Plant growth state prediction method, device, equipment, medium and product

Abstract

The invention discloses a plant growth state prediction method and device, equipment, a medium and a product, and relates to the field of plant growth state prediction.The method comprises the steps that a first electric signal of a plant is obtained, and frequency domain analysis processing is conducted on the first electric signal to obtain a first energy diagram; establishing a data set according to the first energy diagram and a corresponding preset label, and training a preset classification model according to the data set to obtain a growth state prediction model; obtaining a second electric signal of the plant to be detected, and performing frequency domain analysis processing on the second electric signal to obtain a second energy diagram; and inputting the second energy diagram into a growth state prediction model to obtain a prediction result. According to the invention, the plant growth state can be accurately predicted.

Description

Technical Field

[0001] The present application relates to the field of plant growth status prediction, and in particular to a plant growth status prediction method, device, equipment, medium and product. Background Art

[0002] Predicting plant growth patterns is crucial in agricultural, ecological, and botanical research. As the global population grows and food security becomes increasingly important, as climate change intensifies the impact of ecosystems, accurately predicting plant growth patterns becomes even more important.

[0003] As environmental factors such as light intensity, temperature, and humidity change dynamically, a complex and closely linked relationship emerges between plant electrical signals and growth patterns. Changes in these environmental factors trigger a series of physiological and biochemical reactions within the plant, directly influencing its growth. These reactions often manifest themselves as changes in electrical signals. In recent years, scientists have attempted to predict plant growth patterns by studying the characteristics and conduction mechanisms of plant electrical signals.

[0004] The present application provides a plant growth status prediction method, device, medium and product, which can accurately predict the plant growth status. Summary of the Invention

[0005] In a first aspect, the present application provides a method for predicting plant growth status, comprising:

[0006] Acquire a first electrical signal of the plant, and perform frequency domain analysis on the first electrical signal to acquire a first energy map;

[0007] Establishing a data set according to the first energy map and the corresponding preset labels, and training a preset classification model according to the data set to obtain a growth status prediction model;

[0008] Acquire a second electrical signal of the plant to be tested, and perform the frequency domain analysis on the second electrical signal to acquire a second energy map;

[0009] The second energy map is input into the growth state prediction model to obtain a prediction result.

[0010] Optionally, the performing frequency domain analysis on the first electrical signal to obtain a first energy graph includes:

[0011] Preprocessing the first electrical signal to obtain a first preprocessed signal;

[0012] Obtaining a first reconstructed electrical signal by performing wavelet decomposition and noise elimination reconstruction on the first preprocessed signal;

[0013] Performing short-time Fourier transform on the first reconstructed electrical signal to obtain a first energy map.

[0014] Optionally, obtaining a first reconstructed electrical signal by performing wavelet decomposition and noise reduction reconstruction on the first preprocessed signal includes:

[0015] Performing three-level DWT wavelet decomposition on the first preprocessed signal to obtain first decomposition coefficients;

[0016] The first decomposition coefficients are reconstructed by noise elimination to obtain a first reconstructed electrical signal.

[0017] Optionally, the first decomposition coefficients include a first detail coefficient D1, a first detail coefficient D2, a first detail coefficient D3, and a first approximation coefficient A3, and the obtaining of a first reconstructed electrical signal by performing noise removal and reconstruction on the first decomposition coefficients includes:

[0018] Obtaining a high-frequency component estimation noise standard deviation according to the first detail coefficient D1;

[0019] Estimating the noise standard deviation of the high-frequency component and obtaining the optimal threshold value through a universal threshold formula;

[0020] Performing threshold denoising on the first detail coefficient D1, the first detail coefficient D2, and the first detail coefficient D3 using the optimal threshold to obtain a first denoised detail coefficient B1, the first denoised detail coefficient B2, and the first denoised detail coefficient B3;

[0021] The first noise removal detail coefficient B1, the first noise removal detail coefficient B2, the first noise removal detail coefficient B3 and the first approximation coefficient A3 are processed by wavelet inverse transformation to obtain the first reconstructed electrical signal.

[0022] Optionally, the preset classification model is a CNN-LSTM-Attention hybrid neural network model, and the CNN-LSTM-Attention includes a CNN feature extraction module, a Flatten function, an LSTM network, a fully connected layer and an Attention layer connected in sequence.

[0023] In a second aspect, the present application provides a plant growth status prediction device, comprising:

[0024] A first feature module is configured to obtain a first electrical signal of the plant, and perform frequency domain analysis on the first electrical signal to obtain a first energy graph;

[0025] a training module, configured to establish a data set based on the first energy map and corresponding preset labels, and train a preset classification model based on the data set to obtain a growth status prediction model;

[0026] A second feature module is configured to obtain a second electrical signal of the plant to be tested, and perform the frequency domain analysis on the second electrical signal to obtain a second energy map;

[0027] The prediction module inputs the second energy map into the growth state prediction model to obtain a prediction result.

[0028] In a third aspect, the present application provides a plant growth status prediction device, comprising an environment perception module, a processor module, and a display module; wherein:

[0029] The environmental sensing module is used to collect environmental parameters of plant growth and the first electrical signal of the plant; wherein the environmental parameters include soil moisture, CO2 concentration in the air, temperature and light intensity;

[0030] The processor module includes a plant growth state prediction device, and the plant growth state prediction device is used to process the second electrical signal according to the plant growth state prediction method to obtain a prediction result;

[0031] The display module is used to display the environmental parameters and the prediction results in real time.

[0032] In a fourth aspect, the present application provides a computer device comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of any one of the above-mentioned plant growth status prediction methods.

[0033] In a fifth aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of any one of the above-mentioned plant growth status prediction methods.

[0034] In a sixth aspect, the present application provides a computer program product, comprising a computer program, which, when executed by a processor, implements the steps of any one of the above-mentioned plant growth status prediction methods.

[0035] According to the specific embodiments provided in this application, this application discloses the following technical effects:

[0036] The present application provides a plant growth status prediction method, apparatus, device, medium, and product. The method obtains a first energy map by frequency-domain analysis of a first electrical signal. The energy map can reflect the energy distribution of the signal at different frequencies, thereby better reflecting the growth characteristics of the plant. A preset classification model is trained based on the first energy map to obtain a growth status prediction model. This growth status prediction model is used to predict the plant growth status, resulting in more accurate prediction results. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0038] Figure 1 A schematic diagram of a process for predicting plant growth status provided in one embodiment of the present application;

[0039] Figure 2 A schematic flow chart of a plant growth status prediction method provided in another embodiment of the present application;

[0040] Figure 3 for Figure 1 Detailed flow diagram of step 102;

[0041] Figure 4 A schematic diagram of a first energy diagram obtained according to the method of the present application;

[0042] Figure 5 for Figure 3 Detailed flow diagram of step 302;

[0043] Figure 6 A schematic diagram of the functional modules of a plant growth status prediction device provided in one embodiment of the present application;

[0044] Figure 7 A schematic diagram of the functional modules of a plant growth status prediction device provided in one embodiment of the present application;

[0045] Figure 8 A schematic diagram of detailed functional modules of a processor module provided in one embodiment of the present application;

[0046] Figure 9 A schematic diagram of the functional modules of a plant growth status prediction device provided in another embodiment of the present application;

[0047] Figure 10 This is a schematic diagram of the interaction between the remote processor module, the local processor module, the data transmission module and the data monitoring module in one embodiment of the present application;

[0048] Figure 11 A schematic diagram of detailed functional modules of an environment perception module provided in one embodiment of the present application;

[0049] Figure 12 A schematic diagram of the structure of a computer device provided in one embodiment of the present application. DETAILED DESCRIPTION

[0050] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0051] In order to make the above-mentioned purposes, features and advantages of the present application more obvious and easy to understand, the present application is further described in detail below with reference to the accompanying drawings and specific implementation methods.

[0052] In an exemplary embodiment, Figure 1 As shown, a plant growth status prediction method is provided. The method is executed by a computer device, specifically, a computer device such as a terminal or a server, or a terminal and a server. In the embodiment of the present application, the method includes the following steps 101 to 104. Among them:

[0053] Step 101: obtaining a first electrical signal of a plant, and performing frequency domain analysis on the first electrical signal to obtain a first energy graph;

[0054] Furthermore, the first electrical signal is an electrical signal of plants of the same species or type.

[0055] Specifically, a first energy graph is obtained by frequency domain analysis of the first electrical signal. The energy graph can reflect the energy distribution of the signal at different frequencies.

[0056] Step 102: establishing a data set according to the first energy map and the corresponding preset labels, and training a preset classification model according to the data set to obtain a growth state prediction model.

[0057] Specifically, the data set includes multiple sub-data, each sub-data includes a first energy map and a corresponding preset label, and the corresponding preset label is a growth state such as healthy, water shortage, high CO2, hypothermia, and insect pests.

[0058] As an optional embodiment, the data set is divided into a training set and a test set according to a preset ratio. The preset classification model is trained using the training set to obtain a candidate growth state prediction model. The candidate growth state prediction model is evaluated for model capability based on the test set. If the evaluation result meets the preset indicator, the candidate growth state prediction model is used as the growth state prediction model to execute step 103. This ensures that the growth state prediction model has good generalization ability and avoids overfitting.

[0059] Furthermore, if the evaluation result does not meet the preset indicator, the data set is redivided into a training set and a test set according to a preset ratio, and the training is completed according to the above method until the evaluation result does not meet the preset indicator to obtain a growth status prediction model.

[0060] Furthermore, the indicators used to evaluate model capabilities include accuracy, precision, recall, F1 score, etc.

[0061] Step 103: Acquire a second electrical signal of the plant to be tested, and perform frequency domain analysis on the second electrical signal to acquire a second energy graph.

[0062] Furthermore, the second electrical signal is an electrical signal of a plant of the same species or type as the first electrical signal.

[0063] Specifically, the method of performing frequency domain analysis on the second electrical signal to obtain the second energy map is similar to the method of performing frequency domain analysis on the first electrical signal to obtain the first energy map, and will not be described in detail here.

[0064] As another optional embodiment, in step 103, the dataset is divided into a training set, a test set, and a validation set according to a preset ratio. A preset classification model is trained based on the training set and the test set to obtain a growth state prediction model; the first energy map in the validation set is used as the second energy map in this step, and step 104 is executed.

[0065] Step 104: Input the second energy map into a growth state prediction model to obtain a prediction result.

[0066] Specifically, the prediction results obtained include growth status such as health / water shortage / high CO2 / hypothermia.

[0067] This patent implements steps 101 to 104, using frequency domain analysis to process the first electrical signal and accurately obtain a first energy map that intuitively reflects the distribution characteristics of the signal energy in different frequency intervals. This energy map breaks through the limitations of traditional analysis methods and accurately presents subtle changes in plant growth characteristics in a quantitative form, providing key data support for subsequent analysis.

[0068] On this basis, a data set constructed based on the first energy map and its corresponding preset labels covers the multivariate feature information of the plant growth process. The preset classification model is deeply trained using this data set, and the growth status prediction model finally obtained significantly improves the accuracy and reliability of plant growth status prediction through learning and optimization of a large amount of data. Compared with the existing technology, the technical solution of this patent can effectively avoid prediction errors caused by incomplete data feature extraction and insufficient model training, and provides a more practical technical solution for the scientific prediction of plant growth status. It can be widely used in agricultural production, plant research and other fields, and has outstanding technical advantages and application prospects.

[0069] In another exemplary embodiment of the present application, in order to better perform frequency domain processing on the first electrical signal, better eliminate noise, and accurately obtain the first energy map with enhanced features, such as Figure 2 As shown, based on the above steps 101 to 104, step 102 is replaced by the following steps 301 to 304; Figure 3 , steps 301 to 304 specifically include:

[0070] Step 301: Acquire a first electrical signal of a plant, and preprocess the first electrical signal to acquire a first preprocessed signal.

[0071] As an implementation manner, the first electrical signal is processed by Butterworth low-pass filtering to obtain the first preprocessed signal.

[0072] Step 302: Obtain a first reconstructed electrical signal by performing wavelet decomposition and noise reduction reconstruction on the first preprocessed signal.

[0073] Specifically, the first preprocessed signal is subjected to wavelet decomposition to obtain first decomposition coefficients, and the first decomposition coefficients are subjected to threshold denoising and inverse wavelet transform to perform denoising and reconstruction to obtain a first reconstructed electrical signal.

[0074] Step 303: Perform short-time Fourier transform on the first reconstructed electrical signal to obtain a first energy map.

[0075] Specifically, the short-time Fourier transform divides the first reconstructed electrical signal into several short time periods, applies the Fourier transform in each short time period, and outputs a complex matrix in which each column corresponds to the spectrum of the time period. The first energy map can be generated by calculating the energy of each frequency band.

[0076] like Figure 4 This is a schematic diagram of a first energy diagram obtained according to the method of the present application.

[0077] In another exemplary embodiment of the present application, in order to achieve reconstruction of the first decomposition coefficient, the above step 302 is replaced by the following steps 501 to 505, such as Figure 5 As shown, where:

[0078] Step 501 , performing three-level DWT wavelet decomposition on a first preprocessed signal to obtain first decomposition coefficients; the first decomposition coefficients include a first detail coefficient D1 , a first detail coefficient D2 , a first detail coefficient D3 and a first approximation coefficient A3 .

[0079] Specifically, the data obtained after three-level discrete wavelet transform (DWT) decomposition is a set of coefficients with a hierarchical structure. The first level of decomposition yields high-frequency coefficients (D1) and low-frequency coefficients (A1). The second level of decomposition decomposes A1 further to obtain D2 and A2. The third level of decomposition decomposes A2 further to obtain D3 and A3. The final first decomposition coefficients are a set of coefficients: {D1, D2, D3, A3}, which includes the first detail coefficient D1, the first detail coefficient D2, the first detail coefficient D3, and the first approximation coefficient A3.

[0080] Step 502: Obtain a high-frequency component estimated noise standard deviation based on a first detail coefficient D1 of the first decomposition coefficient, and perform threshold processing on the first decomposition coefficient based on the high-frequency component estimated noise standard deviation to obtain a first denoising decomposition coefficient;

[0081] Specifically, the first detail coefficient D1 contains the highest frequency component of the signal, and the standard deviation of the high-frequency component estimation noise is obtained based on the first detail coefficient D1, and the acquisition formula is:

[0082]

[0083] Where σ is the estimated noise standard deviation, MAD is the median absolute deviation, and 0.6745 is the conversion factor between the standard deviation of the Gaussian distribution and the MAD.

[0084] Step 503: The optimal standard deviation is obtained by using a general threshold formula based on the estimated noise standard deviation of the high-frequency component. The formula is:

[0085]

[0086] Where T is the optimal threshold and N is the signal length.

[0087] Step 504 : Perform threshold denoising on the first detail coefficient D1 , the first detail coefficient D2 , and the first detail coefficient D3 using the optimal threshold to obtain first denoised detail coefficients B1 , B2 , and B3 .

[0088] Specifically, the coefficients of the first detail coefficient D1, the first detail coefficient D2 and the first detail coefficient D3 that are not less than the optimal threshold are retained, and the coefficients less than the optimal threshold are set to 0, thereby updating the detail coefficients to obtain the first noise reduction detail coefficient B1, the first noise reduction detail coefficient B2 and the first noise reduction detail coefficient B3.

[0089] Step 505 : Perform wavelet inverse transform processing on the first denoising detail coefficient B1 , the first denoising detail coefficient B2 , the first denoising detail coefficient B3 and the first approximation coefficient A3 to obtain a first reconstructed electrical signal.

[0090] Specifically, after wavelet transform, signal energy is usually concentrated in a few large coefficients, while noise energy is dispersed among many small coefficients. Filtering out small coefficients by thresholding is equivalent to removing the noise-dominated part.

[0091] Specifically, the method of the present application can effectively remove noise without significantly affecting the important features of the original signal, effectively deal with noise problems in different frequency ranges, retain key information related to plant electrical signals, and achieve preliminary separation of signals and noise.

[0092] Furthermore, as an optional embodiment, a CNN-LSTM-Attention hybrid neural network model is used to preset the classification model.

[0093] The convolutional neural network (CNN) is responsible for extracting features from the input energy map. It uses a one-dimensional convolution operation to reduce the complexity of the features. Then, a pooling layer (MaxPooling) is used to reduce the dimensionality of the data, further filtering out important features. Finally, the Flatten() function is used to convert the data dimension for processing in subsequent network layers.

[0094] The long short-term memory (LSTM) network introduces a gated recurrence mechanism, which effectively improves the defects of the traditional RNN model, fully utilizes the relationship between sequences, and captures long-term dependency features.

[0095] In addition, an attention mechanism is added to the model. The attention mechanism can assign different weights to each feature based on its importance to the diagnosis of plant growth status. Adding the attention mechanism after the LSTM network can be used to capture the information features output by the LSTM layer in the prediction model, divide the weight ratio, and extract key information. This allows the model to pay more attention to features that are important for diagnosing plant growth status and ignore unimportant information, thereby improving the diagnostic accuracy of the model.

[0096] This application also provides an application scenario, such as application in modern smart greenhouses, large-scale planting bases, etc. In a smart greenhouse or planting base, the system automatically adjusts the environmental parameters in the greenhouse through real-time prediction of the growth status of plants. For example, when it is predicted that the growth of plants is hindered due to excessively high temperatures, the cooling equipment is automatically turned on; when it is detected that the plants are short of water, the irrigation system is started in time to achieve precise irrigation. In large-scale planting bases, this prediction method can be used to detect plant growth abnormalities in advance, such as early symptoms of diseases and pests, so that planting personnel can take timely prevention and control measures to reduce crop losses. This application scenario applies the above-mentioned plant growth status prediction method. Specifically, it can be applied to scenarios such as agricultural research and plant growth condition management.

[0097] Furthermore, agricultural research: Predicting the growth status of plants through the above-mentioned plant growth status prediction method helps to deeply understand the impact mechanism of different factors on plant growth.

[0098] Furthermore, plant growth condition management: in a greenhouse or controlled environment, environmental conditions (such as light, temperature, humidity, etc.) are adjusted according to the predicted results of plant growth status to promote the best growth environment, predict possible diseases or pests in advance, take preventive measures, and reduce economic losses.

[0099] Based on the same inventive concept, embodiments of the present application also provide a plant growth state prediction device for implementing the aforementioned plant growth state prediction method. The solution provided by this device is similar to the solution described in the aforementioned method. Therefore, the specific limitations of one or more plant growth state prediction device embodiments provided below can be found in the above-described limitations on plant growth state prediction and will not be further elaborated here.

[0100] In an exemplary embodiment, Figure 6 As shown, a plant growth state prediction device is provided, the plant growth state prediction device comprising:

[0101] A first feature module is used to obtain a first electrical signal of the plant, and perform frequency domain analysis on the first electrical signal to obtain a first energy map;

[0102] A training module, configured to establish a data set based on the first energy map and corresponding preset labels, and train a preset classification model based on the data set to obtain a growth state prediction model;

[0103] A second feature module is used to obtain a second electrical signal of the plant to be tested, and perform frequency domain analysis on the second electrical signal to obtain a second energy map;

[0104] The prediction module is used to input the second energy map into the growth state prediction model to obtain a prediction result.

[0105] As an optional embodiment, the first feature module is specifically configured to:

[0106] Preprocessing the first electrical signal to obtain a first preprocessed signal;

[0107] Obtaining a first reconstructed electrical signal by performing wavelet decomposition and noise elimination reconstruction on the first preprocessed signal;

[0108] Performing short-time Fourier transform on the first reconstructed electrical signal to obtain a first energy map.

[0109] As an optional embodiment, the first feature module is further configured to:

[0110] Obtaining a first reconstructed electrical signal by performing wavelet decomposition and noise elimination reconstruction on the first preprocessed signal includes:

[0111] Performing three-level DWT wavelet decomposition on the first preprocessed signal to obtain first decomposition coefficients;

[0112] A first reconstructed electrical signal is obtained by performing noise elimination and reconstruction on the first decomposition coefficient.

[0113] As an optional embodiment, the first decomposition coefficient includes a first detail coefficient D1, a first detail coefficient D2, a first detail coefficient D3, and a first approximation coefficient A3. The first decomposition coefficient is reconstructed by noise removal to obtain a first reconstructed electrical signal. The first feature module is further configured to:

[0114] Obtaining a high-frequency component estimation noise standard deviation according to the first detail coefficient D1;

[0115] The optimal threshold is obtained by estimating the noise standard deviation based on the high-frequency component and using the universal threshold formula;

[0116] Performing threshold denoising on the first detail coefficient D1, the first detail coefficient D2 and the first detail coefficient D3 using an optimal threshold to obtain a first denoised detail coefficient B1, a first denoised detail coefficient B2 and a first denoised detail coefficient B3;

[0117] The first denoising detail coefficient B1, the first denoising detail coefficient B2, the first denoising detail coefficient B3 and the first approximation coefficient A3 are processed by inverse wavelet transform to obtain a first reconstructed electrical signal.

[0118] As an optional embodiment, in the training module, the preset classification model is a CNN-LSTM-Attention hybrid neural network model, and the CNN-LSTM-Attention includes a CNN feature extraction module, a Flatten function, an LSTM network, a fully connected layer, and an Attention layer connected in sequence.

[0119] Based on the same inventive concept, the embodiment of the present application also provides a plant growth state prediction device for implementing the plant growth state prediction method involved above. On the basis of the above-mentioned plant growth state prediction method, the function of analyzing the plant growth state through plant electrical signals can be realized semi-detached from the server, and the hardware equipment can be miniaturized to solve the problems in the prior art of performing operations such as data enhancement, feature engineering, and data format conversion on a large amount of experimental data, such as high requirements for equipment performance, large equipment size, high equipment cost, and difficulty in portability. This device is similar to the implementation scheme described in the above-mentioned method, so the specific limitations in the one or more plant growth state prediction device embodiments provided below can refer to the limitations on plant growth state prediction above, and will not be repeated here.

[0120] In an exemplary embodiment, Figure 7 As shown, a plant growth status prediction device is provided, which includes an environment perception module, a processor module and a display module, wherein:

[0121] An environmental sensing module is used to collect environmental parameters of plant growth and the first electrical signal of the plant; wherein the environmental parameters include soil moisture, CO2 concentration in the air, temperature and light intensity;

[0122] The processor module, the local processor includes a plant growth state prediction device, which is used to process the second electrical signal using the above-mentioned plant growth state prediction method to obtain a prediction result;

[0123] Display module, used to display environmental parameters and prediction results in real time.

[0124] The data transmission module is used to realize data transmission between the remote processor module and the local processor module.

[0125] The data monitoring module analyzes and monitors the status of the data transmitted by the data transmission module, determines whether the data is correct, and transmits it to the local processor module.

[0126] As an optional embodiment, Figure 8 The processor module is divided into plant growth status prediction device, operating system driver unit, data reading and fusion unit, input and output pin configuration unit, interrupt program unit and environmental parameter processing unit according to software functions. Among them:

[0127] The operating system driver unit writes corresponding drivers according to different sensors to implement sensor initialization, data reading and other operations.

[0128] The data reading and fusion unit is used to periodically read sensor data and analyze and fuse different sensor data to obtain more accurate and comprehensive environmental information.

[0129] Input and output pin configuration is used to allocate the data transmission channel of the sensor and control the working mode of the data transmission module.

[0130] The interrupt program unit is used to control the timely response of the data transmission module when uploading and receiving data, and reasonably allocate priorities to ensure that the data is displayed on the display module in a timely manner.

[0131] The environmental parameter processing unit is used to calibrate the sensor data using linear fitting or polynomial fitting methods, balance and optimize the measurement accuracy and response time, and then transmit it to the display module for waveform display.

[0132] As an optional embodiment, Figure 9 and Figure 10 The processor module includes a remote processor module and a local processor module according to the hardware function. Further, the plant growth state prediction device also includes a data transmission module and a data monitoring module, wherein:

[0133] The data transmission module uses spread spectrum modulation technology, matched with finished modules, antennas and interface circuits, such as LoRa modules, to achieve long-distance data transmission between the local processor module and the remote processor module.

[0134] The data monitoring module uses the built-in monitoring function of the MCU to parse and monitor the status of communication data packets to determine whether the data is sent and received correctly.

[0135] The power supply module is the power supply module of the data transmission module, and is equipped with an adaptive linear voltage regulated output circuit according to the specific requirements of the data transmission module.

[0136] As an optional embodiment, the plant growth status prediction device, operating system driver unit, data reading and fusion unit, input and output pin configuration unit and interrupt program unit can be configured in the remote processor module and the local processor module according to needs to achieve the miniaturization of the plant growth status prediction device.

[0137] For example: it can be divided into remote processor module and local processor module according to function, such as configuring the operating system driver unit, data reading and fusion unit, input and output pin configuration unit and interrupt program unit in the local processor module, and configuring the plant growth status prediction device in the remote processor module.

[0138] For another example: the first feature module, training module and second feature module can be configured in the local processor module, and the prediction module can be placed in the remote processor module. The remote processor module can be used to complete the process of processing a large number of data sets, thereby realizing the miniaturization of the portable device.

[0139] Furthermore, if Figure 11 As shown, the environment perception module includes a sensor unit and a data format conversion unit; wherein:

[0140] The sensor unit includes a temperature sensor, a humidity sensor, a CO2 concentration sensor, a light sensor and an electrical signal sensor, which are connected to the corresponding serial ports of the mainboard of the local processor module.

[0141] The data format conversion unit includes a signal conditioning circuit and an A / D converter. Since the sensor output signal is weak and contains noise, it needs to be amplified and filtered by the signal conditioning circuit and converted from analog to digital by the A / D converter.

[0142] As an exemplary embodiment, the process of the device is as follows:

[0143] After power-on, the local processor module configures the environment perception module and data transmission module, initializes the clock signal, hardware resource configuration, and the initial values of the environmental parameters in the system; and uses the communication and serial port debugging program to test the serial port and verify whether the communication between each module is normal.

[0144] The local processor module controls the environmental sensing module to collect environmental parameters for plant growth and the first electrical signal of the plant; wherein the environmental parameters include soil moisture, CO2 concentration in the air, temperature and light intensity;

[0145] The local processor module obtains a prediction result by using the above-mentioned plant growth state prediction method for the second electrical signal, performs operations such as curve fitting and debugging on the environmental parameters, and sends the prediction result and the processed environmental parameters to the display module;

[0146] The display module displays the prediction results and processed environmental parameters in real time, and intuitively displays the plant electrical signal waveform and real-time changes of environmental parameters through the human-computer interaction interface.

[0147] Furthermore, to enable long-distance data transmission between the local and remote processor modules, the selected data transmission module utilizes spread spectrum modulation technology, along with off-the-shelf modules, antennas, and interface circuitry. During data transmission between the local and remote processor modules, the data monitoring module utilizes the MCU's built-in monitoring capabilities to parse and monitor the communication data packets, determining whether the data is being sent and received correctly.

[0148] As an optional embodiment, the remote processor module is a remote computer host, and its hardware environment is: i7-4060 CPU, 8GB memory, and the software environment is: Qt Creator, Qt Quick, VMware Workstation Pro, Ubuntu.

[0149] The remote processor module is the RK3588 development board, and its software environment is the Linux operating system.

[0150] As an optional embodiment, the display module is a MIPI capacitive touch screen module 8001280.

[0151] In an exemplary embodiment, a computer device is provided. The computer device may be a server or a terminal. The internal structure diagram thereof may be as follows: Figure 12As shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, abbreviated as I / O) and a communication interface. The processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store data. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a method for predicting plant growth status is implemented.

[0152] Those skilled in the art will understand that Figure 12 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. Specifically, the computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement.

[0153] In an exemplary embodiment, a computer device is further provided, including a memory and a processor. The memory stores a computer program, and the processor implements the steps in the above method embodiments when executing the computer program.

[0154] In an exemplary embodiment, a computer-readable storage medium is provided, storing a computer program. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented.

[0155] In an exemplary embodiment, a computer program product is provided, including a computer program. When the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0156] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0157] Those skilled in the art will appreciate that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, database or other media used in the embodiments provided in this application may include at least one of non-volatile and volatile memory. Non-volatile memory may include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory may include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM may be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0158] The databases involved in the various embodiments provided herein may include at least one of a relational database and a non-relational database. Non-relational databases may include, but are not limited to, distributed databases based on blockchains. The processors involved in the various embodiments provided herein may include, but are not limited to, general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic units, data processing logic units based on quantum computing, and the like.

[0159] The technical features of the above embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0160] This document uses specific examples to illustrate the principles and implementation methods of this application. The description of the above examples is only intended to help understand the method and core concept of this application. At the same time, those skilled in the art will find that the specific implementation methods and application scopes may vary based on the concept of this application. In summary, the content of this specification should not be construed as limiting this application.

Claims

1. A method for predicting plant growth status, characterized in that: The plant growth state prediction method comprises: Acquire a first electrical signal of the plant, and perform frequency domain analysis on the first electrical signal to acquire a first energy map; Establishing a data set according to the first energy map and the corresponding preset labels, and training a preset classification model according to the data set to obtain a growth status prediction model; Acquire a second electrical signal of the plant to be tested, and perform the frequency domain analysis on the second electrical signal to acquire a second energy map; The second energy map is input into the growth state prediction model to obtain a prediction result.

2. The plant growth state prediction method according to claim 1, characterized in that: The step of analyzing and processing the first electrical signal in the frequency domain to obtain a first energy graph includes: Preprocessing the first electrical signal to obtain a first preprocessed signal; Obtaining a first reconstructed electrical signal by performing wavelet decomposition and noise elimination reconstruction on the first preprocessed signal; Performing short-time Fourier transform on the first reconstructed electrical signal to obtain a first energy map.

3. The plant growth state prediction method according to claim 2, characterized in that: The step of obtaining a first reconstructed electrical signal by performing wavelet decomposition and noise reduction reconstruction on the first preprocessed signal includes: Performing three-level DWT wavelet decomposition on the first preprocessed signal to obtain first decomposition coefficients; The first decomposition coefficients are reconstructed by noise elimination to obtain a first reconstructed electrical signal.

4. The plant growth state prediction method according to claim 3, characterized in that: The first decomposition coefficients include a first detail coefficient D1, a first detail coefficient D2, a first detail coefficient D3, and a first approximation coefficient A3. The first decomposition coefficients are reconstructed by noise elimination to obtain a first reconstructed electrical signal, including: Obtaining a high-frequency component estimation noise standard deviation according to the first detail coefficient D1; Estimating the noise standard deviation of the high-frequency component and obtaining the optimal threshold value through a universal threshold formula; Performing threshold denoising on the first detail coefficient D1, the first detail coefficient D2, and the first detail coefficient D3 using the optimal threshold to obtain a first denoised detail coefficient B1, the first denoised detail coefficient B2, and the first denoised detail coefficient B3; The first noise removal detail coefficient B1, the first noise removal detail coefficient B2, the first noise removal detail coefficient B3 and the first approximation coefficient A3 are processed by wavelet inverse transformation to obtain the first reconstructed electrical signal.

5. The plant growth state prediction method according to claim 1, characterized in that: The preset classification model is a CNN-LSTM-Attention hybrid neural network model, and the CNN-LSTM-Attention includes a CNN feature extraction module, a Flatten function, an LSTM network, a fully connected layer and an Attention layer connected in sequence.

6. A plant growth status prediction device, characterized in that: The plant growth state prediction device includes an environment perception module, a processor module and a display module; wherein: The environmental sensing module is used to collect environmental parameters of plant growth and the first electrical signal of the plant; wherein the environmental parameters include soil moisture, CO2 concentration in the air, temperature and light intensity; The processor module includes a plant growth state prediction device, and the plant growth state prediction device is used to process the second electrical signal according to the plant growth state prediction method according to any one of claims 1 to 5 to obtain a prediction result; The display module is used to display the environmental parameters and the prediction results in real time.

7. The plant growth state prediction device according to claim 6, characterized in that: The processor module includes a remote processor module and a local processor module, and the plant growth state prediction device also includes a data transmission module and a data monitoring module, wherein: The data transmission module is used to realize data transmission between the remote processor and the local processor. The data monitoring module is used to analyze and monitor the status of the data transmitted by the data transmission module, determine whether the data is correct, and transmit the data to the local processor module.

8. A computer device comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the plant growth status prediction method according to any one of claims 1 to 5.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the plant growth state prediction method according to any one of claims 1 to 5 are implemented.

10. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the plant growth state prediction method according to any one of claims 1 to 5 are implemented.

Images