LU505298B1

LU505298B1 - A method, system, device, and storage medium for monitoring the growth conditions of tea plant

Info

Publication number: LU505298B1
Application number: LU505298A
Authority: LU
Inventors: Ze Xu; Haibin Yang
Original assignee: Chongqing Acad Of Agricultural Science
Priority date: 2023-10-17
Filing date: 2023-10-17
Publication date: 2024-04-17

Abstract

This invention discloses a method, system, device, and storage medium for monitoring the growth conditions of tea plants, related to the field of tea plant growth monitoring technology. The method includes: Collecting monitoring data for tea plants; the monitoring data for tea plants includes video images of tea plant growth, spectral data, soil moisture content, and meteorological environmental data; preprocessing the tea plant monitoring data to obtain a monitoring database; inputting the data from the monitoring database into a tea fresh leaf yield image estimation model to predict tea plant yield; the tea fresh leaf yield image estimation model is constructed based on a network model for identifying tea leaf buds and a regression model; the tea leaf bud identification network model is constructed using deep learning networks; based on the tea plant yield and the monitoring database, automatically fertilizing the tea plants.

Description

DESCRIPTION

A METHOD, SYSTEM, DEVICE, AND STORAGE MEDIUM FOR

MONITORING THE GROWTH CONDITIONS OF TEA PLANT

TECHNICAL FIELD

This invention pertains to the field of tea plant growth monitoring technology, specifically focusing on a method, system, device, and storage medium for monitoring the growth conditions of tea plants.

BACKGROUND

The traditional tea garden management relies primarily on manual labor. Estimating tea production, determining the picking period, and monitoring the growth of tea plants still largely depend on accumulated experience. This approach is time-consuming, labor-intensive, and lacks precision, making it unable to provide accurate monitoring of the tea plant's growth status.

SUMMARY

The purpose of this invention is to provide a method, system, equipment, and storage medium for monitoring the growth status of tea plants, with the aim of improving the accuracy of monitoring their growth condition.

To achieve the above objective, this invention offers the following solutions:

A method for monitoring the growth conditions of tea plants, comprising:

Collecting monitoring data for tea plants. The monitoring data for tea plants includes video images of tea plant growth, spectral data, soil moisture content, and meteorological environmental data;

Preprocessing the tea plant monitoring data to obtain a monitoring database;

Inputting the data from the monitoring database into a tea fresh leaf yield image LU505298 estimation model to predict tea plant yield. The tea fresh leaf yield image estimation model is constructed based on a network model for identifying tea leaf buds and a regression model;

The tea leaf bud identification network model is constructed using deep learning networks;

Automatically fertilizing the tea plants based on the tea plant yield and the monitoring database.

Optionally, preprocess the tea plant monitoring data and obtain a monitoring database, it includes the following steps: Removing duplicates, formatting, and filling in missing values in the tea plant monitoring data to obtain complete data; Converting the complete data into a suitable data format to obtain valid data; Using an ETL tool to transform the valid data into a structured format, resulting in a collection database; performing data integration and establishing data relationships within the collection database to obtain the monitoring database.

Optionally, the training process of the tea leaf bud target recognition network model is as follows: Obtain training data; the training data includes tea leaf bud images along with their corresponding bud counts; build a YOLOv3 deep learning network model; input the training data into the YOLOv3 deep learning network model and train it based on the loss function; the trained YOLOv3 deep learning network model is then designated as the tea leaf bud target recognition network model.

Optionally, the tea fresh leaf yield image estimation model is represented as follows:

YF = NFxSLW/100

Yg = YFxAgxC/AF

In the formula for the tea fresh leaf yield image estimation model: Yr represents the predicted tea plant yield within the video image. Nr represents the number of tea leaf buds in the video image. SLW represents the weight of one hundred tea leaf buds (typically expressed as "per one hundred buds"). Y; represents the estimated yield of tea leaf buds on the tea plant.

Ag represents the area of the tea plant cultivation. C represents the tea plant coverage percentage. Ar represents the area of the captured image.

Optionally, automatically fertilize tea plants based on the tea plant yield and the monitoring database, it includes the following steps: generate fertilization control commands LU505298 based on the tea plant yield and a fertilization decision algorithm; determine the fertilization plan by correlating the fertilization control commands with associated data from the monitoring database; implement the established fertilization plan by controlling integrated water and fertilizer equipment to automatically apply fertilizers to the tea plants.

Optionally, data collection module; which is used for collecting monitoring data for tea plants. The tea plant monitoring data includes video images of tea plant growth, spectral data, soil moisture content, and meteorological environmental data; preprocessing module: which is used for preprocessing the tea plant monitoring data to obtain a monitoring database. Yield estimation module which is used to input the data from the monitoring database into a tea fresh leaf yield image estimation model to predict tea plant yield. The tea fresh leaf yield image estimation model is constructed based on a network model for identifying tea leaf buds and a regression model. The tea leaf bud identification network model is constructed using deep learning networks. Automatic fertilization based on the tea plant yield and the monitoring database.

According to the described tea tree yield and the set fertilization plan, automatic fertilization is applied to the tea trees.

This invention also provides an electronic device comprising memory and a processor, where the memory is used to store a computer program, and the processor executes the computer program to enable the electronic device to perform the monitoring method for the tea tree growth conditions as described above.

Furthermore, this invention provides a computer-readable storage medium that stores a computer program, which, when executed by a processor, implements the monitoring method for the tea tree growth conditions as described above.

According to specific embodiments provided by this invention, the following technical advantages are disclosed:

This invention discloses a method, system, device, and storage medium for monitoring the growth conditions of tea trees. The method includes collecting tea tree monitoring data, which comprises video images of tea tree growth, spectral data, soil moisture content, and meteorological environmental data. The monitoring data is preprocessed to obtain a monitoring database. The data from the monitoring database is input into a tea fresh leaf yield LU505298 image estimation model to predict tea tree yield. The tea fresh leaf yield image estimation model is constructed based on a target identification neural network model and a regression model for tea leaf buds. The tea leaf bud target identification neural network model is built using deep learning networks. Based on the tea tree yield and the monitoring database, automatic fertilization is applied to the tea trees. This invention can improve the accuracy of monitoring the growth conditions of tea trees.

BRIEF DESCRIPTION OF THE FIGURES

In order to explain the embodiments of the present invention or the technical scheme in the prior art more clearly, the drawings needed in the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention. For ordinary people in the field, other drawings can be obtained according to these drawings without paying creative labor. Brief description of the drawings

Fig. 1 is a schematic flow chart of the monitoring method of tea tree growth condition of the present invention.

DETAILED DESCRIPTION OF THE INVENTION

In the following, the technical scheme in the embodiment of the invention will be clearly and completely described with reference to the attached drawings. Obviously, the described embodiment is only a part of the embodiment of the invention, but not the whole embodiment.

Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in the field without creative labor belong to the scope of protection of the present invention.

The purpose of this invention is to provide a method, system, device, and storage medium for monitoring the growth status of tea plants, aiming to improve the accuracy of monitoring the growth conditions of tea plants.

To make the objectives, features, and advantages of this invention more apparent and understandable, the following detailed description is provided in conjunction with the drawings and specific embodiments.

As shown in Figure 1, this invention provides a method for monitoring the growth status of tea plants, comprising the following steps:

Step 100: Collect tea plant monitoring data. The tea plant monitoring data includes video LU505298 images of tea plant growth, spectral data, soil moisture content, and meteorological environmental data.

Step 200: Preprocess the tea plant monitoring data to obtain a monitoring database.

Step 300: Input the data from the monitoring database into a tea fresh leaf yield image estimation model to predict the tea plant yield. The tea fresh leaf yield image estimation model is constructed based on a tea leaf bud target identification neural network model and a regression model. The tea leaf bud target identification neural network model is built using deep learning networks.

Step 400: Based on the tea plant yield and the monitoring database, automatically fertilize the tea plants.

As a specific embodiment of Step 200, it includes:

Removing duplicates, formatting, and filling missing values in the tea plant monitoring data to obtain complete data.

Performing data format conversion on the complete data to obtain valid data.

Using an ETL (Extract, Transform, Load) tool to perform collection format conversion on the valid data to obtain a collection database.

Integrating and relating data in the collection database to obtain the monitoring database.

The training process of the tea leaf bud target identification neural network model is as follows:

Obtain training data, which includes tea leaf bud images and the corresponding number of leaf buds.

Construct a YOLOv3 deep learning network model.

Input the training data into the YOLOv3 deep learning network model and train it based on the loss function. The trained YOLOv3 deep learning network model is then designated as the tea leaf bud target identification neural network model.

The image yield estimation model of fresh tea leaf yield is expressed as:

Yr = NrxSLW/100

Yo = YrxAg<C/AF wherein: Yr represents the predicted tea plant yield within the video image. Nr represents the number of tea leaf buds in the video image. SLW represents the weight of one hundred tea LU505298 leaf buds . Y; represents the estimated yield of tea leaf buds on the tea plant. Ag represents the area of the tea plant cultivation. C represents the tea plant coverage percentage. AF represents the area of the captured image.

As a specific implementation of step 400, it includes the following:

Outputting fertilization control commands based on the tea tree yield and fertilization decision algorithm.Determining a set fertilization plan based on the fertilization control commands and associated data from the monitoring database.

Controlling the water and fertilizer integration equipment to automatically fertilize the tea trees according to the set fertilization plan.

Based on the above technical solution, the following implementation example 1s provided:

To implement the above plan, follow these steps:

In the first step, use a tea tree growth monitoring device to intelligently collect atmospheric, soil environment, soil nutrient, and tea tree growth image data in the tea plantation.

In the second step, clean, parse, and transform the raw data collected by the equipment, outputting it in a database-standard storage format and meeting system requirements, and construct a database.

In this implementation example, various data processing tools and script writing techniques are used. Data cleaning involves scripting to remove duplicates, format data, fill missing values, etc, ensuring data accuracy and completeness. Data parsing is done using scripts to transform data into an actionable format, such as converting JSON data into CSV or

XML format for later data analysis. Data transformation involves using ETL (Extract,

Transform, Load) tools to convert data formats or govern data, such as converting data from

MySQL to MongoDB databases. Data integration uses scripts to combine data from multiple sources, such as integrating image data, spectral data, soil moisture data, environmental monitoring data from various smart sensors. Data analysis involves scripting to perform statistical analysis, modeling, and other operations to match correlated data related to tea tree growth, ultimately deriving conclusions about the tea tree's growth status.

The third step involves preprocessing the data using deep learning technology, LU505298 monitoring sensor status, and optimizing the data collection frequency.

In this implementation example, deep learning technology utilizes a neural network structure comprising multiple layers of neurons. Each layer weights and processes input signals before passing them to the next layer for further processing. Deep learning technology trains the network using a large amount of data and backpropagation algorithms, enabling the network to gradually learn abstract features and automatically extract useful features. This allows it to efficiently perform tasks such as image recognition.

The fourth step involves real-time prediction of tea leaf yield, diagnosis of tea leaf quality, and soil nutrient assessment using embedded models: the tea leaf yield estimation model, tea leaf quality analysis model, and soil nutrient diagnostic model.

In this implementation example, the tea leaf yield estimation model calculates tea leaf yield by identifying tea leaf buds (one bud and one leaf), bud weight, yield coefficient, sample area, and coverage. 1. Utilizing deep learning technology to construct a tea leaf bud target recognition neural network model. 2. Employing an untrained test dataset for bud recognition. 3. Evaluating the model using accuracy and recall, with specific evaluation formulas as follows:

P = TP/(TP+FP)

R = TP/(TP+FN)

F1 = (2PxR)/(P+R) wherein,

P —— accuracy of bud identification; TP —— Correctly identify the number of all kinds of tea buds; FP —— Misidentifying the number of all kinds of tea buds; FN —— The number of all kinds of buds is not recognized; R-recall rate; Fl —— Harmonic average of accuracy and recall. 4) The accuracy of model evaluation should be greater than 85%, and the recall rate should be 0-1.

5) Estimating soil nutrient content based on the relationship between soil electrical LU505298 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 correlated with soil nutrient content. 6) The specific model for calculating soil nutrient content using electrical conductivity can vary depending on the actual circumstances. In this implementation example, both simple linear regression models and multiple linear regression models are used:

Simple linear regression model: nutrient content = ax conductivity+b; ; Among them, a and b are regression coefficients, which need to be trained and determined according to actual data. Multiple linear regression model: nutrient content = bO0+b1x conductivity 1+b2x conductivity 2+...; Among them, b0, bl, b2, ... are regression coefficients, and conductivity 1, conductivity 2, ... are selected conductivity parameters related to nutrient content. 7) Calculate and estimate the yield of fresh tea leaves by the number of buds and the weight of 100 buds obtained by the automatic bud identification model, and the formula is as follows:

Yr = NrxSLW/100

Yo = YrxAg<C/AF

Wherein, Yris the predicted tea yield in the video image; Nr is the number of buds in the video image; SLW is 100 bud weight; Y, estimates the yield of tea buds; Ag is the planting area of tea trees; C is the coverage of tea trees; Ar is the area of the captured image.

Step 5, based on the diagnostic results mentioned above, establish a precise management plan for tea picking and tea plantation fertilization. Part of this plan will be implemented in real-time through a mobile application (app) for manual intervention, while another part will be automated through the control system for automatic fertilization in the tea plantation.

An embodiment of automatic fertilization control is as follows: 1. Various sensors installed in the tea plantation, such as light sensors, temperature and humidity sensors, soil moisture sensors, high-definition cameras, etc., collect real-time data on tea tree growth, environmental climate in the tea plantation, soil temperature and humidity, soil pH, and soil nitrogen, phosphorus, and potassium content. This data is transmitted to the central control system through a wireless network.

2. After receiving the data collected by the sensors, the central control system uses LU505298 advanced data processing and analysis algorithms, such as regression models, for real-time analysis of tea tree growth, tea plantation climate, and soil nutrients. For example, it analyzes tea leaf color, leaf surface temperature, and morphology using image recognition algorithms and monitors and evaluates tea plantation climate and soil nutrients online using data mining algorithms. 3. Based on the results of data analysis, the central control system, following predefined fertilization rules and the needs of the tea trees, triggers the fertilization system automatically under appropriate conditions. The fertilization decision algorithm takes into account factors such as the tea tree's growth status, tea plantation climate, and soil nutrients to determine the appropriate fertilization plan and dosage. 4. The automatic fertilization system consists of water and fertilizer integration equipment and a control device. Based on the results of the fertilization decision algorithm, the control device sends commands to the water and fertilizer integration equipment to carry out precise fertilization operations. The water and fertilizer integration equipment adjusts the water-fertilizer ratio and fertilization timing as needed to ensure that the tea trees receive an appropriate supply of water and nutrients.

This embodiment has the following beneficial effects:

This embodiment provides an intelligent monitoring and diagnostic method for the growth status of tea trees through video, spectrometry, soil moisture, and meteorological data.

This method can monitor the growth of tea trees in the tea plantation in real-time, promptly detect changes in the condition of the tea trees, prevent and resolve issues that may arise during the growth process, improve tea tree yield and quality, and provide accurate data support for tea production.

Additionally, the invention also provides a monitoring system for the growth status of tea trees, including:

A data collection module for gathering tea tree monitoring data, which includes video images of tea tree growth, spectral data, soil moisture, and meteorological environmental data.

A preprocessing module for preprocessing the tea tree monitoring data to obtain a monitoring database.

A yield estimation module that inputs data from the monitoring database into the tea leaf LU505298 yield estimation model to predict tea tree yield. The tea leaf yield estimation model is constructed based on a tea leaf bud target recognition neural network model and a regression model. The tea leaf bud target recognition neural network model is constructed using deep learning networks.

Automatic fertilization of tea trees based on the tea tree yield and the set fertilization plan.

The invention also provides an electronic device, including memory and a processor. The memory stores a computer program, and the processor executes the computer program to perform the monitoring method for the growth status of tea trees as described above.

The invention also provides a computer-readable storage medium that stores a computer program. When executed by a processor, the computer program implements the monitoring method for the growth status of tea trees as described above.

The various embodiments in this specification are described progressively, with each embodiment highlighting the differences from other embodiments. Similar or identical parts between the various embodiments can be referenced accordingly.

Specific examples have been used in this document to elucidate the principles and implementation of the invention. The descriptions of the above embodiments are provided to facilitate an understanding of the core concepts of the invention. At the same time, for those skilled in the art, changes may occur in the specific implementation methods and application scope based on the principles of the invention. In summary, the content of this specification should not be construed as limiting the invention.

Claims

CLAIMS LU505298

1. A method for monitoring the growth conditions of tea plants, comprising: collecting monitoring data for tea plants; the monitoring data for tea plants includes video images of tea plant growth, spectral data, soil moisture content, and meteorological environmental data; preprocessing the tea plant monitoring data to obtain a monitoring database; inputting the data from the monitoring database into a tea fresh leaf yield image estimation model to predict tea plant yield; the tea fresh leaf yield image estimation model 1s constructed based on a network model for identifying tea leaf buds and a regression model; the tea leaf bud identification network model is constructed using deep learning networks: automatically fertilizing the tea plants based on the tea plant yield and the monitoring database.

2. The method for monitoring the growth conditions of tea plants, as claimed in claim 1, preprocess the tea plant monitoring data and obtain a monitoring database, it includes the following steps: removing duplicates, formatting, and filling in missing values in the tea plant monitoring data to obtain complete data; converting the complete data into a suitable data format to obtain valid data; using an ETL tool to transform the valid data into a structured format, resulting in a collection database; performing data integration and establishing data relationships within the collection database to obtain the monitoring database.

3. The method for monitoring the growth conditions of tea plants, as claimed in claim 1, the training process of the tea leaf bud target recognition network model is as follows: obtain training data; the training data includes tea leaf bud images along with their corresponding bud counts; build a YOLOv3 deep learning network model; input the training data into the YOLOv3 deep learning network model and train it based on the loss function; the trained YOLOv3 deep learning network model is then designated as the tea leaf bud target recognition network model.

4. The method for monitoring the growth conditions of tea plants, as claimed in claim 1, LU505298 the tea fresh leaf yield image estimation model is represented as follows: YF = NFxSLW/100 Yg = YFxAgxC/AF in the formula for the tea fresh leaf yield image estimation model: Yr represents the predicted tea plant yield within the video image; Nr represents the number of tea leaf buds in the video image; SLW represents the weight of one hundred tea leaf buds; Yz represents the estimated yield of tea leaf buds on the tea plant; As represents the area of the tea plant cultivation; C represents the tea plant coverage percentage; Ar represents the area of the captured image.

5. The method for monitoring the growth conditions of tea plants, as claimed in claim 1, automatically fertilize tea plants based on the tea plant yield and the monitoring database, it includes the following steps: generate fertilization control commands based on the tea plant yield and a fertilization decision algorithm; determine the fertilization plan by correlating the fertilization control commands with associated data from the monitoring database; implement the established fertilization plan by controlling integrated water and fertilizer equipment to automatically apply fertilizers to the tea plants.

6. A monitoring system for the growth conditions of tea plants, characterized by: data collection module; which is used for collecting monitoring data for tea plants; the tea plant monitoring data includes video images of tea plant growth, spectral data, soil moisture content, and meteorological environmental data; preprocessing module: which is used for preprocessing the tea plant monitoring data to obtain a monitoring database; yield estimation module, which is used to input the data from the monitoring database into a tea fresh leaf yield image estimation model to predict tea plant yield; the tea fresh leaf yield image estimation model is constructed based on a network model for identifying tea leaf buds and a regression model; the tea leaf bud identification network model is constructed using deep learning networks; automatic fertilization based on the tea plant yield and the monitoring database.

7. An electronic device, characterized by including a storage unit and a processor, LU505298 wherein the storage unit is used for storing a computer program, and the processor executes the computer program to enable the electronic device to perform the tea plant growth monitoring method as described in any one of claims 1 to 5.

8. À computer-readable storage medium, characterized by storing a computer program, wherein the computer program, when executed by a processor, implements the tea plant growth monitoring method as described in any one of claims 1 to 5.