Method and system for identifying resistance grade of cucumber to meloidogyne hapla based on characteristic electrical signal

Abstract

The present application relates to the technical field of cucumber root-knot nematode resistance grade identification scheme design, and particularly relates to a method and system for identifying the resistance grade of cucumber root-knot nematode based on characteristic electrical signals. The method comprises: applying an excitation electrical signal of a preset frequency band to a cucumber plant and collecting a response signal thereof; extracting an electrical characteristic related to the physiological state of the plant from the response signal; inputting the electrical characteristic into a pre-trained resistance grade identification model to directly output the resistance grade of the plant to root-knot nematode. The system is a special device for implementing the method. The present application utilizes the internal correlation between the electrical physiological characteristics of plants and genetic resistance to achieve intelligent identification of the resistance of cucumber to root-knot nematode at the seedling stage in a fast, non-destructive and high-throughput manner, and overcomes the shortcomings of long identification period, plant destruction and poor universality of molecular markers in traditional inoculation identification, thereby providing an efficient and reliable technical means for cucumber disease resistance breeding.

Description

Technical Field

[0001] This invention relates to the technical field of cucumber root-knot nematode resistance level identification scheme design, specifically to a method and system for cucumber root-knot nematode resistance level identification based on characteristic electrical signals. Background Technology

[0002] Cucumber is an important global vegetable crop; however, its production is constantly threatened by soil-borne diseases, among which the disease caused by the southern root-knot nematode is particularly prominent. This nematode infects cucumber roots, forming root knots and damaging the normal vascular tissue, leading to impaired water and nutrient absorption. Above-ground symptoms include wilting, yellowing, and stunted growth, and in severe cases, significant yield reduction or even total crop failure. The most economical, effective, and environmentally friendly strategy for controlling this disease is to cultivate and plant disease-resistant varieties. Therefore, rapidly and accurately identifying the root-knot nematode resistance level of cucumber germplasm resources or hybrids is a core aspect of disease-resistant breeding.

[0003] Currently, the identification of cucumber resistance to southern root-knot nematodes mainly relies on the traditional greenhouse potted artificial inoculation method. This method typically requires inoculating the test nematode into seedling pots, and after cultivation for weeks or even months, assessing the resistance level by digging up the roots, washing them, and observing and counting the number of root knots or disease indices. Although this method is considered the "gold standard" for resistance identification, its inherent defects are quite obvious: the identification cycle is too long, seriously delaying the breeding process; the operation is cumbersome, requiring a large amount of manual labor and demanding biosafety protection; more importantly, this method is destructive, and the identified plants cannot be used for subsequent breeding programs, which is particularly disadvantageous in the identification of superior strains or valuable germplasm resources in higher generations. Although molecular marker-assisted selection technology can achieve early and non-destructive screening to some extent, its application is limited by the need to pre-develop specific molecular markers closely linked to resistance genes. However, resistance is usually controlled by multiple genes, and its stability varies in different genetic backgrounds, resulting in weak universality and high development and verification costs.

[0004] In recent years, some studies have explored the use of plant physiological or biophysical methods for disease monitoring, such as indirect detection based on leaf spectra or canopy temperature. However, these methods are mostly used for assessing the severity of disease after its occurrence, rather than for early identification of the plant's intrinsic genetic resistance potential. While electrical detection technology, as an emerging approach, has been applied in monitoring plant water stress and frost damage, its use for the direct, non-destructive, and quantitative identification of plant disease resistance, especially the genetic resistance level of soil-borne diseases, remains an underexplored area. Existing sporadic studies lack systematic methodological development and have failed to address key issues such as how to extract characteristic parameters related to resistance stability from complex plant electrical signals and how to establish a reliable quantitative relationship model between electrical characteristics and resistance levels. Therefore, there is an urgent need in this field for a new method and system that can overcome the shortcomings of existing technologies and achieve rapid, non-destructive, high-throughput, and accurate identification of cucumber resistance to root-knot nematodes.

[0005] Therefore, existing technologies still need further development. Summary of the Invention

[0006] The purpose of this invention is to overcome the above-mentioned technical deficiencies and provide a method and system for identifying the resistance level of cucumber root-knot nematodes based on characteristic electrical signals, so as to solve the problems existing in the prior art.

[0007] To achieve the above-mentioned technical objectives, according to a first aspect of the present invention, the present invention provides a method for identifying the resistance level of cucumber root-knot nematodes based on characteristic electrical signals, comprising:

[0008] S1. Apply an excitation electrical signal with a preset waveform and frequency band to a predetermined part of the cucumber plant to be tested through at least two measuring electrodes;

[0009] S2. Collect the response electrical signal generated by the plant in response to the excitation electrical signal;

[0010] S3. Extract at least one electrical feature related to the plant's physiological state from the response electrical signal;

[0011] S4. Input the electrical characteristics into the pre-trained resistance level identification model to obtain and output the root-knot nematode resistance level of the cucumber plant to be tested.

[0012] Specifically, in step S1, the excitation electrical signal is an AC scanning signal containing multiple frequency points.

[0013] Specifically, the frequency range of the AC scanning signal covers 1 kHz to 1 MHz.

[0014] Specifically, in step S3, the electrical characteristics include characteristic parameters extracted from the impedance spectrum.

[0015] Specifically, the characteristic parameters include at least one of the following: impedance amplitude at a specific frequency, impedance phase angle, and relaxation frequency of the spectrum curve.

[0016] Specifically, step S3 further includes: calculating a comprehensive dynamic resistance index based on the extracted feature parameters.

[0017] Specifically, the formula for calculating the Dynamic Resistance Index (DRI) is as follows: ,in, The relaxation frequency, This is the normalized impedance value. The peak value of the phase angle. , , These are the preset weighting coefficients.

[0018] Specifically, in step S1, the predetermined part is the base of the plant stem or a leaf with a specific function.

[0019] Specifically, the resistance level identification model is a machine learning model trained on a sample dataset, which contains electrical characteristic data of multiple cucumber plants with known resistance levels.

[0020] According to a second aspect of the present invention, a system for identifying the resistance level of cucumber root-knot nematodes based on characteristic electrical signals is provided, comprising:

[0021] The signal excitation module is used to apply an excitation electrical signal with a preset waveform and frequency band to a predetermined part of the cucumber plant under test through at least two measuring electrodes.

[0022] The signal acquisition module is used to acquire the response electrical signal generated by the plant in response to the excitation electrical signal;

[0023] The feature extraction module is used to extract at least one electrical feature related to the physiological state of the plant from the response electrical signal;

[0024] The resistance analysis module integrates the resistance level identification model, which is used to receive the electrical characteristics and output the root-knot nematode resistance level of the cucumber plant to be tested.

[0025] Beneficial effects:

[0026] The most significant advantage of this invention lies in its ability to achieve early, rapid, non-destructive, and high-throughput identification of cucumber root-knot nematode resistance levels. This method can complete individual plant testing within minutes of the seedling stage, eliminating the need to wait for disease symptoms to appear, thus significantly shortening the identification cycle and accelerating the breeding screening process. The entire measurement process does not cause any physical damage to the plants; the tested plants can continue to grow normally and be used for subsequent hybridization or propagation, greatly improving the utilization efficiency of valuable breeding materials. This method is simple to operate, easy to standardize, and facilitates rapid screening of large-scale populations, meeting the modern breeding demand for high-throughput phenotypic analysis.

[0027] Another key benefit of this invention is the high accuracy and reliability of its identification results. By employing multi-frequency AC signal scanning, impedance spectrum information that comprehensively reflects the microscopic physiological state of plant tissues is obtained, overcoming the limitations of measuring single electrical parameters. Furthermore, multiple stable electrical characteristic parameters closely related to cell membrane integrity, cell structure, and ionic environment are extracted from the spectrum, and even fused into a comprehensive resistance index. These characteristics are intrinsically linked to the plant's disease resistance physiological mechanisms. Moreover, machine learning models are used to uncover the complex nonlinear mapping relationship between electrical characteristics and resistance levels, avoiding subjective judgment and making the identification conclusions more objective, accurate, and reliable.

[0028] This invention also possesses excellent practicality and robustness. By designing a dedicated background noise acquisition and digital filtering denoising process, environmental interference is effectively suppressed, improving the signal-to-noise ratio and stability measured in unshielded actual breeding environments (such as greenhouses and fields). Integrating the entire method into a portable automated system, forming a "one-click" handheld device, significantly lowers the barrier to entry, enabling breeders without specialized electrical knowledge to easily and accurately complete identification work, greatly promoting the dissemination and application of the technology. In summary, this invention provides a highly efficient technical tool with disruptive potential for cucumber root-knot nematode resistance breeding, fundamentally improving the efficiency and process of traditional resistance identification methods. Attached Figure Description

[0029] Figure 1 This is a flowchart illustrating the method for identifying the resistance level of cucumber root-knot nematodes based on characteristic electrical signals provided in a specific embodiment of the present invention.

[0030] Figure 2 This is a schematic diagram of the system composition of the cucumber root-knot nematode resistance level identification system based on characteristic electrical signals provided in a specific embodiment of the present invention. Detailed Implementation

[0031] To enable those skilled in the art to better understand the technical solutions of the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings. Based on the embodiments in this application, other similar embodiments obtained by those skilled in the art without creative effort should all fall within the scope of protection of this application. Furthermore, directional terms mentioned in the following embodiments, such as "up," "down," "left," and "right," are only for reference to the directions in the accompanying drawings; therefore, the directional terms used are for illustrative purposes and not for limiting the invention.

[0032] The present invention will be further described below with reference to the accompanying drawings and preferred embodiments.

[0033] Please see Figure 1 This invention provides a method for identifying the resistance level of cucumber root-knot nematodes based on characteristic electrical signals, comprising:

[0034] S1. Apply an excitation electrical signal with a preset waveform and frequency band to a predetermined part of the cucumber plant to be tested through at least two measuring electrodes.

[0035] It should be further explained that step S1 includes the following scheme design:

[0036] ① Measuring electrode: An Ag / AgCl disc electrode with a diameter of 8 mm is preferred. Ag / AgCl is chosen because it has the advantages of stable half-cell potential and small polarization effect. An electrode diameter of 8 mm is the preferred value that balances the measurement area (reducing contact impedance) and the size of the plant stem (avoiding short circuits between electrodes).

[0037] ② Electrode Installation: Using elastic bandages, fix two electrodes to the stem between the first and second nodes below the cotyledons of the cucumber plant to be tested, with a spacing of 20 mm. Apply a special conductive paste (e.g., 0.9% physiological saline gel) between the electrodes and the stem skin to ensure continuous and good electrical contact. The base of the stem was chosen because the tissue structure in this area is relatively uniform, the moisture status is stable, and it is less affected by instantaneous environmental changes (such as light and temperature fluctuations). The 20 mm spacing was optimized to ensure that the current path passes through a representative tissue volume and is easy to implement on seedling stems.

[0038] ③ Excitation electrical signal: Generated by a high-precision function generator / impedance analyzer. Its preset waveform is a sine wave, as a sine wave is a pure wave with a single frequency, facilitating frequency domain analysis. The frequency band is preferably set to 10kHz to 500kHz. The signal amplitude (peak voltage) is preferably 100mV. 100mV is chosen because this voltage is far below the excitation threshold of plant cells (usually above 1V), preventing electrical stimulation damage to the plant while still generating a sufficiently strong response signal for detection, resulting in a high signal-to-noise ratio.

[0039] S2. Collect the response electrical signal generated by the plant in response to the excitation electrical signal.

[0040] It should be further explained that step S2 specifically includes:

[0041] The voltage response signal between the two measurement electrodes is acquired using a preamplifier with high input impedance (≥100MΩ). The high input impedance ensures that the current drawn by the measurement circuit from the plant is minimal, avoiding measurement errors caused by load effects.

[0042] Furthermore, the voltage signal is sampled using a 16-bit analog-to-digital converter (ADC). The sampling rate is set to 5 MS / s (5 million samples per second). This sampling rate is chosen to satisfy the Nyquist sampling theorem, which states that the sampling rate must be at least twice the highest frequency of the excitation signal (500 kHz). Here, 10 times (5 MHz) is chosen to ensure waveform integrity. The sampling duration for each frequency point is 100 ms to ensure that multiple complete signal cycles are acquired for averaging and to suppress random noise.

[0043] S3. Extract at least one electrical feature related to the plant's physiological state from the response electrical signal.

[0044] It should be further explained that step S3 specifically includes:

[0045] First, the complex impedance Z is calculated from the acquired time-domain voltage signal and the known excitation current signal (since it is a constant current excitation, the current is known and constant).

[0046] The complex impedance Z can be expressed as: , where R is resistance (real part) and X is reactance (imaginary part). It is the imaginary unit.

[0047] Impedance amplitude and phase angle Calculated using the following formula:

[0048] in, It represents the magnitude of impedance, with the unit being ohms (Ω), which reflects the overall ease or difficulty of an electric current passing through plant tissue. It is the impedance phase angle, measured in degrees (°) or radians (rad). It reflects the phase delay of current relative to voltage and is closely related to the capacitance of tissue cells (mainly caused by the cell membrane).

[0049] Furthermore, by scanning 31 frequency points (logarithmically spaced, e.g., 10, 12.6, 15.8, 20.0, ..., 500 kHz) within the range of 10 kHz to 500 kHz, impedance amplitude spectrum curves and phase angle spectrum curves are obtained. These 31 points are sufficient to clearly depict the spectrum shape while ensuring measurement efficiency.

[0050] S4. Input the electrical characteristics into the pre-trained resistance level identification model to obtain and output the root-knot nematode resistance level of the cucumber plant to be tested.

[0051] It should be further explained that the "pre-trained resistance level identification model" is a mathematical model that maps extracted electrical features to discrete resistance levels. This model is trained on a sample dataset (i.e., the training set) with known resistance levels using a machine learning algorithm.

[0052] Furthermore, resistance levels are typically divided into three grades: "Highly resistant (HR)," "Moderately resistant (MR)," and "Susceptible (S)." The grading criteria are based on the traditional greenhouse inoculation disease index (DI): DI < 10% is HR, 10% ≤ DI < 30% is MR, and DI ≥ 30% is S.

[0053] Furthermore, during implementation, the feature vector extracted in step S3 is input into the model, and the model will output a probability distribution or a direct rank label. For example, the model output might be [HR:0.85, MR:0.14, S:0.01], which would be classified as "high resistance".

[0054] Understandably, this invention provides a standardized and operational electrical identification procedure for cucumber root-knot nematode resistance. By precisely controlling the electrodes, signal parameters, and measurement process, the repeatability and accuracy of the results are ensured, laying a methodological foundation for achieving high-throughput, non-destructive resistance identification.

[0055] Specifically, in step S1, the excitation electrical signal is an AC scanning signal containing multiple frequency points.

[0056] It should be further noted that this invention imposes important limitations on the excitation signal. Specifically, it employs a frequency scanning mode, rather than single-frequency point measurement. The preferred scanning method is "logarithmically equally spaced scanning," meaning that each frequency point is equally spaced on a logarithmic coordinate system. For example, scanning 31 points from 10kHz to 500kHz, the interval between adjacent frequency points on the logarithmic coordinate system is... Logarithmic scanning was chosen because the impedance spectrum of biological tissues varies over a wide frequency range. Logarithmic scanning can obtain sufficient data points in both the low-frequency (drastic changes) and high-frequency (gradual changes) regions, thus more accurately depicting the entire spectral morphology. Specific scanning sequence examples (first 5 points and last point): 10.0kHz, 12.6kHz, 15.8kHz, 20.0kHz, 25.1kHz, ..., 500.0kHz.

[0057] Understandably, logarithmic scanning can efficiently capture impedance characteristics across the entire frequency band with the fewest points, making it particularly suitable for characterizing biological tissues with relaxation processes, and providing a high-quality data foundation for subsequent extraction of robust feature parameters.

[0058] Specifically, the frequency range of the AC scanning signal covers 10kHz to 500kHz.

[0059] It should be further noted that the present invention further preferably uses a frequency range. The specific reasons for choosing 10kHz to 500kHz are as follows:

[0060] 1. Avoiding low-frequency polarization effects: Below 10kHz, the polarization impedance at the electrode-electrolyte interface increases significantly, severely interfering with the measurement of the tissue's own impedance. Above 10kHz, the polarization impedance decreases rapidly, ensuring a good signal-to-noise ratio.

[0061] 2. Covering the Key β-Relaxation Region: The dielectric spectrum of plant tissues exhibits a major relaxation region in the kHz-MHz frequency band, called β-relaxation. This is primarily related to the capacitance charging and discharging of the cell membrane and the interfacial polarization of intracellular and extracellular ions. Nematode infection or resistance responses directly affect microstructural changes such as cell membrane integrity and cell size / density, significantly influencing the characteristic frequencies of β-relaxation (typically in the range of tens to hundreds of kHz). Therefore, 10kHz-500kHz is considered the "golden frequency band" for monitoring these changes.

[0062] 3. Avoid the complexity of high-frequency measurements: When the frequency is higher than 500kHz, the electromagnetic radiation effect is enhanced, the design of the measurement circuit (such as shielding and wiring) becomes more complicated, the cost increases, and the signal is easily shunted, resulting in a decrease in measurement sensitivity.

[0063] Understandably, this frequency range precisely targets the electrical response band most relevant to plant disease resistance physiology, maximizing the acquisition of highly discriminative information while ensuring measurement feasibility and economy.

[0064] Specifically, in step S3, the electrical characteristics include characteristic parameters extracted from the impedance spectrum.

[0065] It should be further explained that this invention clarifies the source of the electrical characteristics. Specifically, the impedance data obtained from scanning at 31 frequency points (31...) Values ​​and 31 Instead of directly using all 61 original data points, a continuous spectral curve is fitted to the data (values), and then a small number of feature parameters with clear biophysical significance are extracted from this curve. This is done to reduce the dimensionality of the data, avoid the "curse of dimensionality," and improve the training efficiency and generalization ability of subsequent models. A commonly used fitting model is the Cole-Cole model, whose complex impedance expression is:

[0066]

[0067] in:

[0068] It is a complex impedance;

[0069] It is the DC frequency ( The limiting resistance at Hz mainly reflects the resistance of the extracellular fluid pathway;

[0070] It is an infinitely high frequency ( The limiting resistance at Hz reflects the total resistance of current passing through both intracellular and extracellular fluid pathways simultaneously.

[0071] It is the relaxation time constant and the relaxation frequency. The relationship is ;

[0072] It is the distribution coefficient (0 < ≤1), describing the distribution of relaxation time, Indicates a single relaxation time;

[0073] , is the angular frequency;

[0074] It is the imaginary unit. By fitting the experimental data to the Cole-Cole model using the nonlinear least squares method, we can obtain... , , , These four characteristic parameters characterize the electrical properties of the tissue from a physical mechanism perspective.

[0075] Understandably, using Cole-Cole model parameters as features gives the features clear physiological and physical meaning, directly correlates with the tissue's microstructure (cell morphology, membrane properties), and enhances the interpretability and reliability of the method.

[0076] Specifically, the feature parameters include at least one of the following:

[0077] Impedance amplitude, impedance phase angle, and relaxation frequency of the spectrum curve at a specific frequency.

[0078] It should be further noted that this invention lists more specific feature parameters. As a supplement or alternative to the Cole-Cole model parameters, intuitive features can be directly extracted from the raw spectral data. Preferred specific feature parameters and their extraction methods are as follows:

[0079] ① Impedance amplitude at a specific frequency ( ): Directly read the impedance amplitude at a frequency of 50kHz. 50kHz was chosen because it is in the core region of β relaxation and is very sensitive to the state of the cell membrane.

[0080] ② Impedance phase angle at a specific frequency ( ): Directly read the phase angle value at a frequency of 100kHz. At 100kHz, the phase angle is usually close to its peak value (negative direction), which can well reflect the capacitance characteristics of the cell.

[0081] ③ The relaxation frequency of the spectrum curve ( This refers to the frequency at which the phase angle spectrum curve reaches its minimum value (most negative point). The precise value can be obtained by finding the index of the minimum value in the phase angle array and then interpolating. . Closely related to cell size, the larger the cell, the better. The lower the value. A preferred set of combined feature vectors is: It includes five characteristics: 10kHz impedance, 100kHz impedance, 50kHz phase, peak phase angle, and relaxation frequency.

[0082] Understandably, these features have clear physical meanings, are simple and quick to calculate, do not require complex curve fitting, and can effectively characterize different aspects of the spectrum, providing rich and complementary information for the model.

[0083] Specifically, step S3 further includes: calculating a comprehensive dynamic resistance index based on the extracted feature parameters.

[0084] It should be further explained that this invention proposes a more advanced feature construction scheme. Specifically, it fuses the aforementioned multiple feature parameters into a single scalar index, called the "dynamic resistance index." Its core advantage lies in compressing multi-dimensional information into one dimension, making the judgment of resistance level very intuitive, and even allowing for rapid initial screening by setting a fixed threshold. A specific, operable fusion formula (linear weighted sum) is as follows:

[0085]

[0086] in:

[0087] The Dynamic Resistance Index (DRI) is a dimensionless numerical value. Empirical evidence shows that a higher DRI value generally indicates stronger resistance.

[0088] Impedance ratio, the ratio of high-frequency impedance to low-frequency impedance, reflects the relative contribution of intracellular and extracellular current pathways and can eliminate some of the influence of individual size and water content. This value may be higher in resistant varieties.

[0089] The absolute value of the peak phase angle characterizes the integrity and activity of the cell membrane; a larger value indicates a better membrane structure and potentially stronger resistance.

[0090] The commonly used logarithm of relaxation frequency indicates resistant varieties with more compact cell structure and smaller volume. It may be too high;

[0091] The weighting coefficients need to be determined using training data, for example, by determining a set of optimal values ​​through principal component analysis (PCA) or correlation analysis with known resistance grades. These weights reflect the relative importance of each feature in distinguishing resistance levels.

[0092] Understandably, the DRI index integrates information from multiple key electrical parameters, making it more robust to measurement errors and individual differences, allowing non-professionals to make quick judgments through simple numerical comparisons.

[0093] Specifically, the formula for calculating the Dynamic Resistance Index (DRI) is as follows:

[0094]

[0095] in, The relaxation frequency, This is the normalized impedance value. The peak value of the phase angle. These are the preset weighting coefficients.

[0096] It should be further noted that this invention provides a specific example of a DRI calculation formula. Each term within it is precisely and practically defined:

[0097] Relaxation frequency. Unit: kHz, as defined in the characteristic parameters. ;

[0098] Normalized impedance value, specifically referring to the impedance amplitude at a frequency of 100kHz. With impedance amplitude at 10kHz The ratio, i.e. This is a dimensionless number;

[0099] Peak phase angle. Measured in degrees (°), it refers to the absolute value of the minimum phase angle (i.e., the peak value in the negative direction) reached across the entire measurement frequency band. For example, if the minimum phase angle is -15°, then... ;

[0100] The weighting coefficients, whose optimal values ​​are determined as follows: A training set containing n samples is collected, each sample having a known resistance level (e.g., 1, 2, 3 representing susceptible, moderately resistant, and highly resistant, respectively). The weighting coefficients for each sample are calculated. , , Then, a multiple linear regression was performed, with the resistance level (1, 2, 3) as the dependent variable Y and the three characteristics as independent variables X, to solve for the coefficients. This makes the equation The goodness of fit (R²) is the highest. A typical set of values ​​obtained through this process might be: Reason: This method is data-driven and ensures optimal linear correlation between DRI and true resistance level.

[0101] Understandably, the formula is clear and easy to calculate. The weights are determined through standard regression analysis, which ensures the scientific nature and objectivity of the DRI index, making it a reliable quantitative indicator for resistance grading.

[0102] Specifically, in step S1, the predetermined part is the base of the plant stem or a leaf with a specific function.

[0103] It should be further explained that the preferred measurement sites in this invention are as follows: For the stem base: as described above, located in the first-to-second internode below the cotyledons. For the specific functional leaf: select the third fully unfolded true leaf counting down from the growing point, avoiding the midrib, and install the electrodes on the leaf mesophyll on both sides of the middle of the leaf. The reason for choosing the third leaf is that this leaf position has strong physiological activity, vigorous metabolism, and is sensitive to stress; it is also fully unfolded, relatively stable in size, and easy to standardize measurement. Regarding the selection of stem and leaf measurements: leaf measurements are preferred in the early seedling stage or when the stems are too thin. Stem measurements generally have better repeatability. In practical applications, a single measurement site can be consistently used on the same batch of materials.

[0104] Understandably, the provision of alternative, validated, and effective measurement sites increases the flexibility and adaptability of the method, making it suitable for cucumber plants at different growth stages.

[0105] Specifically, before step S1, a calibration step is also included: before applying the excitation electrical signal, a segment of the plant's background noise signal under no excitation or minimal excitation is first collected, and in step S3, the response electrical signal is denoised based on the background noise signal.

[0106] It should be further noted that this invention incorporates important anti-interference measures. Detailed implementation:

[0107] 1. Collect background noise:

[0108] Before the frequency scan officially begins, the function generator does not output a signal (the output terminal is connected to a high impedance), and continuously acquires a voltage signal for 10ms. This signal is the noise floor. It mainly includes power frequency interference (50Hz and its harmonics), environmental electromagnetic noise and amplifier noise itself;

[0109] 2. Noise reduction: Noise reduction is performed using frequency domain spectral subtraction. Specific methods include:

[0110] ① The collected response signal and noise signals Perform Fast Fourier Transform (FFT) on each to obtain their frequency domain representations. and ;

[0111] ② Calculate the average power spectrum of the noise signal ;

[0112] ③ Subtract the power spectrum of the noise from the power spectrum of the response signal to obtain the power spectrum estimate of the denoised signal: .in It is an over-subtraction factor, with a preferred value of 1.2, to avoid residual noise caused by estimation errors;

[0113] ④ Keep The phase remains unchanged, combined with the estimated amplitude. Perform Inverse Fast Fourier Transform (IFFT) to reconstruct the denoised time-domain signal. 。;

[0114] ⑤ Subsequent impedance calculations are based on conduct.

[0115] Understandably, this digital signal processing technology can effectively suppress steady-state environmental noise and significantly improve the signal-to-noise ratio, making it possible to conduct accurate measurements in ordinary greenhouse or field environments, thus enhancing the practical value of the method.

[0116] Specifically, the resistance level identification model is a machine learning model trained on a sample dataset, which contains electrical characteristic data of multiple cucumber plants with known resistance levels.

[0117] It should be further noted that this invention defines the core construction method of the model. The specific implementation method is as follows:

[0118] 1. Sample dataset construction:

[0119] ① Collect m (e.g., m=300) plant samples from different cucumber varieties (lines);

[0120] ② Perform impedance measurement on each sample as described in this invention, and extract feature vectors (e.g., as described in the above scheme). (or the impedance amplitude, impedance phase angle, and relaxation frequency of the spectrum curve at a specific frequency as described in the above scheme).

[0121] ③ Perform traditional greenhouse pot inoculation identification on each sample and determine its true resistance grade label (HR, MR, S) based on the disease index (DI);

[0122] ④ Finally, the dataset is obtained. ,in It is an eigenvector. These are the corresponding level labels.

[0123] 2. Model Training (Taking Support Vector Machine (SVM) as an example):

[0124] Reasons for algorithm selection: SVM performs well in small sample and high-dimensional modes, and can handle nonlinear problems through kernel functions;

[0125] Specific steps:

[0126] ① Data preprocessing: Randomly shuffle the dataset D and divide it into a training set (210 samples) and a test set (90 samples) in a 7:3 ratio. Standardize the features of the training set so that the mean of each feature is 0 and the standard deviation is 1. Process the test set using the same standardization parameters.

[0127] ② Model Training: Using the training set data, a multi-class SVM model is trained with the radial basis function (RBF) as the kernel function. The main hyperparameters to be optimized are the penalty coefficient C and the kernel function parameter γ. A grid search is performed on the training set using 5-fold cross-validation to find the optimal parameters. The search range is: , Choose the one with the highest average accuracy in cross-validation. combination.

[0128] ③ Model evaluation: The trained final model is evaluated on the reserved test set (90 samples). Accuracy, precision, recall and F1 score are calculated. When the accuracy on the test set is consistently higher than 85%, the model is considered to have been successfully trained and can be used for the identification of unknown samples.

[0129] 3. Model Deployment: Integrate the final trained SVM model (including model parameters and standardized parameters) into embedded software or host computer software.

[0130] Understandably, automatically learning the complex mapping relationship between "electrical characteristics and resistance level" through data-driven methods avoids the difficulties and inaccuracies of manually formulating discrimination rules, greatly improving the accuracy and objectivity of the identification method.

[0131] Please see Figure 2 The present invention provides another embodiment, which provides a cucumber root-knot nematode resistance level identification system based on characteristic electrical signals. The cucumber root-knot nematode resistance level identification system based on characteristic electrical signals includes:

[0132] The signal excitation module 100 is used to apply an excitation electrical signal with a preset waveform and frequency band to a predetermined part of the cucumber plant to be tested through at least two measuring electrodes.

[0133] It should be further explained that the core component of the signal excitation module 100 is a direct digital frequency synthesizer controlled by a microcontroller (such as the STM32H7 series). This DDS chip (such as AD9833) can generate a sine wave with a frequency resolution of 0.1Hz. It is followed by a voltage-controlled current source circuit to convert the voltage signal generated by the DDS into an AC constant current source with an amplitude stable at 100μA (peak). The selection of a constant current source can ensure that the current flowing through the plant is constant. Then the voltage response signal on the measuring electrode is directly proportional to the impedance, which simplifies the calculation and reduces the error introduced by the change of contact resistance.

[0134] The signal acquisition module 200 is used to acquire the response electrical signal generated by the plant in response to the excitation electrical signal;

[0135] It should be further explained that the core of the signal acquisition module 200 is an instrumentation amplifier (such as INA128) and a 16-bit ADC (such as ADS8688). The gain of the instrumentation amplifier is set to 100 times and the input impedance is 100MΩ. The sampling rate of the ADC is controlled by the microcontroller and can reach up to 5MS / s.

[0136] Feature extraction module 300 is used to extract at least one electrical feature related to the physiological state of the plant from the response electrical signal;

[0137] The resistance analysis module 400 integrates the resistance level identification model, which is used to receive the electrical characteristics and output the root-knot nematode resistance level of the cucumber plant to be tested.

[0138] The feature extraction module 300 and the resistance analysis module 400 are implemented by the ARM Cortex-M7 core inside the microcontroller and the embedded algorithm software it runs on. The software flow includes:

[0139] 1. Control the DDS and ADC to work together to complete frequency scanning and data acquisition;

[0140] 2. Perform FFT, impedance calculation, and characteristic parameter extraction (e.g., calculation of...) Digital signal processing algorithms, such as , etc.

[0141] 3. Call the resistance level identification model (e.g., the parameters of a pre-trained SVM model) that is stored in Flash memory. The model takes the extracted feature vector as input, runs the inference algorithm, and obtains the resistance level.

[0142] It should be further noted that the present invention also provides human-computer interaction, and the system also includes a 3.5-inch LCD touch screen for displaying operation instructions, real-time impedance spectra, and the final resistance rating result (such as "high resistance HR").

[0143] It should be further noted that all circuit modules are integrated into a portable, battery-powered housing, forming a handheld device. The electrodes are connected to the BNC interface of the main body of the device via shielded cables.

[0144] Understandably, integrating the entire method into a dedicated, automated, portable device enables "one-click" operation. Users simply connect the electrodes and click the "measure" button on the screen, and the device can directly display the identification results within tens of seconds. This greatly lowers the technical barrier to entry, making the method easily applicable to breeders and field technicians, and truly commercializing the technology.

[0145] In a preferred embodiment, this application also provides an electronic device, the electronic device comprising:

[0146] The computer device includes a memory and a processor, wherein the memory stores computer-readable instructions that, when executed by the processor, implement the method for identifying the resistance level of cucumber root-knot nematodes based on characteristic electrical signals. The computer device can be broadly categorized as a server, terminal, or any other electronic device with the necessary computing and / or processing capabilities. In one embodiment, the computer device may include a processor, memory, network interface, communication interface, etc., connected via a system bus. The processor of the computer device can be used to provide the necessary computing, processing, and / or control capabilities. The memory of the computer device may include a non-volatile storage medium and internal memory. The non-volatile storage medium may store an operating system, computer programs, etc. The internal memory can provide an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The network interface and communication interface of the computer device can be used to connect and communicate with external devices via a network. When the computer program is executed by the processor, it performs the steps of the method of the present invention.

[0147] This invention can be implemented as a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, causes the steps of the methods of embodiments of the invention to be performed. In one embodiment, the computer program is distributed across multiple network-coupled computer devices or processors, such that the computer program is stored, accessed, and executed in a distributed manner by one or more computer devices or processors. A single method step / operation, or two or more method steps / operations, may be executed by a single computer device or processor or by two or more computer devices or processors. One or more method steps / operations may be executed by one or more computer devices or processors, and one or more other method steps / operations may be executed by one or more other computer devices or processors. One or more computer devices or processors may execute a single method step / operation, or execute two or more method steps / operations.

[0148] Those skilled in the art will understand that the method steps of this invention can be performed by a computer program instructing related hardware, such as a computer device or processor, to perform the steps of this invention when executed. Depending on the context, any references herein to memory, storage, databases, or other media may include non-volatile and / or volatile memory. Examples of non-volatile memory include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), flash memory, magnetic tape, floppy disk, magneto-optical data storage device, optical data storage device, hard disk, solid-state drive, etc. Examples of volatile memory include random access memory (RAM), external cache memory, etc.

[0149] The technical features described above can be combined arbitrarily. Although not all possible combinations of these technical features are described, any combination of these technical features should be considered to be covered by this specification, provided that such combination does not contain contradictions.

[0150] The specific embodiments of the present invention described above do not constitute a limitation on the scope of protection of the present invention. Any other corresponding changes and modifications made in accordance with the technical concept of the present invention should be included within the scope of protection of the claims of the present invention.

Claims

1. A method for identifying cucumber root-knot nematode resistance levels based on characteristic electrical signals, characterized in that, Includes the following steps: S1. Apply an excitation electrical signal with a preset waveform and frequency band to a predetermined part of the cucumber plant to be tested through at least two measuring electrodes; S2. Collect the response electrical signal generated by the plant in response to the excitation electrical signal; S3. Extract at least one electrical feature related to the plant's physiological state from the response electrical signal; S4. Input the electrical characteristics into the pre-trained resistance level identification model to obtain and output the root-knot nematode resistance level of the cucumber plant to be tested. In step S3, the electrical characteristics include characteristic parameters extracted from the impedance spectrum; The characteristic parameters include at least one of the following: Impedance amplitude, impedance phase angle, and relaxation frequency of the spectrum curve at a specific frequency; Step S3 further includes: calculating a comprehensive dynamic resistance index based on the extracted feature parameters; The formula for calculating the Dynamic Resistance Index (DRI) is as follows: ,in, The relaxation frequency, This is the normalized impedance value. The peak value of the phase angle. , , These are the preset weighting coefficients.

2. The method according to claim 1, characterized in that, In step S1, the excitation electrical signal is an AC scanning signal containing multiple frequency points.

3. The method according to claim 2, characterized in that, The frequency range of the AC scanning signal covers 1 kHz to 1 MHz.

4. The method according to claim 1, characterized in that, In step S1, the predetermined part is the base of the plant stem or a leaf with a specific function.

5. The method according to claim 1, characterized in that, The resistance level identification model is a machine learning model trained on a sample dataset, which contains electrical characteristic data of multiple cucumber plants with known resistance levels.

6. A system for identifying cucumber root-knot nematode resistance levels based on characteristic electrical signals, used to implement the method described in any one of claims 1-5, characterized in that, The system includes: The signal excitation module is used to apply an excitation electrical signal with a preset waveform and frequency band to a predetermined part of the cucumber plant under test through at least two measuring electrodes. The signal acquisition module is used to acquire the response electrical signal generated by the plant in response to the excitation electrical signal; The feature extraction module is used to extract at least one electrical feature related to the physiological state of the plant from the response electrical signal; The resistance analysis module integrates the resistance level identification model, which is used to receive the electrical characteristics and output the root-knot nematode resistance level of the cucumber plant to be tested.

Images