Method, system, device, medium and product for predicting black rot of cauliflower leaves
Patent Information
- Application Number
- CN202510349839.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-03-24
- Publication Date
- 2026-09-25
AI Technical Summary
然而，以上技术均需农业专业人员在实验室中进行样本采集和操作，过程耗时耗力，难以实现农田病害的实时快速检测，且通常需要结合光谱学以外的知识进行辅助分析，增加了应用的复杂性
[0040]本申请公开了一种花椰菜叶片的黑腐病预测方法、系统、设备、介质及产品，首先，通过采集待测花椰菜叶片的近红外光谱数据，并结合深度神经网络模型构建花椰菜叶片的黑腐病预测模型，然后，利用花椰菜叶片的黑腐病预测模型对花椰菜叶片的黑腐病进行动态监测与预测。实现了对黑腐病发病过程的早期检测，提升了花椰菜叶片的黑腐病监测的效率和准确性，具有显著的实际应用价值。
Smart Images
Figure CN122814522A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of pest and disease monitoring technology, and in particular to a method, system, equipment, medium and product for predicting black rot disease in cauliflower leaves. Background Technology
[0002] Cauliflower is an important vegetable crop. Black rot is a destructive disease caused by Xanthomonas campestris pv. campestris (Xcc), a species of Xanthomonas, which seriously affects the yield, quality, and overall performance of cauliflower crops. This pathogen mainly invades the host through stomata or plant wounds, rapidly multiplies, and secretes large amounts of extracellular polysaccharides and xanthan gum, which clog the plant's vascular system.
[0003] As the disease progresses, typical "V"-shaped yellowing lesions appear on the leaf edges. As these lesions expand, the veins also turn black, eventually leading to the withering and death of the entire plant. Xcc has an extremely strong ability to spread, rapidly disseminateing through air, water droplets, and agricultural implements. Therefore, early detection of black rot in cauliflower is crucial for effective disease control, ensuring healthy crop growth, and improving agricultural productivity.
[0004] For a long time, the monitoring of black rot in cauliflower leaves has relied heavily on traditional manual surveys, and most of these methods cannot predict the occurrence of black rot in real time. In recent years, with the development of new agricultural technologies, the comprehensive utilization of new technologies such as computer technology, satellite remote sensing monitoring technology, pathogen spore capture technology, and digital image processing technology has enabled efficient and convenient monitoring of the actual incidence of black rot in related fruits and vegetables, such as wheat and soybeans.
[0005] However, current monitoring methods for black rot in fruits and vegetables typically focus on statistical analysis of specific populations and disease incidence, often neglecting dynamic monitoring of the disease's progression. This makes it difficult to detect black rot in its early stages in a timely and accurate manner, resulting in a lack of mature and effective monitoring methods.
[0006] Several methods have emerged for monitoring the incidence and impact of wheat black rot. Some researchers have demonstrated that the hyperspectral reflectance of the wheat canopy can be used to detect wheat black rot and estimate grain yield under different nitrogen input levels by analyzing the effect of wheat black rot severity and nitrogen input levels on spectral reflectance. Furthermore, other researchers have combined canopy reflectance spectroscopy with genome-wide prediction to enhance the selective response of cultivated strawberries to white rot resistance, helping breeders quickly measure relevant traits, improve selection accuracy, and thus increase breeding efficiency. However, all of these techniques require agricultural professionals to collect and operate samples in laboratories, a time-consuming and labor-intensive process that makes real-time rapid detection of farmland diseases difficult. Moreover, they often require auxiliary analysis using knowledge beyond spectroscopy, increasing the complexity of their application.
[0007] Therefore, it is necessary to provide a method for predicting black rot in cauliflower leaves to solve the above problems. Summary of the Invention
[0008] The purpose of this application is to provide a method, system, equipment, medium, and product for predicting black rot in cauliflower leaves, which enables early detection of the disease process of black rot in cauliflower leaves and improves the efficiency and accuracy of black rot monitoring.
[0009] To achieve the above objectives, this application provides the following solution:
[0010] In a first aspect, this application provides a method for predicting black rot disease in cauliflower leaves, the method comprising:
[0011] Near-infrared spectral scanning was performed on the cauliflower leaves to be tested, and the near-infrared spectral data of the cauliflower leaves to be tested were obtained.
[0012] The near-infrared spectral data of the cauliflower leaves to be tested were preprocessed to obtain the preprocessed near-infrared spectral data of the cauliflower leaves to be tested.
[0013] The near-infrared spectral data of the pre-processed cauliflower leaves to be tested were filtered by band to obtain near-infrared spectral data of the cauliflower leaves to be tested that meet the preset band range.
[0014] Near-infrared spectral data of cauliflower leaves within a preset wavelength range are input into a black rot prediction model for cauliflower leaves, and the prediction results for black rot are output. The black rot prediction results for cauliflower leaves include: the predicted time of onset of black rot and the predicted severity of black rot. The black rot prediction model for cauliflower leaves is obtained by training a deep neural network model using a training set. The onset time of black rot is the difference between the time when black rot symptoms become obvious and the time when the black rot fungus is inoculated.
[0015] Optionally, the near-infrared spectral data of the cauliflower leaves to be tested are preprocessed to obtain preprocessed near-infrared spectral data of the cauliflower leaves to be tested, specifically including:
[0016] The near-infrared spectral data of the cauliflower leaves to be tested were processed using the multivariate scattering correction method to obtain the corrected near-infrared spectral data of the cauliflower leaves to be tested.
[0017] The near-infrared spectral data of the cauliflower leaves under test were processed using the standard normal variable method to obtain preprocessed near-infrared spectral data.
[0018] Optionally, the near-infrared spectral data of the preprocessed cauliflower leaves to be tested can be filtered by using a competitive adaptive reweighted sampling method.
[0019] Optionally, the training process for the cauliflower leaf black rot prediction model specifically includes:
[0020] A training set is constructed, comprising: near-infrared spectral data of cauliflower leaves satisfying multiple preset sample bands, the true values of the onset time of black rot in cauliflower leaves, and the true values of the severity of black rot in cauliflower leaves; wherein, the preset sample bands are obtained by merging the characteristic bands corresponding to the inoculation time of black rot fungi and the characteristic bands corresponding to the severity of black rot, and the onset time of black rot in cauliflower leaves is the difference between the time when black rot symptoms are obvious and the time of inoculation with black rot fungi.
[0021] Construct a deep neural network model;
[0022] The deep neural network model was trained using the training set to obtain a prediction model for black rot disease in cauliflower leaves.
[0023] Optionally, a training set is constructed, specifically including:
[0024] The sample cauliflower leaves were inoculated with black rot pathogens.
[0025] After the inoculation period is set, the near-infrared spectral scanning of the inoculated cauliflower leaves is performed daily to obtain the near-infrared spectral data of the inoculated cauliflower leaves. The true values of the onset time of black rot and the true values of the severity of black rot in the cauliflower leaves are also recorded.
[0026] The near-infrared spectral data of the inoculated cauliflower leaves were preprocessed to obtain the preprocessed near-infrared spectral data of the cauliflower leaves.
[0027] The near-infrared spectral data of the preprocessed cauliflower leaf samples were filtered by band to obtain near-infrared spectral data of cauliflower leaf samples that met the preset bands.
[0028] The training set is constructed based on near-infrared spectral data of cauliflower leaves that meet multiple preset wavelength bands, the true values of the onset time of black rot in cauliflower leaves, and the true values of the severity of black rot in cauliflower leaves.
[0029] Optionally, the deep neural network model is trained using the training set to obtain a prediction model for black rot in cauliflower leaves, specifically including:
[0030] Using the near-infrared spectral data of cauliflower leaves that meet the preset wavelength bands as input, and the true values of the onset time and severity of black rot in the corresponding cauliflower leaves as output, the deep neural network model is trained until the training count reaches the stopping condition, thus obtaining the black rot prediction model for cauliflower leaves.
[0031] Secondly, this application provides a black rot prediction system for cauliflower leaves. The cauliflower leaf black rot prediction system is used to implement the aforementioned cauliflower leaf black rot prediction method. The cauliflower leaf black rot prediction system includes:
[0032] The data acquisition unit is used to acquire near-infrared spectral data of the cauliflower leaves to be tested;
[0033] The preprocessing unit is used to preprocess the near-infrared spectral data of the cauliflower leaves to be tested, and obtain the preprocessed near-infrared spectral data of the cauliflower leaves to be tested.
[0034] The screening unit is used to perform band screening on the near-infrared spectral data of the pre-processed cauliflower leaves to be tested, so as to obtain the near-infrared spectral data of the cauliflower leaves to be tested that meet the preset band range.
[0035] The prediction unit is used to input near-infrared spectral data of cauliflower leaves that meet the preset band range into the black rot prediction model of cauliflower leaves, and output the black rot prediction results of cauliflower leaves; the black rot prediction results of cauliflower leaves include: the predicted value of the onset time of black rot of cauliflower leaves and the predicted value of the severity of black rot; the black rot prediction model of cauliflower leaves is obtained by training a deep neural network model using a training set; the onset time of black rot of cauliflower leaves is the difference between the time when the black rot symptoms are obvious and the time when the black rot fungus is inoculated.
[0036] Thirdly, this application provides a computer device, including: 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 method for predicting black rot of cauliflower leaves as described in any of the above claims.
[0037] Fourthly, this application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method for predicting black rot in cauliflower leaves as described in any of the preceding claims.
[0038] Fifthly, this application provides a computer program product, including a computer program that, when executed by a processor, implements the method for predicting black rot in cauliflower leaves as described in any of the preceding claims.
[0039] According to the specific embodiments provided in this application, this application has the following technical effects:
[0040] This application discloses a method, system, equipment, medium, and product for predicting black rot in cauliflower leaves. First, near-infrared spectral data of the cauliflower leaves to be tested are collected, and a black rot prediction model is constructed using a deep neural network model. Then, the black rot prediction model is used to dynamically monitor and predict black rot in cauliflower leaves. This enables early detection of the black rot disease process, improves the efficiency and accuracy of black rot monitoring in cauliflower leaves, and has significant practical application value. Attached Figure Description
[0041] To more clearly illustrate the technical solutions in the embodiments of this application 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 this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0042] Figure 1 A schematic flowchart of a method for predicting black rot in cauliflower leaves provided in an embodiment of this application;
[0043] Figure 2 This is a schematic diagram of near-infrared spectral data image provided in an embodiment of this application;
[0044] Figure 3 This is a schematic diagram of a cauliflower leaf with a relatively obvious degree of black rot disease, provided in one embodiment of this application;
[0045] Figure 4 This is a schematic diagram of a deep neural network model structure provided in an embodiment of this application;
[0046] Figure 5This is a schematic diagram showing the current stage of prediction results for cauliflower leaves.
[0047] Figure 6 A schematic diagram showing the predicted severity of black rot in cauliflower leaves;
[0048] Figure 7 This is a schematic diagram of the structure of a computer device provided in an embodiment of this application. Detailed Implementation
[0049] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.
[0050] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments.
[0051] To address the limitations of existing methods, which require high levels of specialized knowledge and cannot be widely applied, this application provides a method for predicting black rot in cauliflower leaves based on machine learning and near-infrared spectroscopy. This method collects near-infrared spectral data of cauliflower leaves and combines it with a neural network algorithm to dynamically monitor and predict black rot. This enables early detection of the disease process, improves monitoring efficiency and accuracy, and has significant practical application value.
[0052] In one exemplary embodiment, such as Figure 1 As shown, a method for predicting black rot in cauliflower leaves is provided, including the following steps: (The method includes...)
[0053] Step S1: Near-infrared spectral scanning is performed on the cauliflower leaves to be tested to obtain the near-infrared spectral data of the cauliflower leaves to be tested.
[0054] Step S2: Preprocess the near-infrared spectral data of the cauliflower leaves to be tested to obtain the preprocessed near-infrared spectral data of the cauliflower leaves to be tested.
[0055] To eliminate background noise and interference caused by instrument and position changes during near-infrared spectral data acquisition, it is necessary to preprocess the acquired near-infrared spectral data of the cauliflower leaves. Preprocessing the raw near-infrared spectral data of the cauliflower leaves using Multiplicative Scatter Correction (MSC) and Standard Normal Variation (SNV) methods can eliminate noise and baseline drift interference.
[0056] Step S3: Perform band screening on the preprocessed near-infrared spectral data of the cauliflower leaves to be tested to obtain near-infrared spectral data of the cauliflower leaves to be tested that meet the preset band range.
[0057] Step S4: Input the near-infrared spectral data of the cauliflower leaves to be tested, which meet the preset band range, into the black rot prediction model of cauliflower leaves, and output the black rot prediction results of cauliflower leaves; the black rot prediction results of cauliflower leaves include: the predicted value of the onset time of black rot of cauliflower leaves and the predicted value of the severity of black rot; the black rot prediction model of cauliflower leaves is obtained by training a deep neural network model using a training set.
[0058] As an optional implementation method, step S2 specifically includes:
[0059] Step S21: The near-infrared spectral data of the cauliflower leaves to be tested are processed using the multivariate scattering correction method to obtain the corrected near-infrared spectral data of the cauliflower leaves to be tested.
[0060] Specifically, the multiple scattering correction method corrects for the effects of multiple scattering by constructing a ratio between the sample near-infrared spectral data (i.e., the spectral numerical matrix obtained from actual measurements of cauliflower leaves under a near-infrared spectrometer) and the reference near-infrared spectral data (i.e., standard near-infrared spectral data without any baseline drift or interference). Multiple scattering leads to an increase in optical path length and a decrease in signal intensity. The basic formula for MSC correction is as follows:
[0061] R MSC =(R sample -R min ) / (R ref -R min (1)
[0062] Among them, R MSC These are the corrected near-infrared spectral data of the cauliflower leaves to be tested, R sample These are the near-infrared spectral data of the cauliflower leaves to be tested, R ref It is based on near-infrared spectral data, R minThis is the minimum value of the near-infrared spectral data. The purpose of MSC correction is to remove residual scattering components in the spectrum and improve the accuracy and comparability of the near-infrared spectral data of the samples.
[0063] Step S22: The near-infrared spectral data of the cauliflower leaves to be tested are processed using the standard normal variable method to obtain preprocessed near-infrared spectral data.
[0064] Specifically, the standard normal distribution method is a normalization method used to eliminate differences caused by variations in light intensity and baseline drift. The basic formula for SNV processing is as follows:
[0065] R SNV =(R MSC -μ) / σ (2)
[0066] Among them, R SNV σ represents the preprocessed near-infrared spectral data; μ is the average value of the corrected near-infrared spectral data of the cauliflower leaves; and σ is the standard deviation of the corrected near-infrared spectral data of the cauliflower leaves. For each spectrum, the SNV method subtracts the average value from each data point in the near-infrared spectral data and then divides by the standard deviation. In this way, overall intensity variations and baseline drift in the near-infrared spectral data are eliminated, thus highlighting the spectral characteristics.
[0067] As an optional implementation, in step S3, the near-infrared spectral data of the preprocessed cauliflower leaves to be tested are filtered by band using a competitive adaptive reweighted sampling method.
[0068] As an optional implementation, step S4, the training process of the cauliflower leaf black rot prediction model, specifically includes:
[0069] Step S41, construct a training set; the training set includes: near-infrared spectral sample data of cauliflower leaves that meet the preset sample bands, the true value of the onset time of black rot in cauliflower leaves, and the true value of the severity of black rot in cauliflower leaves; wherein, the preset sample bands are obtained by merging the characteristic bands corresponding to the inoculation time of black rot fungi and the characteristic bands corresponding to the severity of black rot, and the onset time of black rot in cauliflower leaves is the difference between the time when black rot symptoms are obvious and the time when black rot fungi are inoculated.
[0070] Step S42: Construct a deep neural network model.
[0071] Step S43: Use the training set to train the deep neural network model to obtain a prediction model for black rot disease in cauliflower leaves.
[0072] As an optional implementation, step S41 specifically includes:
[0073] Step S411: Inoculate the sample cauliflower leaves with black rot fungi.
[0074] Specifically, 1) Normal cauliflower leaves provided by a certain research institute were used as inoculation samples for black rot. The leaves were taken from cauliflower seedlings in the seedling stage. True leaves of seedlings from the same batch of pre-germinated seedlings were placed in a plastic petri dish with a diameter of 12cm. Black rot spores were prepared into 1×10⁻⁶ spores using distilled water. 5 ml -1 The spore suspension was used to spread 1 ml of bacterial solution evenly on the surface of cauliflower leaves using an inoculation stick.
[0075] Step S412: After the preset inoculation period, perform near-infrared spectral scanning on the inoculated cauliflower leaves every day to obtain the near-infrared spectral data of the inoculated cauliflower leaves, and record the true value of the onset time of black rot and the true value of the severity of black rot on the cauliflower leaves.
[0076] Specifically, 1) Before conducting near-infrared spectral analysis of cauliflower leaves, place them in a dark environment to allow them to adapt to the lighting conditions in the measurement room. This step aims to eliminate the interference of ambient light on near-infrared spectral data acquisition and ensure the accuracy and consistency of the test results.
[0077] 2) To obtain near-infrared spectral data of cauliflower leaves, a Thermo Nicolet ANTARIS II Fourier transform near-infrared (FT-NIR) analyzer was used to scan the cauliflower leaves. Scanning was performed under constant temperature (24℃) and constant humidity (60%) conditions. The spectral range was 1000–2500 nm, with each sample scanned 64 times at a resolution of 8 cm⁻¹. -1 The analyzer was mounted 20 cm from the sample cauliflower leaves. Spectral images were acquired under artificial illumination, consisting of four 15-watt 12-volt bulbs, two located on either side of the lens. The main specifications of the hyperspectral camera included a Firewire (IEEE 1394b) interface, 14-bit digital output, and a 7° field of view. The objective lens, with a 17 mm focal length (maximum aperture F1.4), was hyperspectrally optimized to obtain reflectance data in 462 spectral bands from 386 to 1004 nm, achieving a spectral resolution of 1.3 nm. ENVI 5.3 software (ITT, Visual Information Solutions, Boulder, Colorado, USA) was used to extract reflectance values for each band from the region of interest, resulting in near-infrared spectral data images as shown below. Figure 2 As shown.
[0078] 3) Before fixing the cauliflower leaves onto the conveyor of the near-infrared spectroscopy detection equipment, the equipment needs to be preheated, zeroed, and calibrated. These procedures should be repeated before each day's experiments to ensure the equipment operates normally and the accuracy of measurements. After the equipment has completed preheating and calibration, remove the leaves from the petri dish and gently wipe off excess moisture from the leaf surface with a non-woven cloth to reduce the impact of moisture reflection on the spectral data. Then, place the treated leaves on the conveyor, ready for near-infrared spectroscopy detection.
[0079] Step S413: Preprocess the near-infrared spectral data of the inoculated cauliflower sample leaves to obtain the preprocessed near-infrared spectral data of the cauliflower sample leaves.
[0080] Step S414: Band selection is performed on the near-infrared spectral data of the preprocessed sample cauliflower leaves to obtain near-infrared spectral data of multiple sample cauliflower leaves that meet the preset bands.
[0081] Specifically, the near-infrared spectral data of the inoculated cauliflower leaves can be expressed as follows:
[0082]
[0083] Where, x i It represents the number of near-infrared bands in the i-th cauliflower leaf.
[0084] To obtain the optimal characteristic band, the target output Y needs to be defined. i,1 and Y i,2 The former is the difference between the set experimental period and the current number of experimental days, while the latter is the degree of black rot in cauliflower leaves (for ease of modeling and analysis, it is divided into three categories: 0 represents no black rot, 1 represents mild black rot, and 2 represents significant black rot). A schematic diagram of cauliflower leaves with a significant degree of black rot is shown below. Figure 3 As shown.
[0085] Based on the competitive adaptive reweighted sampling method, the near-infrared spectral data of the preprocessed cauliflower leaf samples were screened twice to obtain two sets of corresponding outputs of characteristic bands, namely the characteristic band number1 corresponding to the inoculation time and the characteristic band number2 corresponding to the degree of black rot. Then, the selected bands were merged to obtain the final characteristic band group number3 (i.e., the near-infrared spectral data of the cauliflower leaf samples that meet the preset bands of the samples).
[0086] Step S415: Based on the near-infrared spectral data of sample cauliflower leaves that meet multiple preset sample bands, the true values of the onset time of black rot in sample cauliflower leaves, and the true values of the severity of black rot in sample cauliflower leaves, the training set is constructed.
[0087] As an optional implementation, step S43 specifically includes:
[0088] Using the near-infrared spectral data of cauliflower leaves that meet the preset wavelength bands as input, and the true values of the onset time and severity of black rot in the corresponding cauliflower leaves as output, the deep neural network model is trained until the training count reaches the stopping condition, thus obtaining the black rot prediction model for cauliflower leaves.
[0089] As an optional implementation, step S42 specifically includes:
[0090] like Figure 4 As shown, the deep neural network model has 5 layers, including an input layer (which takes into account the length of the final feature band group numbers3), three intermediate layers (which use the ReLU activation function and the solver is "adam"), and an output layer (which outputs the difference Y between the set experimental period and the current number of experimental days). i,1 And the degree of black rot in leaves Y i,2 ).
[0091] Step S421, Construct the input layer:
[0092] The input layer takes in near-infrared spectral data in the wavelength range of 900nm to 1700nm collected by a near-infrared spectroscopy device (i.e., near-infrared spectral data of cauliflower leaves that meet the preset wavelength band of the sample), and takes each row of data (i.e. the sample of the label value) except for the first column (label column) as input to the network structure.
[0093] Step S422, construct the hidden layer:
[0094] The deep neural network model in this embodiment contains multiple hidden layers (i.e., intermediate layers), each with a different number of neurons. The first hidden layer has 250 neurons, and the last two hidden layers have 500 and 250 neurons, respectively. Each hidden layer is followed by a ReLU activation function and a Dropout layer. All hidden layers utilize the ReLU (Rectified Linear Unit) activation function to process the input data; 'nn.Dropout(0.05)' is added after each hidden layer to reduce fitting by randomly turning off a portion of the neurons. Overall, the deep neural network model in this embodiment is a feedforward neural network.
[0095] Step S423, construct the output layer:
[0096] The output layer consists of one neuron that predicts the target value (i.e., the prediction of black rot in cauliflower leaves), which is the content of a principal component. The output layer has no activation function because this is a regression task, and the output is a real value.
[0097] In addition, since the model's predicted output cannot completely correspond to the defined degree of black rot occurrence in cauliflower leaves, an upper and lower threshold range is given for the prediction of the degree of black rot, allowing for fluctuations of ±0.5.
[0098] The beneficial effects of this application are:
[0099] The method for predicting black rot in cauliflower leaves presented in this application has the advantages of being rapid, accurate, and non-destructive. It can predict and determine whether black rot has occurred in cauliflower leaves collected from the same batch within a single cycle. This method is suitable for early warning and control of the severity of black rot in large-scale field operations before symptoms appear, providing technical support for accurate monitoring and prediction of black rot in cauliflower.
[0100] Furthermore, in an exemplary embodiment, by analyzing the near-infrared spectral data of 49 cauliflower leaves that were inoculated and uninoculated within 10 days, the 490 near-infrared spectral data were divided into a training set and a test set at a ratio of 5:1. The black rot prediction model of cauliflower leaves was used to predict the current disease severity and disease time of the cauliflower leaves corresponding to each data set.
[0101] In predicting the number of days until symptoms of black rot appear in cauliflower leaves (i.e., the onset time), the cauliflower leaf black rot prediction model achieved an R² value of 0.953 and an RMSE of 0.606 on the test set. This indicates that the cauliflower leaf black rot prediction model can learn from the training set the Y-value for any test object in the test set. i,1 The prediction; at the same time, combined with the degree of the expected symptom Y i,2 The prediction of the severity of the disease can be used to obtain the time from the onset of the disease (obtained from experience) and whether the disease has occurred and the severity of the disease with an accuracy of 83%.
[0102] Furthermore, the effectiveness of the predictive model for black rot in cauliflower leaves was tested by calculating the R-value between the actual and measured values. 2 RMSE test against Y i,1 The accuracy of parameter predictions, and the degree of black rot symptoms, are determined by whether the predicted values fall within the allowable range. The prediction chart for the current stage of cauliflower leaves is shown below. Figure 5 As shown in the diagram, this is a prediction of the severity of black rot symptoms in cauliflower leaves. Figure 6 As shown.
[0103] Based on the same inventive concept, this application also provides a cauliflower leaf black rot prediction system for implementing the aforementioned method for predicting black rot in cauliflower leaves. The solution provided by this system is similar to the solution described in the above method. Therefore, the specific limitations of one or more embodiments of the cauliflower leaf black rot prediction system provided below can be found in the limitations of the cauliflower leaf black rot prediction method described above, and will not be repeated here.
[0104] In one exemplary embodiment, a black rot prediction system for cauliflower leaves is provided, comprising:
[0105] The data acquisition unit is used to acquire near-infrared spectral data of the cauliflower leaves to be tested.
[0106] The preprocessing unit is used to preprocess the near-infrared spectral data of the cauliflower leaves to be tested, and obtain the preprocessed near-infrared spectral data of the cauliflower leaves to be tested.
[0107] The screening unit is used to perform band screening on the pre-processed near-infrared spectral data of the cauliflower leaves to be tested, so as to obtain near-infrared spectral data of the cauliflower leaves to be tested that meet the preset band range.
[0108] The prediction unit is used to input the near-infrared spectral data of the cauliflower leaves to be tested, which meet the preset band range, into the black rot prediction model of cauliflower leaves, and output the black rot prediction results of cauliflower leaves; the black rot prediction results of cauliflower leaves include: the predicted value of the onset time of black rot of cauliflower leaves and the predicted value of the severity of black rot; the black rot prediction model of cauliflower leaves is obtained by training a deep neural network model using a training set.
[0109] In one exemplary embodiment, a computer device is provided, including: 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 a method for predicting black rot in cauliflower leaves.
[0110] In one exemplary embodiment, a computer-readable storage medium is provided having a computer program stored thereon that, when executed by a processor, implements a method for predicting black rot in cauliflower leaves.
[0111] In one exemplary embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements a method for predicting black rot in cauliflower leaves.
[0112] In one exemplary embodiment, a computer device is provided, which may be a server or a terminal, and its internal structure diagram may be as follows. Figure 7 As shown, this computer device includes a processor, memory, input / output (I / O) interfaces, and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides the environment for the operating system and computer programs stored in the non-volatile storage media. The I / O interfaces are used for exchanging information between the processor and external devices. The communication interface is used for communicating with external terminals via a network connection. When the computer program is executed by the processor, it implements a method for predicting black rot in cauliflower leaves.
[0113] Those skilled in the art will understand that Figure 7 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.
[0114] 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, data stored, data displayed, 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 the relevant data must comply with relevant regulations.
[0115] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can 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 can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM).
[0116] The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.
[0117] The technical features of the above embodiments can be combined in any way. For the sake of brevity, 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.
[0118] This document uses specific examples to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. Furthermore, those skilled in the art will recognize that, based on the ideas of this application, 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 this application.
Claims
1. A method for predicting black rot disease in cauliflower leaves, characterized in that, The method for predicting black rot in cauliflower leaves includes: Near-infrared spectral scanning was performed on the cauliflower leaves to be tested, and the near-infrared spectral data of the cauliflower leaves to be tested were obtained. The near-infrared spectral data of the cauliflower leaves to be tested were preprocessed to obtain the preprocessed near-infrared spectral data of the cauliflower leaves to be tested. The near-infrared spectral data of the pre-processed cauliflower leaves to be tested were filtered by band to obtain near-infrared spectral data of the cauliflower leaves to be tested that meet the preset band range. Near-infrared spectral data of cauliflower leaves within a preset wavelength range are input into a black rot prediction model for cauliflower leaves, and the prediction results for black rot are output. The black rot prediction results for cauliflower leaves include: the predicted time of onset of black rot and the predicted severity of black rot. The black rot prediction model for cauliflower leaves is obtained by training a deep neural network model using a training set. The onset time of black rot is the difference between the time when black rot symptoms become obvious and the time when the black rot fungus is inoculated.
2. The method for predicting black rot in cauliflower leaves according to claim 1, characterized in that, The near-infrared spectral data of the cauliflower leaves to be tested were preprocessed to obtain the preprocessed near-infrared spectral data of the cauliflower leaves, specifically including: The near-infrared spectral data of the cauliflower leaves to be tested were processed using the multivariate scattering correction method to obtain the corrected near-infrared spectral data of the cauliflower leaves to be tested. The near-infrared spectral data of the cauliflower leaves under test were processed using the standard normal variable method to obtain preprocessed near-infrared spectral data.
3. The method for predicting black rot in cauliflower leaves according to claim 1, characterized in that, The near-infrared spectral data of preprocessed cauliflower leaves were filtered by competitive adaptive reweighted sampling method.
4. The method for predicting black rot in cauliflower leaves according to claim 1, characterized in that, The training process of the prediction model for black rot in cauliflower leaves specifically includes: Construct a training set; the training set includes: near-infrared spectral sample data of cauliflower leaves that meet the preset sample bands, the true value of the onset time of black rot in cauliflower leaves, and the true value of the severity of black rot in cauliflower leaves; wherein, the preset sample bands are obtained by merging the characteristic bands corresponding to the inoculation time of black rot fungi and the characteristic bands corresponding to the severity of black rot. Construct a deep neural network model; The deep neural network model was trained using the training set to obtain a prediction model for black rot disease in cauliflower leaves.
5. The method for predicting black rot in cauliflower leaves according to claim 4, characterized in that, Constructing the training set specifically includes: The sample cauliflower leaves were inoculated with black rot pathogens. After the inoculation period is set, the near-infrared spectral scanning of the inoculated cauliflower leaves is performed daily to obtain the near-infrared spectral data of the inoculated cauliflower leaves. The true values of the onset time of black rot and the true values of the severity of black rot in the cauliflower leaves are also recorded. The near-infrared spectral data of the inoculated cauliflower leaves were preprocessed to obtain the preprocessed near-infrared spectral data of the cauliflower leaves. The near-infrared spectral data of the preprocessed cauliflower leaf samples were filtered by band to obtain near-infrared spectral data of cauliflower leaf samples that met the preset bands. The training set is constructed based on near-infrared spectral data of cauliflower leaves that meet multiple preset wavelength bands, the true values of the onset time of black rot in cauliflower leaves, and the true values of the severity of black rot in cauliflower leaves.
6. The method for predicting black rot in cauliflower leaves according to claim 5, characterized in that, The deep neural network model is trained using the training set to obtain a prediction model for black rot in cauliflower leaves, specifically including: Using the near-infrared spectral data of cauliflower leaves that meet the preset wavelength bands as input, and the true values of the onset time and severity of black rot in the corresponding cauliflower leaves as output, the deep neural network model is trained until the training count reaches the stopping condition, thus obtaining the black rot prediction model for cauliflower leaves.
7. A system for predicting black rot in cauliflower leaves, characterized in that, The black rot prediction system for cauliflower leaves is used to implement the black rot prediction method for cauliflower leaves according to any one of claims 1-6, wherein the black rot prediction system for cauliflower leaves comprises: The data acquisition unit is used to acquire near-infrared spectral data of the cauliflower leaves to be tested; The preprocessing unit is used to preprocess the near-infrared spectral data of the cauliflower leaves to be tested, and obtain the preprocessed near-infrared spectral data of the cauliflower leaves to be tested. The screening unit is used to perform band screening on the near-infrared spectral data of the pre-processed cauliflower leaves to be tested, so as to obtain the near-infrared spectral data of the cauliflower leaves to be tested that meet the preset band range. The prediction unit is used to input near-infrared spectral data of cauliflower leaves that meet the preset band range into the black rot prediction model of cauliflower leaves, and output the black rot prediction results of cauliflower leaves; the black rot prediction results of cauliflower leaves include: the predicted value of the onset time of black rot of cauliflower leaves and the predicted value of the severity of black rot; the black rot prediction model of cauliflower leaves is obtained by training a deep neural network model using a training set; the onset time of black rot of cauliflower leaves is the difference between the time when the black rot symptoms are obvious and the time when the black rot fungus is inoculated.
8. A computer device, comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that the processor executes the computer program to implement the method for predicting black rot in cauliflower leaves according to any one of claims 1-6.
9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the computer program implements the method for predicting black rot in cauliflower leaves as described in any one of claims 1-6.
10. A computer program product, comprising a computer program, characterized in that, When executed by a processor, the computer program implements the method for predicting black rot in cauliflower leaves as described in any one of claims 1-6.