An automatic detection method and system for rice diseases

Abstract

The application discloses a kind of automatic detection method and system of rice disease, it is related to crop detection field, including, preparation biodegradable optical marker, biodegradable optical marker is sprayed to rice leaf using high-pressure microfluidic jet device, mark rice leaf is formed;The vascular tissue in marked rice leaf is scanned, and GFP fluorescence signal and CQD fluorescence signal are captured, generate three-dimensional fluorescence intensity diagram and using ImageJ image analysis software is handled, obtain the disease positioning fluorescence diagram of rice leaf;According to the disease positioning fluorescence diagram of rice leaf, the reflectance spectrum of marked rice leaf is scanned, obtain the absorption characteristic of reflectance spectrum and compare with pre-constructed disease fingerprint library, obtain rice disease diagnosis data.The application is prepared by double-labeled biodegradable optical marker, reduces the interference between fluorescence signal, also improves the accuracy of three-dimensional reconstruction of rice disease area.

Description

Technical Field

[0001] This invention relates to the field of crop detection, and in particular to an automatic detection method and system for rice diseases. Background Technology

[0002] In recent years, automated detection and management technologies for rice diseases have developed rapidly in the agricultural field. With the widespread application of near-infrared spectroscopy, fluorescence imaging, and biomarkers, various disease detection methods based on spectral analysis and image processing have been developed. For example, traditional near-infrared spectrometers scan the reflectance spectrum of rice leaves and, combined with chemometric models, identify the absorption characteristics of common diseases such as rice blast. Fluorescent labeling technology captures the fluorescence signals of GFP or CQD to accurately locate diseased areas.

[0003] Existing automated detection methods for rice diseases have unresolved issues. While conventional spectral analysis methods can identify disease types, their ability to accurately locate diseased areas is limited, increasing operational complexity in paddy fields. Furthermore, fluorescence imaging technology relies on a single labeler (such as GFP), which limits signal resolution due to tissue penetration depth and signal crosstalk, resulting in insufficient accuracy in reconstructing three-dimensional disease distribution. Summary of the Invention

[0004] In view of the aforementioned existing problems, the present invention is proposed.

[0005] Therefore, the present invention provides an automatic detection method for rice diseases, which solves the problems of low detection and location accuracy and insufficient comprehensive diagnosis in existing automatic detection methods for rice diseases.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0007] In a first aspect, the present invention provides an automatic detection method for rice diseases, comprising: preparing a biodegradable optical marker and spraying the biodegradable optical marker onto rice leaves using a high-pressure microfluidic spraying device to form marked rice leaves;

[0008] The vascular tissue of labeled rice leaves was scanned to capture GFP and CQD fluorescence signals, and a three-dimensional fluorescence intensity map was generated. The image was then processed using ImageJ image analysis software to obtain a fluorescence map of disease location in rice leaves.

[0009] Based on the fluorescence map of disease location in rice leaves, the reflectance spectrum of labeled rice leaves is scanned to obtain the absorption characteristics of the reflectance spectrum and compared with the pre-constructed disease fingerprint database to obtain rice disease diagnostic data.

[0010] The system uses a smart sprayer to automatically dispense pesticides, spray them onto the disease-infected areas in the disease location fluorescence map of rice leaves, and tracks the degradation process of the biodegradable optical markers to generate a rice disease management report.

[0011] As a preferred embodiment of the automatic detection method for rice diseases described in this invention, the preparation of the biodegradable optical marker includes inserting GFP into Bacillus subtilis using electroporation, screening out positive transformants, culturing the screened positive transformants, and obtaining a Bacillus subtilis suspension.

[0012] CQD was synthesized by hydrothermal carbonization and then encapsulated in chitosan to form a chitosan-CQD hydrogel microsphere solution.

[0013] A biodegradable optical marker was obtained by mixing Bacillus subtilis suspension and chitosan-CQD hydrogel microsphere solution.

[0014] As a preferred embodiment of the automatic detection method for rice diseases described in this invention, the method for forming marked rice leaves includes configuring a high-pressure microfluidic spraying device, setting spraying parameters, and spraying a biodegradable optical marker onto the stomatal region on the back of the rice leaf.

[0015] As a preferred embodiment of the automatic detection method for rice diseases described in this invention, the method for capturing GFP fluorescence signals and CQD fluorescence signals includes selecting Thorlabs Bergamo II as a two-photon microscope and registering it, and focusing the laser beam generated by the registered two-photon microscope onto the marked rice leaves.

[0016] As a preferred embodiment of the automatic detection method for rice diseases described in this invention, the method for obtaining the disease location fluorescence image of rice leaves includes using the optical path unit of a registered two-photon microscope to separate the GFP fluorescence signal and the CQD fluorescence signal, and recording the fluorescence intensity of the separated GFP fluorescence signal and CQD fluorescence signal respectively to obtain a rice leaf fluorescence image dataset.

[0017] The rice leaf fluorescence image dataset was pseudo-color encoded to obtain an encoded fluorescence image dataset. The Bio-Formats plugin was then used to stack the encoded fluorescence image dataset in multiple channels to form a three-dimensional dataset.

[0018] By setting a fluorescence intensity threshold, the fluorescence intensity of GFP and CQD fluorescence signals in the three-dimensional dataset is statistically analyzed using ImageJ software and compared with the fluorescence intensity threshold to determine the diseased areas of rice.

[0019] By statistically analyzing the volume and centroid coordinates of voxels in the diseased areas of rice, a fluorescence map of the disease location on rice leaves was obtained.

[0020] As a preferred embodiment of the automatic detection method for rice diseases described in this invention, the method for obtaining rice disease diagnostic data includes configuring an infrared handheld detection device, using the infrared handheld detection device to scan the rice disease area in the disease location fluorescence map of rice leaves, capturing the reflected light intensity of the marked rice leaves, and generating reflected light intensity data points.

[0021] The reflectance of the reflected light intensity data points is calculated and saved as a TXT file to obtain the reflectance spectrum dataset of rice leaves;

[0022] Savitzky-Golay filtering was applied to the reflectance spectrum dataset of rice leaves using an ARM Cortex-M4 microcontroller to obtain a smoothed reflectance spectrum dataset.

[0023] Set an absorption indentation threshold and compare it with the intensity of reflected light in the smoothed reflectance spectrum dataset to obtain the absorption characteristics of the reflectance spectrum. Then, normalize the data to generate an absorption spectrum feature dataset.

[0024] Rice leaves with diseased samples were collected to construct a disease fingerprint database. By calculating the cosine similarity between the reflected light intensity in the disease fingerprint database and the normalized reflected light intensity in the absorption spectral feature dataset, the diagnosis results of rice diseases were obtained and the severity of rice diseases was assessed.

[0025] The rice disease diagnosis results, the severity of rice diseases, and the centroid coordinates are integrated into a CSV file to obtain rice disease diagnosis data.

[0026] As a preferred embodiment of the automatic detection method for rice diseases described in this invention, the output of the rice disease management report includes using a hierarchical clustering algorithm to group the centroid coordinates in the rice disease diagnostic data and generate a minimum coverage rectangle.

[0027] An initial spraying path is generated within the minimum coverage rectangle using an S-shaped path algorithm, and the initial spraying path is optimized using RTK-GPS navigation to generate the final spraying path.

[0028] The pesticide was sprayed into the disease-infected area using a smart sprayer along the final spraying path to obtain pesticide application data, and the degradation process of the biodegradable optical marker was monitored using a soil sensor network to obtain CQD fluorescence data.

[0029] The CQD fluorescence data and pesticide application data are integrated to generate a rice disease management report.

[0030] Secondly, the present invention provides an automatic detection system for rice diseases, comprising a preparation module for preparing a biodegradable optical marker and spraying the biodegradable optical marker onto rice leaves using a high-pressure microfluidic spraying device to form marked rice leaves.

[0031] The scanning module scans the vascular tissue of labeled rice leaves, captures GFP and CQD fluorescence signals, generates a three-dimensional fluorescence intensity map, and processes it using ImageJ image analysis software to obtain a fluorescence map of disease location on rice leaves.

[0032] The comparison module scans the reflectance spectrum of labeled rice leaves based on the fluorescence map of disease location on rice leaves, obtains the absorption characteristics of the reflectance spectrum, and compares it with the pre-constructed disease fingerprint database to obtain rice disease diagnostic data.

[0033] The report generation module uses a smart sprayer to automatically dispense pesticides, spray them onto the disease-infected areas in the disease location fluorescence map of rice leaves, and tracks the degradation process of the biodegradable optical markers to output a rice disease management report.

[0034] Thirdly, the present invention provides a computer device including a memory and a processor, wherein the memory stores a computer program, wherein when the computer program is executed by the processor, it implements any step of the automatic detection method for rice diseases as described in the first aspect of the present invention.

[0035] Fourthly, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, it implements any step of the automatic detection method for rice diseases as described in the first aspect of the present invention.

[0036] The beneficial effects of this invention are as follows: This invention prepares a GFP and CQD dual-labeled biodegradable optical marker by mixing Bacillus subtilis suspension and chitosan-CQD hydrogel microsphere solution. Through signal separation and degradation monitoring, it solves the problems of low localization accuracy and insufficient diagnostic comprehensiveness. Furthermore, the synergistic effect of the dual markers reduces interference between fluorescence signals, improves the accuracy of three-dimensional reconstruction of rice disease areas, and ensures uniform adhesion of the markers on rice leaves, which is superior to the low signal resolution of single markers. Attached Figure Description

[0037] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0038] Figure 1 This is a flowchart of an automatic detection method for rice diseases.

[0039] Figure 2 This is a schematic diagram of the preparation of biodegradable optical labeling agents.

[0040] Figure 3 A schematic diagram illustrating the generation of fluorescence intensity maps for locating diseases on rice leaves.

[0041] Figure 4 This is a flowchart for reflectance spectral analysis and disease diagnosis. Detailed Implementation

[0042] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0043] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0044] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0045] Reference Figures 1-4 This is one embodiment of the present invention, which provides an automatic detection method for rice diseases, including the following steps:

[0046] S1. Preparation of biodegradable optical markers, and spraying the biodegradable optical markers onto rice leaves using a high-pressure microfluidic spraying device to form marked rice leaves, includes the following steps.

[0047] S1.1 The green fluorescent protein (GFP) gene was inserted into Bacillus subtilis using electroporation. The procedure was as follows: First, an expression plasmid containing the GFP gene was constructed. The GFP expression plasmid was then transformed into the WB800N strain of Bacillus subtilis using an electroporator. After electroporation, the strain was cultured in chloramphenicol-containing LB medium under specific environmental conditions, such as 37°C and 200 rpm shaking, to screen for positive transformants.

[0048] The selected positive transformants were cultured again. After culture, the cells of the selected positive transformants were collected by centrifugation and resuspended in sterile phosphate buffer at a concentration of, for example, 10⁻⁶.9 CFU / mL was used to obtain a Bacillus subtilis suspension.

[0049] S1.2. Carbon quantum dots (CQDs) are synthesized by hydrothermal carbonization and encapsulated in chitosan to form a chitosan-CQD hydrogel microsphere solution.

[0050] Specifically, citric acid and ethylenediamine are dissolved in deionized water, stirred evenly, and then transferred to a stainless steel reactor lined with polytetrafluoroethylene for chemical reaction. After the reaction is completed, the mixture is cooled to room temperature and filtered through a filter membrane to obtain a carbon quantum dot solution.

[0051] Mix a portion of carbon quantum dot solution and chitosan solution, add a small amount of glutaraldehyde (e.g., 0.5 mL) as a crosslinking agent, and stir to form a chitosan-CQD hydrogel microsphere solution.

[0052] S1.3. Mix Bacillus subtilis suspension and chitosan-CQD hydrogel microsphere solution to obtain biodegradable optical marker.

[0053] The specific operation is as follows: Take a portion of Bacillus subtilis suspension and a portion of chitosan-CQD hydrogel microsphere solution and mix them in a 1:1 volume ratio. At the same time, add Tween-80 as a dispersant. After stirring for 30-40 minutes, add carboxymethyl cellulose (CMC) as an adhesive. Stir again for 10-15 minutes to form the final biodegradable optical labeling agent.

[0054] S1.4. Use a high-pressure microfluidic spraying device to spray biodegradable optical markers onto the stomatal area on the back of rice leaves to form marked rice leaves.

[0055] Specifically, a customized high-pressure microfluidic spraying device is selected, equipped with a precision flow controller and GPS positioning system. The biodegradable optical marker is loaded into the storage tank of the high-pressure microfluidic spraying device, and the spraying parameters, including spraying pressure and spraying flow rate, are set. The high-pressure microfluidic spraying device moves along the field at a constant speed to spray, with the spraying focus on the back of the rice leaves.

[0056] S2. Scanning the vascular tissue of labeled rice leaves to capture GFP and CQD fluorescence signals, generating a three-dimensional fluorescence intensity map, and processing it using ImageJ image analysis software to obtain a disease localization fluorescence map of rice leaves includes the following steps.

[0057] S2.1 Select the Thorlabs Bergamo II equipped with a Ti:sapphire laser as the two-photon microscope and register the two-photon microscope. The specific registration process is as follows: Set the parameters of the Ti:sapphire laser, including wavelength, pulse width, laser frequency and repetition frequency. The laser frequency is controlled at about 20mW by an adjustable attenuator to avoid photodamage to rice leaf tissue.

[0058] The two-photon microscope is equipped with a water immersion objective and focal length, covering the vascular tissue of rice leaves, including the xylem and phloem. A Hamamatsu H7422P-40 photomultiplier tube with a bandpass filter is used to mount the two-photon microscope on a portable track device. The track device's running speed and scanning area are set, and the track is laid along the rice paddy with a servo motor-controlled movement speed.

[0059] S2.2 Using a registered two-photon microscope, GFP and CQD in the biodegradable optical labeling agent are scanned to obtain GFP fluorescence signal and CQD fluorescence signal.

[0060] The specific operation is as follows: A laser beam generated by a registered two-photon microscope is focused onto the xylem and phloem of a rice leaf through a water immersion objective. Due to the photon absorption characteristics of GFP and CQD, GFP and CQD fluorescence signals are generated under the influence of the laser beam. These signals are then separated by the optical path unit of the two-photon microscope. The separated GFP and CQD fluorescence signals are guided to a photomultiplier tube (PMT) via a bandpass filter. The PMT records the fluorescence intensity of the separated GFP and CQD signals in real time at a sampling rate of approximately 1 MHz, within a range of 10. 3 ~10 6 RFU (Reactive Fusion Method). Each scan generates a 1000×1000 pixel two-dimensional image, with pixel values ​​corresponding to fluorescence intensity, stored in grayscale format. GFP and CQD fluorescence signal images are recorded as independent channels, separated by bandpass filters to avoid signal crosstalk. During scanning, the two-photon microscope displays the fluorescence intensity distribution in real time. High-intensity areas, corresponding to diseased parts, are observed on the screen. A single rice leaf generates 75MB of fluorescence image data, equivalent to 75 layers × 1000 × 1000 pixels, forming a rice leaf fluorescence image dataset.

[0061] The rice leaf fluorescence image dataset was pseudo-color encoded to obtain the encoded fluorescence image dataset.

[0062] Using two-photon microscopy, the grayscale images of the GFP fluorescence signal and the CQD fluorescence signal were assigned pseudo-color images, for example, the GFP channel was assigned a green tone and the CQD channel was assigned a blue tone.

[0063] The grayscale images of GFP and CQD fluorescence signals are mapped to the RGB color space using a lookup table (LUT).

[0064] The specific steps are as follows: ThorImage software is used to read the grayscale images of the GFP and CQD channels. Green and blue LUTs are then set. The green LUT maps the grayscale image of the GFP channel to green brightness, and the blue LUT maps the grayscale image of the CQD channel to blue brightness, forming a single-channel color image, i.e., a green image and a blue image. Subsequently, the MergeChannels function of ThorImage software is used to overlay the green and blue images, forming a composite pseudo-color image, which is then saved in TIFF format to obtain the encoded fluorescence image dataset.

[0065] The purpose of pseudo-color coding is to visually distinguish the spatial distribution of GFP fluorescence signals and CQD fluorescence signals, highlighting the fluorescence characteristics of diseased parts, such as signal overlap in bacterial blight areas.

[0066] S2.3. Use ImageJ software to stack the encoded fluorescence image dataset into a three-dimensional dataset.

[0067] Specifically, launch ImageJ software, load the Bio-Formats plugin, select the encoded fluorescence image dataset, and the Bio-Formats plugin reads all images from the encoded fluorescence image dataset and sorts them according to the scanning order. In the Bio-Formats interface, check the "Group files with similar names" and "Open ashyperstack" options. ImageJ software will automatically stack all images into a multi-channel stack, which is displayed as "Hyperstack" in the ImageJ window. To verify the correctness of the stacking, check the three-layer cross-section of the multi-channel stack using ImageJ software's OrthogonalViews function to ensure that the alignment error between layers is less than 1 μm. The three-dimensional dataset consists of voxels, each containing grayscale values ​​for the GFP and CQD channels.

[0068] The 3DViewer plugin in ImageJ software is used to render a 3D dataset.

[0069] The specific steps are as follows: In ImageJ software, open the 3D dataset, launch the 3DViewer plugin via Plugins > 3DViewer. After launch, the 3DViewer plugin interface will display a 3D rendering window. Select the Volume rendering mode in the 3D rendering window, set the rendering parameters, that is, assign green to the GFP channel and blue to the CQD channel, and set the transparency to about 50%. Use the 3DViewer plugin to map each voxel in the 3D dataset to the color and brightness corresponding to the fluorescence intensity, and generate a preliminary 3D fluorescence intensity map.

[0070] Specifically, the grayscale value of each voxel is read one by one, and the voxel is projected along the viewing direction, i.e. the z-axis, using a stereo projection algorithm. Color, brightness and transparency are superimposed one by one according to the set rendering parameters. The mapping process is displayed in real time in the ImageJ 3DViewer window. The mapped voxels form a preliminary three-dimensional fluorescence intensity map and are stored in OBJ format.

[0071] S2.4. Use ImageJ software to mark the preliminary three-dimensional fluorescence intensity map, find the diseased areas of rice leaves, and output the disease location fluorescence map of rice leaves.

[0072] The specific steps are as follows: In ImageJ software, load the 3D fluorescence intensity map and set the fluorescence intensity threshold, which is determined by the maximum fluorescence intensity of the GFP and CQD fluorescence signals in the GFP and CQD channels, i.e., the maximum fluorescence intensity. Use the Measure function in ImageJ software to statistically analyze the fluorescence intensity of the GFP and CQD fluorescence signals in the 3D dataset. For example, the maximum intensity in the GFP channel is 60,000 grayscale values, and in the CQD channel, it is 55,000 grayscale values. When the fluorescence intensity of the GFP and CQD fluorescence signals exceeds the fluorescence intensity threshold, the corresponding area of ​​the rice leaf is marked as a diseased area, such as xylem blockage in bacterial leaf blight or infection sites in rice blast. This marking is achieved using the Analyze plugin in ImageJ software. The Analyze plugin automatically identifies fluorescence intensities above the fluorescence intensity threshold and generates the outline and intensity level of the diseased area; for example, red indicates a fluorescence intensity >90%. After marking, the 3DObjectCounter function is used to count the volume and centroid coordinates of voxels in the diseased area and store them as a CSV file, including the diseased area ID, X coordinate, Y coordinate, Z depth, and voxel volume, forming a fluorescence map of disease location on rice leaves.

[0073] S3. Based on the disease localization fluorescence map of rice leaves, scan the reflectance spectrum of the labeled rice leaves, obtain the absorption characteristics of the reflectance spectrum, and compare it with the pre-constructed disease fingerprint database to obtain rice disease diagnostic data. This includes the following steps.

[0074] S3.1 Configuring an infrared handheld detection device with a diode laser and an InGaAs photodetector.

[0075] Specifically, the infrared handheld detection device is connected to the lithium battery and powered on for a few minutes to warm up and stabilize the diode laser output. After warm-up, the diode laser parameters need to be calibrated, namely the laser wavelength, output power, and output spot diameter, and then focused using the built-in collimating lens. The InGaAs photodetector parameters also need to be calibrated, including the detection band, sampling rate, and cross-group gain.

[0076] The infrared handheld detection device is equipped with a touchscreen that displays spectral curves and an interface in real time. The touchscreen loads a fluorescence image of disease location on rice leaves, and the bodily centroid coordinates of this image are imported into the device's GPS navigation to guide the scanning of targeted diseased areas. A reflectivity standard plate is used as a calibration benchmark for the reflected light intensity, and the reference reflected light intensity is measured. A reflectivity standard plate is a high-reflectivity diffuse reflective material; due to its uniform reflection of near-infrared light, it is used as a calibration reference for spectrometers or detectors to provide known reflectivity characteristics.

[0077] S3.2 Based on the fluorescence map of disease location on rice leaves, use the configured infrared handheld detection device to scan the reflectance spectrum of the rice leaf surface and output the absorption characteristics of the reflectance spectrum.

[0078] The reflectance spectrum of rice leaves refers to the continuous distribution curve of the intensity of reflected light produced by incident laser light on the surface of rice leaves in the near-infrared band as a function of wavelength. The reflectance spectrum reflects the optical properties of the waxy layer on the surface of rice leaves. For healthy rice leaves, the reflectance spectrum will appear smooth because the waxy layer is intact and has weak absorption. For diseased rice leaves, the reflectance spectrum will show absorption depressions because pathogens will destroy the cell structure and waxy layer of the rice leaves.

[0079] The specific scanning process is as follows: The configured infrared handheld detection device moves along the veins (main vein or first-order lateral vein) of the rice leaf at a constant speed to scan the diseased areas in the disease location fluorescence map of the rice leaf one by one. The rice leaf is irradiated by a laser beam generated by a diode laser, and the InGaAs photodetector captures the intensity of reflected light from the rice leaf to generate multiple reflected light intensity data points.

[0080] The reflectance of the reflected light intensity data points is calculated using the following expression:

[0081] ;

[0082] in, The reflectivity represents the intensity of reflected light. This indicates the intensity of reflected light from rice leaves. This represents the reference reflected light intensity of the reflectivity standard plate. For example, suppose the measured reflected light intensity of a rice leaf is 8.5 × 10⁻⁶. 5 μW / cm², while the reference reflected light intensity of the reflectivity standard plate is 10 μW / cm². 6 The reflectance of the reflected light intensity data point is calculated using this formula, and the reflectance is 85%. The reflectance of healthy rice leaves is usually greater than 85%, while the reflectance of diseased rice leaves is usually less than 65%.

[0083] The reflectance of all reflected light intensity data points is saved in TXT format to form a reflectance spectrum dataset of rice leaves, including wavelength points, reflectance intensity, and reflectance.

[0084] S3.3 Identify and extract the absorption characteristics of the reflectance spectra of rice leaf reflectance spectra dataset.

[0085] The absorption characteristics of reflectance spectrum refer to the decrease in reflectance and the concave pattern in the spectral curve caused by rice diseases in a specific spectral band, such as 1050~1100nm. Specific absorption characteristics include rice blast, sheath blight, and bacterial blight. Rice blast presents a double-valley curve with two absorption peaks, sheath blight presents a single broad peak, and bacterial blight presents a single narrow peak. These three absorption characteristics reflect the tissue changes caused by rice diseases. For example, sheath blight causes necrosis of rice tissue, forming a broad peak. Each absorption characteristic includes wavelength, peak width, and absorption depth.

[0086] The reflectance spectrum dataset of rice leaves was imported into the ARM Cortex-M4 microcontroller, and Savitzky-Golay filtering was applied to the dataset using SpectralWorks software to remove noise.

[0087] Specifically, filtering parameters are set, namely the window width and the fitting polynomial. For each wavelength point, SpectralWorks software selects multiple wavelength points before and after it to construct a sliding window. The sliding window moves along the spectral band, and the fitting polynomial is repeatedly fitted to each wavelength point to generate a smooth spectrum, which can preserve the shape of the absorption dip in the spectral curve. For wavelength points at the edge, such as 1052.5nm, an asymmetric window is used, that is, multiple wavelength points before or after are used for processing, and a smooth reflectance spectrum dataset is output.

[0088] Absorption features in the smoothed reflectance spectrum dataset were identified using SpectralWorks software. An absorption depression threshold was set based on the reflected light intensity and signal-to-noise ratio obtained from field experiments. When the reflected light intensity in the smoothed reflectance spectrum dataset was less than the absorption depression threshold, the reflected light intensity was marked as an absorption feature. Simultaneously, SpectralWorks software recorded the wavelength position, wavelength depth, and wavelength width of the absorption feature.

[0089] The absorption features are normalized to obtain an absorption spectral feature dataset.

[0090] Specifically, the highest and lowest reflected light intensities in the spectral bands are found using SpectralWorks software. The difference between the highest and lowest reflected light intensities is calculated. The lowest reflected light intensity is subtracted from the reflected light intensity of each absorption feature and divided by the difference between the highest and lowest reflected light intensities to obtain the normalized reflected light intensity, which is then stored in JSON format to form an absorption spectral feature dataset.

[0091] S3.4. Constructing a Disease Fingerprint Database: This database stores the absorption characteristics of common rice diseases. The specific construction process is as follows: 1000 sets of disease-affected leaf samples were collected from rice-growing areas, covering different rice varieties, such as indica and japonica rice, and the infection stages of the rice diseases. The disease-affected leaf samples were scanned using an OceanOptics NIRQuest near-infrared spectrometer, and the reflectance spectra of the disease samples were recorded. Each set of samples was repeated multiple times, and the average value was taken. Savitzky-Golay filtering was applied to the reflectance spectra of the disease samples to identify absorption characteristics. These absorption characteristics were then compiled into a JSON file, containing the disease type, reflected light intensity, spectral curve, peak position, and peak width, forming the disease fingerprint database.

[0092] S3.5. Compare the absorption spectral feature dataset with the disease fingerprint database, run the cosine similarity algorithm using an ARM Cortex-M4 microcontroller, calculate the cosine similarity, and generate rice disease diagnostic data.

[0093] The normalized reflected light intensity in the absorption spectral feature dataset is compared one by one with the reflected light intensity in the disease fingerprint database, and the cosine similarity can be calculated by the following expression:

[0094] ;

[0095] in, This represents the cosine similarity score, which ranges from 0 to 1, with 1 indicating a perfect match. This represents the normalized reflected light intensity in the absorption spectral feature dataset. Indicates the intensity of reflected light in the disease fingerprint database;

[0096] The similarity threshold is set based on a combination of field variability and ROC curves. Field variability refers to the slight changes in the spectral curve of rice leaves caused by the combined effects of variety, growth stage and environment. For example, setting the similarity threshold to 0.85 can tolerate ±5% variation and ensure field applicability.

[0097] The highest cosine similarity is selected using SpectralWorks software. When the highest cosine similarity is greater than a similarity threshold, the rice disease diagnosis result is obtained, which corresponds to the disease type in the disease fingerprint database with the highest cosine similarity output by the cosine similarity algorithm. For example, if the cosine similarity between the normalized reflected light intensity of the absorption spectral feature dataset and the reflected light intensity of the disease fingerprint database is 0.95, and higher than other types, such as sheath blight (0.6), then the diagnosis result is rice blast.

[0098] The severity of rice diseases is assessed based on the reflectance of the reflectance spectrum data of rice leaves, and is classified into severe infection, moderate infection, and mild infection.

[0099] Specifically, based on previously collected disease samples, handheld detection devices were used to measure reflectance, and the necrotic area on rice leaf tissue was observed under a microscope to obtain measurement results. For example, if a certain disease causes severe damage to the waxy layer of rice leaves and cell wall disintegration, the average reflectance is between 60% ± 5%, and the corresponding necrotic area is greater than 50%. In this case, the disease can be classified as severe infection. The same logic applies to moderate and mild infection. For example, in the early stage of bacterial blight, if the necrotic area of ​​leaf tissue is less than 20% and the reflectance is higher than 75%, it indicates that only the waxy layer of rice leaves is slightly damaged, and the reflectance is close to that of healthy leaves. Bacterial blight can be classified as mild infection.

[0100] The rice disease diagnosis results, the severity of rice diseases, and the centroid coordinates of the disease location fluorescence map on rice leaves are integrated into a CSV file to form rice disease diagnosis data, which is then transmitted to the smart sprayer via LoRa wireless communication.

[0101] S4. Using a smart sprayer, the pesticide is automatically dispensed and sprayed onto the disease-infected areas in the disease location fluorescence map of rice leaves. The degradation process of the biodegradable optical marker is tracked, and a rice disease management report is generated, including the following steps.

[0102] S4.1 Calculate the required pesticide concentration for the smart sprayer. Specifically, the smart sprayer's controller receives rice disease diagnostic data via a LoRa receiver. After receiving the data, it sets a pesticide formulation table based on the diagnostic results, the severity of the rice disease, and the standards of the Ministry of Agriculture and Rural Affairs. For example, for severe rice blast, use 0.3 g / L tricyclazole; for moderate rice sheath blight, use 0.5 g / L hexaconazole; and for mild bacterial blight, use 0.2 g / L thiabendazole. The mass of pesticide to be mixed is calculated based on the mixing parameters. For example, assuming the required spray volume is 2 liters and 0.3 g / L tricyclazole is needed for rice blast, the mass of pesticide to be mixed is 0.3 × 2 = 0.6 g.

[0103] The concentration of the drug is calculated based on the calculated mass of the drug to be dispensed, using the following expression:

[0104] ;

[0105] in, Indicates the concentration of the medicine. Indicates the quality of the dispensed medication. Indicates the volume of water;

[0106] Hierarchical clustering algorithm was used to group the centroid coordinates in rice disease diagnosis data to generate minimum coverage rectangles. Specifically:

[0107] The Euclidean distance between each pair of centroid coordinates is calculated using the following expression:

[0108] ;

[0109] in, This represents the Euclidean distance between each pair of centroid coordinates. Represents the x-axis number of... Coordinates of each disease point Represents the x-axis number of... Coordinates of each disease point Represents the first on the y-axis Coordinates of each disease point Represents the first on the y-axis Coordinates of the diseased points.

[0110] When the Euclidean distance between each pair of centroid coordinates is less than a fixed value, such as 2cm, the pair of centroid coordinates are clustered into the same group.

[0111] The S-shaped path algorithm is used to generate a spraying path within the minimum coverage rectangle to cover the disease-infected area.

[0112] Specifically, set the spray path parameters, including spray width and step size. Calculate the length of the spray path by dividing the area of ​​the minimum coverage rectangle by the spray width, and use this length as the initial spray path.

[0113] The initial spraying path is optimized to generate the final spraying path.

[0114] Specifically, RTK-GPS navigation is used to correct the position of the smart sprayer, ensuring that the starting point of the initial spraying path is aligned with the lower left corner of the minimum coverage rectangle. The smart sprayer's controller identifies healthy areas of rice leaves by comparing reflectance in rice disease diagnosis data and removes non-disease-affected areas (reflectance > 75%). For example, if the reflectance is 80% in the 12-12.5cm segment within the minimum coverage rectangle, the initial spraying path skips this segment. Cubic spline interpolation is applied to each turnaround point (approximately every 5cm) to generate a smooth trajectory, reducing jitter in the initial spraying path.

[0115] S4.2. Use a smart sprayer to spray pesticides on the disease-infected area and obtain pesticide application data.

[0116] Specifically, the smart sprayer is placed in the field, and the final spraying path and pesticide status are confirmed via a touch screen. At the same time, the controller activates the nozzle and preheats it for about 20 to 30 seconds.

[0117] After preparation for spraying is completed, the intelligent sprayer moves along the final spraying path at a constant speed. Simultaneously, GPS navigation tracks the centroid coordinates, and the nozzles dynamically adjust the solenoid valves based on these coordinates, spraying only the diseased areas of the rice. During spraying, the spraying amount per acre is recorded, and the real-time spraying amount is controlled. The expression for the real-time spraying amount is:

[0118] ;

[0119] in, This indicates the real-time spray volume. Indicates the area of ​​the diseased region. This indicates the amount of sprayed per acre of land. It indicates the total area of ​​farmland.

[0120] After spraying is completed, the controller of the smart sprayer records the application data, including rice leaf ID, disease type, pesticide type, pesticide concentration, centroid coordinates, and application amount.

[0121] S4.3. Monitoring the biodegradation process of biodegradable optical markers through a soil sensor network to obtain CQD fluorescence data includes the following steps:

[0122] Since the chitosan-CQD hydrogel microsphere solution in the biodegradable optical labeling agent undergoes hydrolysis within its half-life due to pesticide irrigation or rainwater, releasing CQD, the remaining concentration of chitosan in the hydrogel microspheres can be calculated using the following expression:

[0123] ;

[0124] in, This indicates the remaining concentration of chitosan in the hydrogel microspheres. This indicates the initial chitosan concentration, measured from the previously prepared chitosan-CQD hydrogel microsphere solution. The multiple of the half-life is defined as the hydrolysis time of chitosan from the start of pesticide spraying to the present moment, divided by the half-life.

[0125] To further explain, this formula is used to estimate when chitosan degrades to a safe level, and the subsequent sensor sampling frequency is adjusted based on the remaining concentration, thereby reducing blind monitoring.

[0126] A soil sensor network was selected as the monitoring device and deployed in the field, covering the disease-infected areas of rice leaves. The nodes of the soil sensor network measured the CQD fluorescence intensity in runoff four times a day and converted it into CQD concentration. The conversion formula is as follows:

[0127] ;

[0128] in, Indicates the concentration of CQD. Indicates the fluorescence intensity of CQD in runoff water. The calibration slope, representing the relationship between the fluorescence intensity and concentration of CQD in runoff, reflects the linear relationship between the fluorescence intensity and concentration of CQD in runoff. It is obtained by measuring the fluorescence intensity using a YSIEXO2 fluorescence probe with a known concentration of CQD solution, and linearly fitting the fluorescence intensity using a laboratory standard curve (R² = 0.99). The intercept representing the linear relationship between the fluorescence intensity of CQD in runoff water and its concentration is determined by laboratory measurements of CQD-free pure water. It is 0.

[0129] When the CQD concentration drops to a stable value, such as below 0.01 μg / L, it indicates 100% degradation. The CQD concentration is then stored in CSV format to obtain CQD fluorescence data.

[0130] S4.4 Integrate CQD fluorescence data and pesticide application data, and upload them synchronously to the cloud database using a 4G network to generate a rice disease management report. The report specifically includes pesticide application records, namely leaf ID, disease type, pesticide type, pesticide concentration, application amount, degradation progress, and verification results. For example, 20 days after application, the CQD concentration is <0.01μg / L, indicating 100% clearance.

[0131] This embodiment also provides an automatic detection system for rice diseases, including: a preparation module for preparing a biodegradable optical marker and spraying the biodegradable optical marker onto rice leaves using a high-pressure microfluidic spraying device to form marked rice leaves;

[0132] The scanning module scans the vascular tissue of labeled rice leaves, captures GFP and CQD fluorescence signals, generates a three-dimensional fluorescence intensity map, and processes it using ImageJ image analysis software to obtain a fluorescence map of disease location on rice leaves.

[0133] The comparison module scans the reflectance spectrum of labeled rice leaves based on the fluorescence map of disease location on rice leaves, obtains the absorption characteristics of the reflectance spectrum, and compares it with the pre-constructed disease fingerprint database to obtain rice disease diagnostic data.

[0134] The report generation module uses a smart sprayer to automatically dispense pesticides, spray them onto the disease-infected areas in the disease location fluorescence map of rice leaves, and tracks the degradation process of the biodegradable optical markers to output a rice disease management report.

[0135] This embodiment also provides a computer device applicable to the automatic detection method for rice diseases, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to realize the automatic detection method for rice diseases as proposed in the above embodiment.

[0136] The computer device can be a terminal, comprising a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, carrier networks, NFC (Near Field Communication), or other technologies. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the computer device's casing, or an external keyboard, touchpad, or mouse.

[0137] This embodiment also provides a storage medium storing a computer program, which, when executed by a processor, implements the automatic detection method for rice diseases as proposed in the above embodiments. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Red-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0138] In summary, this invention prepares a biodegradable optical marker dual-labeled with GFP and CQD by mixing Bacillus subtilis suspension and chitosan-CQD hydrogel microsphere solution. Through signal separation and degradation monitoring, it addresses the problems of low localization accuracy and insufficient diagnostic comprehensiveness. Furthermore, the synergistic effect of the dual markers reduces interference between fluorescence signals, improves the accuracy of three-dimensional reconstruction of rice disease areas, and ensures uniform adhesion of the markers to rice leaves, thus overcoming the low signal resolution of single markers.

[0139] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for automatic detection of rice disease, characterized by: The application relates to a method for diagnosing rice diseases. The method comprises the following steps: preparing a biodegradable optical marker, spraying the biodegradable optical marker on rice leaves by using a high-pressure microfluidic jet device to form marked rice leaves, and scanning the leaf vascular tissue in the marked rice leaves to capture GFP fluorescent signals and CQD fluorescent signals, generate a three-dimensional fluorescence intensity map, and process the three-dimensional fluorescence intensity map by using ImageJ image analysis software to obtain a disease positioning fluorescence map of the rice leaves. The method for preparing the biodegradable optical marker comprises the following steps: inserting GFP into Bacillus subtilis by using an electroporation method, screening positive transformants, and culturing the screened positive transformants to obtain a Bacillus subtilis suspension. The method comprises the following steps: synthesizing CQD by using a hydrothermal carbonization method, encapsulating the CQD in chitosan to form a chitosan-CQD hydrogel microsphere solution, and mixing the Bacillus subtilis suspension and the chitosan-CQD hydrogel microsphere solution to obtain the biodegradable optical marker. The method comprises the following steps: mixing part of the Bacillus subtilis suspension and part of the chitosan-CQD hydrogel microsphere solution according to a volume ratio of 1:1, adding Tween-80 as a dispersing agent, stirring for 30-40 minutes, adding carboxymethyl cellulose (CMC) as an adhesive, stirring again for 10-15 minutes, and obtaining the final biodegradable optical marker. The method comprises the following steps: scanning the leaf vascular tissue in the marked rice leaves to capture GFP fluorescent signals and CQD fluorescent signals, generating a three-dimensional fluorescence intensity map, and processing the three-dimensional fluorescence intensity map by using ImageJ image analysis software to obtain a disease positioning fluorescence map of the rice leaves. The method comprises the following steps: scanning the reflection spectrum of the marked rice leaves according to the disease positioning fluorescence map of the rice leaves, obtaining the absorption characteristics of the reflection spectrum, comparing the absorption characteristics with a pre-constructed disease fingerprint library, and obtaining rice disease diagnosis data. The method comprises the following steps: using an intelligent sprayer to automatically dispense medicine, spraying the medicine to a disease infection area in the disease positioning fluorescence map of the rice leaves, tracking the degradation process of the biodegradable optical marker, and outputting a rice disease management report. The method for forming the marked rice leaves comprises the following steps: configuring a high-pressure microfluidic jet device, setting jet parameters, and spraying the biodegradable optical marker into the stomatal area on the back of the rice leaves.

2. The automatic detection method of the rice disease according to claim 1, characterized by: The method for capturing the GFP fluorescent signals and the CQD fluorescent signals comprises the following steps: selecting Thorlabs Bergamo II as a two-photon microscope, performing registration, and focusing a laser beam generated by the registered two-photon microscope on the marked rice leaves.

3. The automatic detection method of the rice disease according to claim 2, characterized by: The method for obtaining the disease positioning fluorescence map of the rice leaves comprises the following steps: separating the GFP fluorescent signals and the CQD fluorescent signals by using a light path unit of the registered two-photon microscope, recording the fluorescence intensities of the separated GFP fluorescent signals and CQD fluorescent signals, and obtaining a rice leaf fluorescence image data set.

4. The automatic detection method of rice disease according to claim 3, characterized in that: The method comprises the following steps: pseudo-color encoding the rice leaf fluorescence image data set to obtain an encoded fluorescence image data set, and using a Bio-Formats plug-in to perform multi-channel stacking on the encoded fluorescence image data set to form a three-dimensional data set. The method comprises the following steps: setting a fluorescence intensity threshold, using ImageJ software to count the fluorescence intensities of the GFP fluorescent signals and the CQD fluorescent signals in the three-dimensional data set, comparing the fluorescence intensities with the fluorescence intensity threshold, and determining a disease area of the rice. The method comprises the following steps: counting the volume and centroid coordinates of voxels in the disease area of the rice to obtain a disease positioning fluorescence map of the rice leaves. ​ 5. The automatic detection method of the rice disease according to claim 4, characterized by: The obtained rice disease diagnosis data comprises configuring an infrared handheld detection device, using the infrared handheld detection device to scan the rice disease area in the disease positioning fluorescence map of the rice leaf, capturing the reflected light intensity of the marked rice leaf, and generating a reflected light intensity data point; The reflectivity of the reflected light intensity data point is calculated to obtain a reflectance spectrum data set of the rice leaf; A Savitzky-Golay filter is applied to the reflectance spectrum data set of the rice leaf using an ARM Cortex-M4 microcontroller to obtain a smoothed reflectance spectrum data set; An absorption dip threshold is set and compared with the reflected light intensity in the smoothed reflectance spectrum data set to obtain absorption characteristics of the reflectance spectrum and perform normalization processing to generate an absorption spectrum feature data set; Rice leaves of disease samples are collected to construct a disease fingerprint library, and the cosine similarity between the reflected light intensity in the disease fingerprint library and the normalized reflected light intensity in the absorption spectrum feature data set is calculated to obtain rice disease diagnosis results and evaluate the severity of the rice disease; The rice disease diagnosis results, the severity of the rice disease, and the centroid coordinates are integrated into a CSV file to obtain rice disease diagnosis data.

6. The automatic detection method of rice disease according to claim 5, wherein: The output rice disease management report comprises configuring a hierarchical clustering algorithm to group the centroid coordinates in the rice disease diagnosis data to generate a minimum bounding rectangle; A preliminary spraying path is generated in the minimum bounding rectangle using an S-shaped path algorithm, and the preliminary spraying path is optimized using RTK-GPS navigation to generate a final spraying path; A smart sprayer is used to spray pesticides along the final spraying path to the disease infection area to obtain application data, and a soil sensor network is used to monitor the degradation process of the biodegradable optical marker to obtain CQD fluorescence data; The CQD fluorescence data and the application data are integrated to generate a rice disease management report.

7. An automatic detection system for rice diseases, based on the automatic detection method for rice diseases according to any one of claims 1 to 6, characterized in that: It comprises, A preparation module for preparing a biodegradable optical marker and spraying the biodegradable optical marker onto rice leaves using a high-pressure microfluidic jet device to mark the rice leaves; A scanning module for scanning the leaf vascular tissue in the marked rice leaves, capturing GFP fluorescence signals and CQD fluorescence signals, generating a three-dimensional fluorescence intensity map, and processing the map using ImageJ image analysis software to obtain a disease positioning fluorescence map of the rice leaf; A comparison module for scanning the reflectance spectrum of the marked rice leaf according to the disease positioning fluorescence map of the rice leaf, obtaining the absorption characteristics of the reflectance spectrum, and comparing them with a pre-constructed disease fingerprint library to obtain rice disease diagnosis data; A report generation module for automatically dispensing pesticides using a smart sprayer, spraying them into the disease infection area in the disease positioning fluorescence map of the rice leaf, and tracking the degradation process of the biodegradable optical marker to output a rice disease management report.

8. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that: The processor executes the computer program to implement the steps of the automatic detection method of rice disease according to any one of claims 1-6.

9. A computer readable storage medium having stored thereon a computer program, characterized in that: The computer program is executed by the processor to implement the steps of the automatic detection method of rice disease according to any one of claims 1-6.

Images