Methods and devices for identifying wheat diseases

Abstract

This invention provides a method and apparatus for identifying wheat diseases, applied in the field of crop disease identification technology. The method includes: acquiring a training sample set; determining the mass-to-charge ratio (M / C ratio) signal intensity data of volatile organic compounds in wheat leaf samples from the training sample set, and determining the maximum eigenvalue of each M / C ratio channel in the M / C ratio signal intensity data; mapping the maximum eigenvalues ​​of all M / C ratio channels to pixels to obtain a two-dimensional pixel matrix of wheat leaf samples; extracting the parent ion and its corresponding fragment ions from the two-dimensional pixel matrix, and rearranging the positions of the extracted pixels in the two-dimensional pixel matrix to obtain a reconstructed feature map; inputting the reconstructed feature map into a wheat disease identification model for network training, wherein the wheat disease identification model is used to identify the disease type and severity of wheat based on the M / C ratio signal features in the reconstructed feature map.

Description

Technical Field

[0001] This invention relates to the field of crop disease identification technology, and in particular to a method and apparatus for identifying wheat diseases. Background Technology

[0002] Traditional disease detection techniques, including polymerase chain reaction (PCR), enzyme-linked immunosorbent assay (ELISA), and fluorescence in situ hybridization (FISH), rely heavily on the appearance of disease symptoms, leading to technical bottlenecks in early and accurate detection. Because plant diseases are characterized by rapid onset and spread, the effective control window is often extremely short. The aforementioned techniques, due to their inherent limitations such as long detection cycles and delayed response, are no longer adequate to meet the urgent needs of modern agriculture for rapid disease early warning and intervention. Summary of the Invention

[0003] This invention provides a method and apparatus for identifying wheat diseases, which solves the problems of long detection cycles and slow response of existing technologies, making it difficult to meet the urgent needs of modern agriculture for rapid disease early warning and early intervention.

[0004] This invention provides a method for identifying wheat diseases, comprising: acquiring a training sample set, the training sample set including wheat leaf samples in different health states, each wheat leaf sample labeled with disease type and disease severity; determining the mass-to-charge ratio signal intensity data of volatile organic compounds in the wheat leaf samples in the training sample set, and determining the maximum eigenvalue of each mass-to-charge ratio channel in the mass-to-charge ratio signal intensity data, mapping the maximum eigenvalues ​​of all mass-to-charge ratio channels to pixels, obtaining a two-dimensional pixel matrix of wheat leaf samples, one wheat leaf sample corresponding to one two-dimensional pixel matrix; extracting the pixels containing the parent ion and its corresponding fragment ions from the two-dimensional pixel matrix, and rearranging the positions of the extracted pixels in the two-dimensional pixel matrix to obtain a reconstructed feature map; inputting the reconstructed feature map into a wheat disease identification model for network training, the wheat disease identification model being used to identify the disease type and disease severity of wheat based on the mass-to-charge ratio signal features in the reconstructed feature map.

[0005] According to a wheat disease identification method provided by the present invention, the step of extracting the pixels containing the parent ion and its corresponding fragment ions from the two-dimensional pixel matrix and rearranging the positions of the extracted pixels in the two-dimensional pixel matrix to obtain a reconstructed feature map includes: when prior knowledge of the target disease is determined, extracting the pixels corresponding to the mass-to-charge ratio channel for feature reconstruction based on the correlation between the parent ion and its fragment ions in the prior knowledge of the target disease; when prior knowledge of the target disease is not determined, reconstructing the pixels containing the parent ion and its corresponding fragment ions based on the generalized feature combination of volatile organic compounds released from wheat leaf samples.

[0006] According to the wheat disease identification method provided by the present invention, the prior knowledge of the target disease is derived from the analysis literature using gas chromatography-mass spectrometry.

[0007] According to the present invention, a wheat disease identification method is provided, wherein the wheat disease identification model adopts a lightweight convolutional neural network, the lightweight convolutional neural network comprising: an input layer, three convolutional layers, three max pooling layers, a flattening layer, two fully connected layers, and an output layer.

[0008] According to a wheat disease identification method provided by the present invention, determining the mass-to-charge ratio signal intensity data of volatile organic compounds in wheat leaf samples from the training sample set includes:

[0009] The mass-to-charge ratio signal intensity data of volatile organic compounds in wheat leaf samples were detected by proton transfer reaction mass spectrometry.

[0010] This invention also provides a wheat disease identification device, comprising the following modules: an acquisition module and a processing module; the acquisition module is used to acquire a training sample set, the training sample set including wheat leaf samples in different health states, each wheat leaf sample being labeled with disease type and disease severity; the processing module is used to determine the mass-to-charge ratio signal intensity data of volatile organic compounds in the wheat leaf samples in the training sample set, and determine the maximum eigenvalue of each mass-to-charge ratio channel in the mass-to-charge ratio signal intensity data, mapping the maximum eigenvalues ​​of all mass-to-charge ratio channels to pixels, obtaining a two-dimensional pixel matrix of wheat leaf samples, one wheat leaf sample corresponding to one two-dimensional pixel matrix; extracting the pixels containing the parent ion and its corresponding fragment ions from the two-dimensional pixel matrix, and rearranging the positions of the extracted pixels in the two-dimensional pixel matrix to obtain a reconstructed feature map; inputting the reconstructed feature map into a wheat disease identification model for network training, the wheat disease identification model being used to identify the disease type and disease severity of wheat based on the mass-to-charge ratio signal features in the reconstructed feature map.

[0011] According to a wheat disease identification device provided by the present invention, the processing module is used to extract the pixel points of the corresponding mass-to-charge ratio channel for feature recombination based on the correlation between the parent ion and its fragment ions in the prior knowledge of the target disease when the prior knowledge of the target disease is determined; and to reconstruct the features of the pixel points where the parent ion and its corresponding fragment ions are located based on the generalized feature combination of volatile organic compounds released from wheat leaf samples when the prior knowledge of the target disease is not determined.

[0012] According to the present invention, a wheat disease identification device is provided in which the prior knowledge of the target disease is derived from the analysis literature using gas chromatography-mass spectrometry.

[0013] According to the present invention, a wheat disease identification device is provided, wherein the wheat disease identification model adopts a lightweight convolutional neural network, the lightweight convolutional neural network comprising: an input layer, three convolutional layers, three max pooling layers, a flattening layer, two fully connected layers, and an output layer.

[0014] According to the present invention, a wheat disease identification device is provided, wherein the processing module is used to detect the mass-to-charge ratio signal intensity data of volatile organic compounds in wheat leaf samples by proton transfer reaction mass spectrometry.

[0015] The present invention also provides an electronic 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 wheat disease identification method as described above.

[0016] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the wheat disease identification method as described above.

[0017] The present invention also provides a computer program product, including a computer program that, when executed by a processor, implements the wheat disease identification method as described above.

[0018] The wheat disease identification method and apparatus provided by this invention, on the one hand, effectively suppresses background noise and signal fluctuation interference by extracting the maximum feature values ​​of each mass-to-charge ratio channel and mapping them into a two-dimensional pixel matrix, highlighting key features related to diseases and improving the sensitivity of the wheat disease identification model to weak disease signals; on the other hand, by extracting and rearranging the pixels of the parent ion and its fragment ions, the dispersed chemical signals are integrated into a structured feature map, significantly enhancing the expressive ability of disease features. Thus, high-precision early identification of wheat diseases can be achieved, providing key technical support for rapid disease warning and early intervention in modern agriculture. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0020] Figure 1 This is one of the flowcharts of the wheat disease identification method provided by the present invention.

[0021] Figure 2 This is the second flowchart of the wheat disease identification method provided by the present invention.

[0022] Figure 3 This is a schematic diagram of the structure of the wheat disease identification device provided by the present invention.

[0023] Figure 4 This is a schematic diagram of the structure of the electronic device provided by the present invention. Detailed Implementation

[0024] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions 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, 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.

[0025] It should be noted that in the embodiments of this application, the words "exemplary" or "for example" are used to indicate examples, illustrations, or explanations. Any embodiment or design scheme described as "exemplary" or "for example" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or design schemes. Specifically, the use of the words "exemplary" or "for example" is intended to present the relevant concepts in a specific manner.

[0026] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element. Furthermore, it should be noted that the scope of the methods and apparatuses in the embodiments of this application is not limited to performing functions in the order shown or discussed, but may also include performing functions substantially simultaneously or in the reverse order, depending on the functions involved. For example, the described methods may be performed in a different order than described, and various steps may be added, omitted, or combined. Additionally, features described with reference to certain examples may be combined in other examples.

[0027] To facilitate a clear description of the technical solutions of the embodiments of this application, the terms "first" and "second" are used in the embodiments of this application to distinguish the same or similar items with essentially the same function and effect. Those skilled in the art can understand that the terms "first" and "second" are not intended to limit the quantity or execution order.

[0028] This application describes some exemplary embodiments for illustrative purposes. It should be understood that this application may be implemented in other ways not specifically shown in the accompanying drawings.

[0029] like Figure 1 As shown, this application provides a method for identifying wheat diseases, which can be applied to a wheat disease identification device. The method may include steps S101-S104:

[0030] S101. The wheat disease identification device acquires a training sample set, which includes wheat leaf samples in different health states, and each wheat leaf sample is labeled with the disease type and degree of infection.

[0031] Specifically, to construct an accurate wheat disease identification model, it is necessary to scientifically collect and label wheat leaf samples at different health states. The core of this process stems from the knowledge accumulated by plant pathologists over long-term research: different wheat diseases exhibit unique visual symptoms on leaves. For example, in the early stages of wheat powdery mildew, small white mold spots appear on the leaf surface. As the disease progresses, the mold spots gradually enlarge and merge, forming a thick layer of white powdery mildew. In the early stages of stripe rust, chlorotic spots appear on the leaves, followed by the growth of bright yellow uredinia. In severe cases, the uredinia cover the entire leaf, causing the leaves to wither and turn yellow. Based on these characteristics, experts can make scientific judgments about the diseases.

[0032] In practice, sample collection must follow a rigorous process. First, representative wheat plants are selected from wheat fields at different growth stages and in different plots. For healthy wheat leaves, samples are chosen that are intact, bright green, and free of any lesions. For diseased leaves, samples are collected according to the severity of the disease at different infection stages. For example, in the early stages of powdery mildew infection, leaves with only a few small white mold spots are carefully selected; in the severe stages, samples covered with a white mold layer, with curled or even yellowed leaves are collected. During collection, the location, time, and wheat variety of the sample are recorded. After collection, the samples are sent to a professional laboratory where pathologists, based on the morphology, distribution, and color of the leaf lesions, combined with their professional knowledge and experience, label the disease type and severity of each sample. During labeling, microscopes and other tools can be used to observe the microstructure of the lesion tissue to further confirm the disease type and ensure the accuracy of the labeling.

[0033] Based on the above approach, on the one hand, it comprehensively covers samples of various disease types and their different infection stages, providing a rich data foundation for the constructed training sample set. This enables the subsequently constructed wheat disease identification model to learn more comprehensive and diverse disease characteristics, improving the model's identification accuracy and generalization ability. On the other hand, relying on the professional knowledge of pathology experts for annotation ensures the scientific rigor and reliability of sample annotation, avoiding the problem of model training deviation due to inaccurate annotation.

[0034] S102. The wheat disease identification device determines the mass-to-charge ratio signal intensity data of volatile organic compounds in wheat leaf samples from the training sample set, and determines the maximum eigenvalue of each mass-to-charge ratio channel in the mass-to-charge ratio signal intensity data. The maximum eigenvalues ​​of all mass-to-charge ratio channels are mapped to pixels to obtain a two-dimensional pixel matrix of wheat leaf samples. One wheat leaf sample corresponds to one two-dimensional pixel matrix.

[0035] Studies have found that plants release specific volatile organic compounds (VOCs) when subjected to pathogen stress. These VOCs, as important signaling molecules between plants or between plants and the environment, can reflect the physiological state of plants before diseases show obvious symptoms.

[0036] Optionally, the wheat disease identification device can use proton transfer reaction mass spectrometry (PTR-MS) to detect the mass-to-charge ratio signal intensity data of volatile organic compounds (VOCs) in wheat leaf samples. Specifically, wheat can release VOCs of varying compositions and concentrations under healthy conditions and different disease infection states. PTR-MS technology can utilize hydrated hydrogen ions (… As a reagent ion, it undergoes a proton transfer reaction with VOCs molecules in wheat leaf samples, causing the VOCs molecules to ionize and be detected by a mass spectrometer, thus obtaining mass-to-charge ratio signal intensity data.

[0037] like Figure 2 As shown, after acquiring the mass-to-charge ratio (MCR) signal intensity data, the wheat disease identification device can perform data preprocessing. Since the signal is unstable in the initial detection phase, this data needs to be discarded, retaining only the MCR signal intensity data from the stable phase. Then, the maximum eigenvalue of each MCR channel is extracted as the effective eigenvalue. Finally, the maximum eigenvalues ​​of all MCR channels are mapped to pixels in an RGB three-channel pseudo-color image, arranged in a fixed order within a two-dimensional image grid of size H×W. For example, if the image size is set to 100×100, the maximum eigenvalues ​​of different MCR channels are mapped to corresponding pixel positions, forming a two-dimensional pixel matrix of the wheat leaf sample.

[0038] The specific processing procedure is as follows: First, wheat leaf samples labeled with disease type and severity are placed one by one into a 50mL sealed gas collection bottle. For example, healthy leaves, leaves infected with early-stage powdery mildew, and leaves severely infected with powdery mildew are placed in different gas collection bottles and left to stand at room temperature for 30-40 minutes to allow the leaves to fully release VOCs. The gas collection bottle is equipped with inlet and outlet ports for easy connection to the PTR-MS sample introduction system. After enrichment, the inlet of the PTR-MS system is connected to the outlet of the gas collection bottle, the detection program is started, and the measurement is performed continuously for 2 minutes. Key parameters are set during detection: the mass-to-charge ratio detection range is 20–150, which covers the mass-to-charge ratio range of common VOCs; the drift tube voltage is 450V, and the single mass-to-charge ratio signal scan time is 100ms to ensure the sensitivity and accuracy of the detection.

[0039] Based on the above scheme, in terms of detection efficiency, compared with the traditional method relying on manual observation of leaf morphology, the PTR-MS detection and data processing workflow can complete the analysis of a large number of samples in a shorter time, significantly improving detection efficiency. Regarding accuracy, VOCs, as an early physiological response indicator of plant diseases, can detect disease information before obvious visible lesions appear on the leaves. Combined with the high sensitivity of PTR-MS, early and accurate diagnosis of wheat diseases can be achieved. Furthermore, converting the data into an RGB format two-dimensional pixel matrix allows for compatibility with existing mature image analysis models, eliminating the need to redevelop complex algorithms, lowering the technical threshold, and improving the stability and reliability of model training and disease identification. This facilitates the rapid deployment of wheat disease identification systems in agricultural production, enabling timely prevention and control measures and reducing economic losses caused by diseases.

[0040] S103. The wheat disease identification device extracts the parent ion and the corresponding fragment ion of the pixel from the two-dimensional pixel matrix, and rearranges the position of the extracted pixel in the two-dimensional pixel matrix to obtain a recombined feature map.

[0041] Optionally, the wheat disease identification device extracts the pixels containing the parent ion and its corresponding fragment ions from the two-dimensional pixel matrix, and rearranges the positions of the extracted pixels in the two-dimensional pixel matrix to obtain a reconstructed feature map, including: when prior knowledge of the target disease is determined, extracting the pixels of the corresponding mass-to-charge ratio channel for feature reconstruction based on the correlation between the parent ion and its fragment ions in the prior knowledge of the target disease; when prior knowledge of the target disease is not determined, performing feature reconstruction on the pixels containing the parent ion and its corresponding fragment ions based on the generalized feature combination of volatile organic compounds released from wheat leaf samples.

[0042] Specifically, when using PTR-MS to detect VOCs in wheat leaves for disease identification, the chemical ionization of PTR-MS causes long-chain VOC molecules to break down into fragment ions. These fragment ions overlap with the parent ion signal, making it difficult to identify individual VOCs and affecting the accuracy of disease identification. Furthermore, in practical applications, there may be a lack of prior knowledge about the relevant VOCs for certain target diseases, which also poses a challenge to accurate identification.

[0043] However, on the one hand, it is known that when compounds of different chemical structures undergo structural fragmentation during PTR-MS ionization, the released fragment ions often exhibit a regular distribution, and there is a correlation between the parent ion of the compound and its main fragment ions. This provides a basis for feature recombination based on prior knowledge. On the other hand, common VOCs released by wheat have a certain degree of universality, such as aldehydes, alcohols, and terpenes. Based on this, a generalized feature combination strategy can be constructed to address situations where prior knowledge of VOCs related to target diseases is lacking.

[0044] The specific processing procedure is as follows: First, determine whether there is prior knowledge about VOCs relevant to the target disease. For example... Figure 2 As shown, if such features exist, the corresponding mass-to-charge ratio channel signals are extracted from the initially constructed two-dimensional pixel matrix for feature recombination based on the correlation between the parent ion and its main fragment ions. For example, given the known powdery mildew-specific VOCs, their positions in the image are rearranged according to preset rules based on their parent ions and corresponding main fragment ions. If prior knowledge of relevant VOCs for the target disease is lacking, a generalized feature combination strategy is constructed based on common wheat VOC categories (such as olefins, cycloalkanes, alkylbenzenes, alcohols, aldehydes, terpenes, etc.).

[0045] When prior knowledge of the target disease exists, the relative molecular mass of a compound plus one ([M+H]+) represents the mass-to-charge ratio characteristic of its parent ion in PTR-MS detection. Since PTR-MS uses chemical ionization, the target compound is prone to structural fragmentation during ionization. Fragment ions released during fragmentation of compounds with different chemical structures often exhibit a regular distribution. Table 1 summarizes the typical fragment ion distributions of six common VOCs, which can serve as a basis for characteristic recombination.

[0046] To address the lack of prior knowledge about relevant VOCs in the target diseases, a generalized feature combination strategy is used for recombination. This strategy targets common VOCs released by wheat, including alkylbenzenes (corresponding parent ions 105, 121, 129, 135), aldehydes (corresponding parent ions 107, 143, etc.), alkenes (corresponding parent ions 109, 111, etc.), alcohols (corresponding parent ions 117, 127, etc.), cycloalkanes (corresponding parent ion 119), and monoterpenes (corresponding parent ion 137). Typical fragment ions corresponding to each category are shown in Table 1, which allows for the construction of a structurally sound and generalizable feature combination scheme.

[0047] Table 1

[0048]

[0049] Optionally, the aforementioned prior knowledge of the target disease is derived from analytical literature using gas chromatography-mass spectrometry.

[0050] Based on the above scheme, firstly, by incorporating prior knowledge into feature recombination, the expression intensity of disease-related signals in the model is effectively enhanced, significantly reducing misjudgments caused by fragment interference. Secondly, it supports two combination methods: the specificity of known disease-related VOCs and the generality of unknown disease-related VOCs, greatly improving the model's adaptability. When the relevant VOCs released by the target disease are known, combining specific features based on prior knowledge can improve the specific recognition capability; when relevant knowledge is lacking, constructing a generalized combination scheme based on common VOC structure categories can improve the model's scenario adaptability. Furthermore, this processing achieves a structured representation of VOCs data, providing more interpretable input features for model training, which helps improve the model's generalization ability and the accuracy of disease identification, providing strong support for the early and accurate identification of wheat diseases.

[0051] S104. The wheat disease identification device inputs the recombined feature map into the wheat disease identification model for network training.

[0052] The wheat disease identification model is used to identify the type and severity of wheat diseases based on the mass-to-charge ratio signal features in the recombinant feature map.

[0053] Specifically, such as Figure 2As shown, the reconstructed feature map is input into the wheat disease identification model for training, establishing a mapping relationship between the mass-to-charge ratio signal features and the degree of wheat disease, and constructing an intelligent discrimination model for wheat health status. After the wheat disease identification model is established, the wheat leaves to be tested are placed in a gas collecting bottle to enrich the released VOCs. The mass-to-charge ratio signal intensity data is obtained using the PTR-MS system, and feature extraction, reconstruction, and image conversion are completed according to the established identification process. Finally, the trained wheat disease identification model outputs the disease type and infection degree corresponding to the sample, realizing rapid identification and intelligent classification of wheat diseases.

[0054] Optionally, the wheat disease identification model employs a lightweight convolutional neural network, which includes: an input layer, three convolutional layers, three max pooling layers, a flattening layer, two fully connected layers, and an output layer.

[0055] Specifically, the convolutional layers can use 3×3 kernels for local feature extraction with a stride of 1 and "same" padding to maintain the feature map size. To enhance the model's generalization ability and prevent overfitting, a Dropout layer is introduced after each max-pooling layer, randomly discarding 20% ​​of the neuron connections. This network structure design effectively extracts spatial correlation information from the two-dimensional mass-to-charge ratio feature map and improves the discrimination performance for samples of different disease levels.

[0056] In this embodiment, on the one hand, by extracting the maximum eigenvalues ​​of each mass-to-charge ratio channel and mapping them to a two-dimensional pixel matrix, background noise and signal fluctuation interference are effectively suppressed, highlighting key features related to diseases and improving the sensitivity of the wheat disease identification model to weak disease signals. On the other hand, by extracting and rearranging the pixels of the parent ion and its fragment ions, the dispersed chemical signals are integrated into a structured feature map, significantly enhancing the expressive power of disease features. Thus, high-precision early identification of wheat diseases can be achieved, providing key technical support for rapid disease warning and early intervention in modern agriculture.

[0057] The foregoing mainly describes the solutions provided by the embodiments of this application from a methodological perspective. To achieve the above functions, it includes corresponding hardware structures and / or software modules for executing each function. Those skilled in the art should readily recognize that, in conjunction with the units and algorithm steps of the various examples described in the embodiments disclosed herein, the embodiments of this application can be implemented in hardware or a combination of hardware and computer software. Whether a function is executed in hardware or by computer software driving hardware depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0058] The wheat disease identification method provided in this application can be executed by a wheat disease identification device or a control module for wheat disease identification within that device. This application uses the example of a wheat disease identification device executing the wheat disease identification method to illustrate the wheat disease identification device provided in this application.

[0059] It should be noted that the embodiments of this application can divide the wheat disease identification device into functional modules according to the above method examples. For example, each function can be divided into its own functional modules, or two or more functions can be integrated into one processing module. The integrated modules can be implemented in hardware or as software functional modules. Optionally, the module division in the embodiments of this application is illustrative and only represents one logical functional division; other division methods may be used in actual implementation.

[0060] like Figure 3 As shown in the figure, this application embodiment provides a wheat disease identification device 300. The wheat disease identification device 300 includes an acquisition module 301 and a processing module 302. The acquisition module 301 is used to acquire a training sample set, which includes wheat leaf samples in different health states, each wheat leaf sample labeled with disease type and disease severity. The processing module 302 is used to determine the mass-to-charge ratio signal intensity data of volatile organic compounds in the wheat leaf samples in the training sample set, and to determine the maximum eigenvalue of each mass-to-charge ratio channel in the mass-to-charge ratio signal intensity data. The maximum eigenvalues ​​of all mass-to-charge ratio channels are mapped to pixels to obtain a two-dimensional pixel matrix of wheat leaf samples, with one wheat leaf sample corresponding to one two-dimensional pixel matrix. The parent ion and its corresponding fragment ion are extracted from the two-dimensional pixel matrix, and the positions of the extracted pixels in the two-dimensional pixel matrix are rearranged to obtain a reconstructed feature map. The reconstructed feature map is input into a wheat disease identification model for network training. The wheat disease identification model is used to identify the disease type and disease severity of wheat based on the mass-to-charge ratio signal features in the reconstructed feature map.

[0061] Optionally, the processing module 302 is used to extract the pixels of the corresponding mass-to-charge ratio channel for feature recombination based on the correlation between the parent ion and its fragment ions in the prior knowledge of the target disease when the prior knowledge of the target disease is determined; and to reconstruct the features of the pixels where the parent ion and its corresponding fragment ions are located based on the generalized feature combination of volatile organic compounds released from wheat leaf samples when the prior knowledge of the target disease is not determined.

[0062] Optionally, the prior knowledge of the target disease is derived from analytical literature using gas chromatography-mass spectrometry.

[0063] Optionally, the wheat disease identification model employs a lightweight convolutional neural network, which includes: an input layer, three convolutional layers, three max pooling layers, a flattening layer, two fully connected layers, and an output layer.

[0064] Optionally, the processing module 302 is used to detect the mass-to-charge ratio signal intensity data of volatile organic compounds in wheat leaf samples using proton transfer reaction mass spectrometry.

[0065] In this embodiment, on the one hand, by extracting the maximum eigenvalues ​​of each mass-to-charge ratio channel and mapping them to a two-dimensional pixel matrix, background noise and signal fluctuation interference are effectively suppressed, highlighting key features related to diseases and improving the sensitivity of the wheat disease identification model to weak disease signals. On the other hand, by extracting and rearranging the pixels of the parent ion and its fragment ions, the dispersed chemical signals are integrated into a structured feature map, significantly enhancing the expressive power of disease features. Thus, high-precision early identification of wheat diseases can be achieved, providing key technical support for rapid disease warning and early intervention in modern agriculture.

[0066] Figure 4 An example is a schematic diagram of the physical structure of an electronic device, such as... Figure 4 As shown, the electronic device may include: a processor 410, a communications interface 420, a memory 430, and a communications bus 440, wherein the processor 410, the communications interface 420, and the memory 430 communicate with each other through the communications bus 440. The processor 410 can call logic instructions in the memory 430 to execute a wheat disease identification method. This method includes: acquiring a training sample set, which includes wheat leaf samples in different health states, each labeled with a disease type and degree of infection; determining the mass-to-charge ratio signal intensity data of volatile organic compounds in the wheat leaf samples within the training sample set, and determining the maximum eigenvalue of each mass-to-charge ratio channel in the mass-to-charge ratio signal intensity data; mapping the maximum eigenvalues ​​of all mass-to-charge ratio channels to pixels to obtain a two-dimensional pixel matrix of wheat leaf samples, with one wheat leaf sample corresponding to one two-dimensional pixel matrix; extracting the parent ion and its corresponding fragment ions from the two-dimensional pixel matrix, and rearranging the positions of the extracted pixels in the two-dimensional pixel matrix to obtain a reconstructed feature map; and inputting the reconstructed feature map into a wheat disease identification model for network training, the wheat disease identification model being used to identify the disease type and degree of infection of wheat based on the mass-to-charge ratio signal features in the reconstructed feature map.

[0067] Furthermore, the logical instructions in the aforementioned memory 830 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0068] On the other hand, the present invention also provides a computer program product, which includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the wheat disease identification method provided by the above methods. The method includes: acquiring a training sample set, which includes wheat leaf samples in different health states, each wheat leaf sample being labeled with a disease type and degree of infection; determining the mass-to-charge ratio signal intensity data of volatile organic compounds in the wheat leaf samples in the training sample set, and determining the maximum eigenvalue of each mass-to-charge ratio channel in the mass-to-charge ratio signal intensity data, mapping the maximum eigenvalues ​​of all mass-to-charge ratio channels to pixels to obtain a two-dimensional pixel matrix of wheat leaf samples, with one wheat leaf sample corresponding to one two-dimensional pixel matrix; extracting the pixels containing the parent ion and its corresponding fragment ions from the two-dimensional pixel matrix, and rearranging the positions of the extracted pixels in the two-dimensional pixel matrix to obtain a reconstructed feature map; inputting the reconstructed feature map into a wheat disease identification model for network training, the wheat disease identification model being used to identify the disease type and degree of infection of wheat based on the mass-to-charge ratio signal features in the reconstructed feature map.

[0069] In another aspect, the present invention also provides a non-transitory computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the wheat disease identification method provided by the above methods. The method includes: acquiring a training sample set, the training sample set including wheat leaf samples in different health states, each wheat leaf sample labeled with disease type and degree of infection; determining the mass-to-charge ratio signal intensity data of volatile organic compounds in the wheat leaf samples in the training sample set, and determining the maximum eigenvalue of each mass-to-charge ratio channel in the mass-to-charge ratio signal intensity data, mapping the maximum eigenvalues ​​of all mass-to-charge ratio channels to pixels, obtaining a two-dimensional pixel matrix of wheat leaf samples, one wheat leaf sample corresponding to one two-dimensional pixel matrix; extracting the pixels containing the parent ion and its corresponding fragment ions from the two-dimensional pixel matrix, and rearranging the positions of the extracted pixels in the two-dimensional pixel matrix to obtain a reconstructed feature map; inputting the reconstructed feature map into a wheat disease identification model for network training, the wheat disease identification model being used to identify the disease type and degree of infection of wheat based on the mass-to-charge ratio signal features in the reconstructed feature map.

[0070] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0071] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0072] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for identifying wheat diseases, characterized in that, include: Obtain a training sample set, which includes wheat leaf samples in different health states, with each wheat leaf sample labeled with the disease type and degree of infection; The mass-to-charge ratio signal intensity data of volatile organic compounds in wheat leaf samples in the training sample set are determined, and the maximum eigenvalue of each mass-to-charge ratio channel in the mass-to-charge ratio signal intensity data is determined. The maximum eigenvalues ​​of all mass-to-charge ratio channels are mapped to pixels to obtain a two-dimensional pixel matrix of wheat leaf samples. One wheat leaf sample corresponds to one two-dimensional pixel matrix. The parent ion and its corresponding fragment ion are extracted from the two-dimensional pixel matrix, and the positions of the extracted pixels in the two-dimensional pixel matrix are rearranged to obtain a reconstructed feature map. The recombined feature map is input into the wheat disease identification model for network training. The wheat disease identification model is used to identify the type and degree of wheat disease based on the mass-to-charge ratio signal features in the recombined feature map. The step of extracting the parent ion and its corresponding fragment ion pixels from the two-dimensional pixel matrix, and rearranging the positions of the extracted pixels in the two-dimensional pixel matrix to obtain a reconstructed feature map includes: Given prior knowledge of the target disease, based on the correlation between the parent ion and its fragment ions in the prior knowledge of the target disease, the pixels of the corresponding mass-to-charge ratio channel are extracted for feature recombination. Without prior knowledge of the target disease, feature recombination is performed on the pixels containing the parent ion and its corresponding fragment ions based on the generalized combination of volatile organic compounds released from wheat leaf samples.

2. The wheat disease identification method according to claim 1, characterized in that, The prior knowledge of the target disease is derived from analytical literature using gas chromatography-mass spectrometry.

3. The wheat disease identification method according to claim 1, characterized in that, The wheat disease identification model employs a lightweight convolutional neural network, which includes: an input layer, three convolutional layers, three max pooling layers, a flattening layer, two fully connected layers, and an output layer.

4. The wheat disease identification method according to claim 1, characterized in that, The determination of the mass-to-charge ratio signal intensity data of volatile organic compounds in wheat leaf samples from the training sample set includes: The mass-to-charge ratio signal intensity data of volatile organic compounds in wheat leaf samples were detected by proton transfer reaction mass spectrometry.

5. A wheat disease identification device, characterized in that, include: Acquisition module and processing module; The acquisition module is used to acquire a training sample set, which includes wheat leaf samples in different health states, and each wheat leaf sample is labeled with the disease type and degree of infection. The processing module is used to determine the mass-to-charge ratio signal intensity data of volatile organic compounds in wheat leaf samples in the training sample set, and to determine the maximum eigenvalue of each mass-to-charge ratio channel in the mass-to-charge ratio signal intensity data. The maximum eigenvalues ​​of all mass-to-charge ratio channels are mapped to pixels to obtain a two-dimensional pixel matrix of wheat leaf samples, with one wheat leaf sample corresponding to one two-dimensional pixel matrix. The module extracts the parent ion and its corresponding fragment ion pixels from the two-dimensional pixel matrix, and rearranges the positions of the extracted pixels in the two-dimensional pixel matrix to obtain a reconstructed feature map. The reconstructed feature map is input into a wheat disease identification model for network training. The wheat disease identification model is used to identify the disease type and severity of wheat based on the mass-to-charge ratio signal features in the reconstructed feature map. The processing module is used to extract the pixels of the corresponding mass-to-charge ratio channel for feature recombination based on the correlation between the parent ion and its fragment ions in the prior knowledge of the target disease, after determining the prior knowledge of the target disease. Without prior knowledge of the target disease, feature recombination is performed on the pixels containing the parent ion and its corresponding fragment ions based on the generalized combination of volatile organic compounds released from wheat leaf samples.

6. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the wheat disease identification method as described in any one of claims 1 to 4.

7. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the wheat disease identification method as described in any one of claims 1 to 4.

8. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the wheat disease identification method as described in any one of claims 1 to 4.

Images