicted tea tree yield within the video image; F SLW represents the number of buds in the video image; Y represents the weight of 100 buds. g Estimating the yield of tea buds; A g C represents the area planted with tea trees; A represents the coverage of tea trees; C represents the tea tree coverage. F The area of ​​the captured image.

2. The method for monitoring the growth status of tea trees according to claim 1, characterized in that, The tea tree monitoring data is preprocessed to obtain a monitoring database, specifically including: The tea tree monitoring data was deduplicated, formatted, and missing values ​​were filled to obtain complete data; The complete data is converted to a different format to obtain valid data. Using an ETL tool, the valid data is converted into a collection format to obtain a collection database; The data in the aforementioned collection database are integrated and correlated to obtain a monitoring database.

3. The method for monitoring the growth status of tea trees according to claim 1, characterized in that, Based on the tea tree yield and the monitoring database, automatic fertilization of the tea trees is performed, specifically including: Based on the tea tree yield and fertilization decision algorithm, output fertilization control commands; The fertilization plan is determined based on the fertilization control command and the associated data in the monitoring database, and the water and fertilizer integration equipment is controlled to automatically fertilize the tea trees according to the fertilization plan.

4. A system for monitoring the growth status of tea trees, using the method described in any one of claims 1-3, characterized in that, include: The data acquisition module is used to collect tea tree monitoring data, which includes video images of tea tree growth, spectral data, soil moisture and meteorological environmental data. The preprocessing module is used to preprocess the tea tree monitoring data to obtain a monitoring database; The yield prediction module is used to input the data from the monitoring database into the tea fresh leaf yield image estimation model to predict the tea tree yield; the tea fresh leaf yield image estimation model is constructed based on the tea bud target recognition network model and regression model; the tea bud target recognition network model is constructed based on a deep learning network. The tea trees are automatically fertilized based on the tea tree yield and the monitoring database.

5. An electronic device, characterized in that, The device includes a memory and a processor, the memory being used to store a computer program, and the processor running the computer program to enable the electronic device to perform the method for monitoring the growth status of tea trees according to any one of claims 1-3.

6. A computer-readable storage medium, characterized in that, It stores a computer program, which, when executed by a processor, implements a method for monitoring the growth status of tea trees as described in any one of claims 1-3.

Images