Mulberry sclerotiniose detecting and grading system based on near infrared spectrum and gene analysis
By combining near-infrared spectroscopy with gene analysis, early and rapid screening and accurate grading of mulberry sclerotinia disease have been achieved, solving the problems of low efficiency and high cost in existing technologies and providing precise decision support for disease management.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-12-16
- Publication Date
- 2026-03-24
AI Technical Summary
Existing technologies are insufficient for quickly and accurately distinguishing and classifying mulberry sclerotinia rot, especially large-sized and small-sized sclerotinia rot, resulting in low efficiency in field identification and time-consuming and costly laboratory testing, which cannot meet the needs of early warning and large-scale screening.
Combining the near-infrared spectroscopy non-destructive testing module and the gene analysis verification module, images of mulberry fruits are acquired using a near-infrared camera for preliminary identification, and real-time quantitative PCR is used for precise confirmation. The data fusion and classification decision center then outputs comprehensive disease classification results.
It enables early and rapid screening and accurate grading of mulberry sclerotinia disease, improves field monitoring efficiency, reduces costs, and provides precise decision support for disease management.
Smart Images
Figure CN121720972A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of agricultural information technology, specifically to a mulberry sclerotinia disease detection and grading system based on near-infrared spectroscopy and gene analysis. Background Technology
[0002] Sclerotinia sclerotinia is a devastating disease that severely impacts mulberry yield and quality. Currently, three types are recognized: large-fruited sclerotinia sclerotinia, small-fruited sclerotinia sclerotinia, and small-fruited sclerotinia sclerotinia, respectively. The pathogens are *Ciboria shiraiana*, *Ciboria carunculoides*, and *Scleromitrula shiraiana*, respectively. Large-fruited and small-fruited sclerotinia ... Therefore, accurately, quickly, and efficiently distinguishing between large-sized and small-sized sclerotinia stem rot in mulberry is crucial for disease control. Currently, field identification mainly relies on manual observation based on experience, which is inefficient and cannot provide early warning. While laboratories can use methods such as microbial isolation and culture or molecular biological detection for accurate identification, the former is time-consuming, complex, and requires specialized laboratory conditions and personnel, failing to meet the needs of rapid diagnosis and timely control in the field. The latter usually requires in vitro and destructive treatment of samples, which is cumbersome, has low throughput, and is costly, making it difficult to conduct rapid and economical early screening and warning of large numbers of field samples.
[0003] Therefore, there is an urgent need in this field to develop a comprehensive technical solution that integrates early, rapid, large-scale screening, accurate confirmation, and tiered decision-making to overcome the limitations of a single technical approach and achieve early, accurate, and efficient monitoring and control of mulberry sclerotinia disease. Summary of the Invention
[0004] To address the problems mentioned in the background section, this invention provides the following technical solution: a mulberry sclerotinia stem rot detection and grading system based on near-infrared spectroscopy and gene analysis, comprising:
[0005] The near-infrared non-destructive testing module is used to acquire near-infrared spectral images of mulberry fruits and perform rapid and non-destructive preliminary disease identification of the fruits based on the built-in primary discrimination model, and output the primary discrimination results.
[0006] The gene analysis verification module is used to perform minimally invasive sampling on fruit samples identified by the near-infrared non-destructive detection module, and to perform real-time fluorescence quantitative PCR detection using specific primers for the pathogen of mulberry sclerotinia stem rot, so as to accurately identify the pathogen type and quantitatively calculate the pathogen load.
[0007] The data fusion and classification decision center is communicatively connected to the near-infrared non-destructive testing module and the gene analysis verification module. It is used to receive the primary discrimination results and the pathogen type and load data, run the advanced fusion classification model based on decision tree, and output the comprehensive disease classification results.
[0008] Preferably, the near-infrared non-destructive testing module includes a near-infrared camera with a working wavelength of 800-1700nm, a pixel size of 5µm*5µm, and a signal-to-noise ratio greater than 50dB. It is used to acquire near-infrared images of mulberry fruits under indoor lighting conditions, with an exposure time of 100,000µs. The primary discrimination model analyzes the acquired images through image processing algorithms, calculates the ratio of the abnormal area of the fruit to the total area of the fruit as the disease severity factor α, and classifies the fruit as "healthy fruit", "early suspected", or "disease-infected fruit" according to a preset α threshold range.
[0009] Preferably, the image processing algorithm includes grayscale normalization, energy spectrum analysis and filtering of the near-infrared image (preferably with a filtering value of 0.14), and extraction of fruit area and abnormal area through target recognition algorithm; wherein, the calculation formula of the disease severity factor α is: α = s1 / s, where s1 is the area of abnormal area and s is the total area of fruit.
[0010] Preferably, the gene analysis verification module is used to accurately identify the pathogen species and quantitatively analyze the nucleic acid load of the sampled sample using specific molecular detection technology targeting the mulberry sclerotium rot pathogen.
[0011] Preferably, the comprehensive disease classification results output by the data fusion and classification decision center include at least the following categories:
[0012] Category I: Healthy Fruit;
[0013] Category II: Early suspected cases;
[0014] Category III: Early stage of sclerotinia stem rot;
[0015] Category IV: Mid-stage sclerotinia disease;
[0016] Category V: Late stage of sclerotinia stem rot;
[0017] The determination of categories III to V combines the primary discrimination results with the positive results of gene verification and pathogen load information.
[0018] Preferably, the system is used for early differentiation and severity classification of mulberry sclerotinia stem rot and mulberry small-grain sclerotinia stem rot.
[0019] Preferably, the preset α threshold range is as follows: α < 1% is identified as "healthy fruit"; 1% ≤ α ≤ 3% is identified as "early suspected"; α > 3% is identified as "disease-infected fruit"; wherein, the samples identified as "disease-infected fruit" are further distinguished into different degrees of severity according to the range of α values.
[0020] Preferably, for fruits identified as "early suspected", the processing method is further determined based on the number of fruits identified as "early suspected" within a set range around them: if the number is less than a preset threshold, they are marked and short-term retested; if the number reaches or exceeds the preset threshold, they are directly sampled and sent to the gene analysis verification module.
[0021] Preferably, the molecular detection technology used in the gene analysis verification module has the sensitivity to detect trace amounts of pathogenic bacterial nucleic acid.
[0022] An early detection method for mulberry sclerotinia stem rot based on near-infrared spectroscopy and gene analysis, using the aforementioned detection system, the method includes the following steps:
[0023] The mulberry fruit was imaged and preliminarily identified using a near-infrared non-destructive testing module.
[0024] Minimally invasive sampling was performed on fruit samples marked as requiring verification in the initial identification, and the pathogens were accurately identified and quantified through the gene analysis verification module.
[0025] By integrating preliminary judgment results and gene verification results through the data fusion and classification decision center, the final comprehensive disease classification information is output.
[0026] Compared with existing technologies, this invention provides a detection and grading system for mulberry sclerotinia stem rot based on near-infrared spectroscopy and gene analysis, which has the following beneficial effects:
[0027] 1. This mulberry sclerotinia stem rot detection and grading system, based on near-infrared spectroscopy and gene analysis, creatively integrates near-infrared spectroscopy non-destructive detection technology with targeted molecular verification technology in a hierarchical manner. The near-infrared module can non-destructively and rapidly detect the internal biochemical and physical changes in the fruit caused by the disease in its early stages, enabling preliminary screening and early risk warning for a wide range of samples; the molecular verification module, on the other hand, performs precise confirmation and pathogen typing only for high-risk samples. This solution, through the working paradigm of "rapid non-destructive initial screening driven by targeted and precise verification," fundamentally solves the industry problem of the inability of traditional single methods to effectively connect early warning sensitivity, large-scale screening efficiency, and laboratory-level confirmation accuracy.
[0028] 2. This mulberry sclerotinia stem rot detection and grading system, based on near-infrared spectroscopy and gene analysis, integrates phenotypic severity indicators (such as the percentage of abnormal fruit area α) reflected by near-infrared spectroscopy with quantitative pathogen load information (such as Ct values) provided by gene analysis at the decision center. Through a decision tree-based classification model, it outputs refined classification results such as "healthy, early suspected, early-stage, mid-stage, and late-stage disease." This provides direct quantitative evidence for implementing differentiated and precise field management (such as variable-rate pesticide application), avoiding the indiscriminate overuse of pesticides.
[0029] 3. This mulberry sclerotinia rot detection and grading system, based on near-infrared spectroscopy and gene analysis, employs a workflow of "rapid and non-destructive initial screening + targeted and precise verification." Near-infrared initial screening can be automated, greatly covering the screening scope; subsequent gene verification is initiated only for a small number of suspected or infected samples. This strategy avoids expensive and time-consuming molecular testing on all samples, significantly improving the overall efficiency of large-scale monitoring while ensuring accuracy and reducing labor and material costs.
[0030] 4. This mulberry sclerotinia rot detection and grading system, based on near-infrared spectroscopy and gene analysis, deeply integrates crop phenomics (near-infrared spectroscopy) information with pathogen molecular detection information through a data fusion and classification decision center, forming an intelligent decision-making system. This system completes a full technical loop from early suspected detection to precise pathogen identification, and then to severity assessment and classification, demonstrating strong system innovation and practical value. Attached Figure Description
[0031] Figure 1 This is a comparison of the imaging effects of mulberry sclerotium rot in the visible light band (left) and near-infrared band (right) according to the present invention;
[0032] Figure 2 This is a near-infrared image processing and recognition diagram of mulberry fruit according to the present invention;
[0033] Figure 3 This is a schematic diagram illustrating the sensitivity verification of a molecular detection method used in the system of this invention;
[0034] Figure 4 This invention provides a schematic diagram of a standard curve for the quantitative analysis of pathogenic nucleic acid in the present invention.
[0035] Figure 5 This is a schematic diagram of the overall system architecture and data fusion classification decision-making process of the present invention. Detailed Implementation
[0036] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0037] Example
[0038] This embodiment provides a detection and grading system for mulberry sclerotinia stem rot based on near-infrared spectroscopy and gene analysis (hereinafter referred to as the "system"). The system is particularly suitable for early identification, type differentiation and severity grading of two diseases, mulberry large-sized sclerotinia stem rot and mulberry small-sized sclerotinia stem rot, which have similar early symptoms but different pathogens.
[0039] like Figures 1 to 5 As shown, this system mainly includes three functional modules: near-infrared non-destructive testing module, gene analysis and verification module, and data fusion and classification decision center.
[0040] 1. Implementation of the near-infrared non-destructive testing module
[0041] The core task of this module is to achieve rapid and non-destructive initial screening of mulberry fruits. The specific implementation steps are as follows:
[0042] Hardware Configuration and Image Acquisition: A 1.3-megapixel short-wave infrared camera with a GigE interface and a corresponding lens was used. Its effective operating wavelength is 800-1700 nm, pixel size is 5µm*5µm, and signal-to-noise ratio is greater than 50dB. The camera was fixed on an indoor, well-lit detection platform, ensuring that the mulberry fruit (which could be placed on a sample tray) was within its field of view. The camera exposure time was set to 100,000 microseconds, and images were captured under normal indoor daylight conditions to obtain raw near-infrared images of the mulberry fruit. Compared to visible light imaging (see...), this method... Figure 1 (Left) Near-infrared imaging can more clearly reveal early changes in biochemical components inside the fruit caused by disease (see left). Figure 1 (Right), thus revealing the early lesion area.
[0043] Image Processing and Feature Extraction: The acquired raw near-infrared image is input into the built-in primary discrimination model for processing. First, the image undergoes grayscale normalization to eliminate the influence of uneven illumination. Then, the image's energy spectrum is analyzed, and a normalized filter (preferably 0.14) is applied to reduce noise and enhance the image. The processing effect is as follows: Figure 2As shown. Next, target recognition algorithms (such as threshold-based segmentation or edge detection algorithms) are used to segment the complete fruit region from the processed image, and further identify abnormal regions of the fruit (i.e., suspected lesion regions). Let the area of the abnormal region be s1, and the total area of the fruit be s.
[0044] Primary identification and grading: Calculate the disease severity factor α = s1 / s. Based on a preset threshold range, quickly identify the fruit:
[0045] If α < 1%, it is judged as "healthy fruit".
[0046] If 1% ≤ α ≤ 3%, it is judged as "early suspected" and marked as a key focus.
[0047] If α > 3%, the fruit is considered "infected". The severity can be further subdivided based on the α value. For example, 3% < α ≤ 10% is "early stage of sclerotinia disease", 10% < α ≤ 20% is "mid stage of sclerotinia disease", and α > 20% is "late stage of sclerotinia disease".
[0048] Post-processing strategy:
[0049] For samples determined to be "healthy fruit", no further action is usually required; simply record the results.
[0050] For samples identified as "early suspected," the system will employ different strategies based on the assessment of surrounding fruits: It will count the number of other fruits identified as "early suspected" within a certain radius (e.g., within a 10 cm radius) around the suspected fruit. If the number is less than 10, only that area will be marked, and a retest is recommended after 2-3 days. The final judgment will be made based on the trend of the α value (a significant increase indicates disease infection, while no change indicates healthy). If the number reaches or exceeds 10, it is considered a high risk of localized disease, and the operator will be immediately notified to harvest the suspected fruits and send them to the gene analysis verification module.
[0051] For samples identified as "infected fruit", the system directly prompts for harvesting and sends them to the gene analysis verification module for precise identification of the pathogen.
[0052] 2. Implementation of the gene analysis validation module
[0053] This module, as the system's precise confirmation unit, is responsible for performing qualitative and quantitative pathogen analysis on high-risk samples (i.e., samples identified as "early suspected" or "infected fruit") screened by the near-infrared module.
[0054] Sample processing and nucleic acid extraction: The labeled fruit was sampled minimally (e.g., a small amount of pulp tissue) and total nucleic acids were extracted from the samples. Extraction methods can employ conventional plant genomic DNA extraction techniques, such as commercially available extraction kits.
[0055] Targeted molecular detection: Detection is performed using nucleic acid amplification techniques capable of specifically identifying the main pathogens of mulberry sclerotinia rot (such as *Ciboria shiraiana* and *Ciboria carunculoides*). As a preferred embodiment, real-time quantitative PCR technology can be used, employing specific primers targeting the aforementioned pathogens for amplification. Those skilled in the art can design and screen suitable primer pairs based on the known genomic sequences of the pathogens.
[0056] Results Interpretation: By analyzing signal changes during the amplification process (e.g., monitoring Ct values and amplification curves in real-time quantitative PCR), the presence or absence of specific pathogens can be determined. Furthermore, by comparing with standards, relative or absolute quantification of the pathogen's nucleic acid load can be achieved, providing molecular-level quantitative indicators of disease severity. This module is designed to ensure high specificity and sufficient detection sensitivity for initially screened positive samples, enabling closed-loop precision validation.
[0057] 3. Implementation of the Data Fusion and Classification Decision Center
[0058] The center is a software algorithm module that runs on a local server or cloud computing platform. It is responsible for integrating information from the first two modules and making the final decision.
[0059] Data Reception and Storage: The center continuously receives and stores preliminary discrimination results (healthy / early suspected / disease-infected) and key feature data (such as α value, image feature vector) from the near-infrared module. Simultaneously, it receives and stores verification results (pathogen type: negative, *Acer mulberryii* positive, *Acer sarcodactylum* positive) and quantitative data (Ct value, pathogen load) from the gene analysis module.
[0060] Fusion classification model operation: The center runs an advanced fusion classification model trained based on a decision tree algorithm. This model is trained using a large number of samples with "near-infrared features + gene verification ground truth" data. The model uses the α value (or richer spectral features) of the current sample and the gene verification results (pathogen type, load) as input features.
[0061] Example of decision-making logic:
[0062] If the near-infrared spectroscopy indicates "health," genetic verification is usually not initiated, and the final classification is Category I (healthy fruit).
[0063] If near-infrared spectroscopy determines it to be "early suspected", but genetic verification is not performed or is not required (e.g., few surrounding cases, no change in α after retesting), it is ultimately classified as Category II (early suspected) and becomes a monitoring target.
[0064] If the near-infrared spectroscopy indicates "early suspected" or "early stage of infection", and the gene verification is positive (for any pathogen), and the pathogen load is low, then the final classification is Category III (early stage of sclerotinia disease - confirmed).
[0065] If near-infrared spectroscopy indicates "mid-stage infection" and gene verification is positive with a moderate pathogen load, the final classification is Category IV (mid-stage sclerotinia disease - confirmed).
[0066] If near-infrared spectroscopy indicates "late stage of infection" and gene verification is positive with a high pathogen load, then the final classification is Category V (late stage of sclerotinia disease - confirmed).
[0067] The classification logic diagram can be found here. Figure 5 .
[0068] Output and Decision Support: The system visualizes the final comprehensive classification results, pathogen types, and load information to the user (e.g., via a display screen or mobile terminal), and can generate disease distribution maps and severity statistical reports. Based on different classification levels and pathogen types, the system can provide differentiated control recommendations, such as: focusing on monitoring and preventative application of pesticides for "Category III" areas; and recommending the use of targeted fungicides and precise variable-rate spraying for "Category IV / V" areas caused by specific pathogens.
[0069] Figure 1 Center: Visible light (left) and near-infrared band (right) imaging effects of mulberry sclerotium rot.
[0070] Figure 2 In the middle: A. Healthy fruit; B. Suspected early stage of mulberry sclerotinia disease; C. Mid-stage of sclerotinia disease.
[0071] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.
Claims
1. A detection and grading system for mulberry sclerotinia stem rot based on near-infrared spectroscopy and gene analysis, characterized in that, include: The near-infrared non-destructive testing module is used to acquire near-infrared spectral images of mulberry fruits and perform rapid and non-destructive preliminary disease identification of the fruits based on the built-in primary discrimination model, and output the primary discrimination results. The gene analysis verification module is used to perform minimally invasive sampling on fruit samples identified by the near-infrared non-destructive detection module, and to perform real-time fluorescence quantitative PCR detection using specific primers for the pathogen of mulberry sclerotinia stem rot, so as to accurately identify the pathogen type and quantitatively calculate the pathogen load. The data fusion and classification decision center is communicatively connected to the near-infrared non-destructive testing module and the gene analysis verification module. It is used to receive the primary discrimination results and the pathogen type and load data, run the advanced fusion classification model based on decision tree, and output the comprehensive disease classification results.
2. The mulberry sclerotinia stem rot detection and grading system based on near-infrared spectroscopy and gene analysis according to claim 1, characterized in that, The near-infrared non-destructive testing module includes a near-infrared camera with a working wavelength of 800-1700nm, a pixel size of 5µm*5µm, and a signal-to-noise ratio greater than 50dB. It is used to acquire near-infrared images of mulberry fruits under indoor lighting conditions, with an exposure time of 100000µs. The primary discrimination model analyzes the collected images using image processing algorithms, calculates the ratio of the area of abnormal fruit regions to the total fruit area as the disease severity factor α, and classifies the fruit as "healthy fruit", "early suspected", or "disease-infected fruit" based on a preset α threshold range.
3. The mulberry sclerotinia rot detection and grading system based on near-infrared spectroscopy and gene analysis according to claim 2, characterized in that, The image processing algorithm includes grayscale normalization, energy spectrum analysis and filtering of the near-infrared image (preferred filtering value is 0.14), and extraction of fruit area and abnormal area through target recognition algorithm; wherein, the calculation formula of the disease severity factor α is: α = s1 / s, where s1 is the area of abnormal area and s is the total area of fruit.
4. The mulberry sclerotinia stem rot detection and grading system based on near-infrared spectroscopy and gene analysis according to claim 3, characterized in that, The gene analysis verification module is used to accurately identify the pathogen species and quantitatively analyze the nucleic acid load of the sampled samples using specific molecular detection technology targeting the mulberry sclerotium rot pathogen.
5. The mulberry sclerotinia stem rot detection and grading system based on near-infrared spectroscopy and gene analysis according to claim 4, characterized in that, The comprehensive disease classification results output by the data fusion and classification decision center include at least the following categories: Category I: Healthy Fruit; Category II: Early suspected cases; Category III: Early stage of sclerotinia stem rot; Category IV: Mid-stage sclerotinia disease; Category V: Late stage of sclerotinia stem rot; The determination of categories III to V combines the primary discrimination results with the positive results of gene verification and pathogen load information.
6. The mulberry sclerotinia stem rot detection and grading system based on near-infrared spectroscopy and gene analysis according to claim 5, characterized in that, The system is used for early differentiation and severity classification of mulberry sclerotinia stem rot and mulberry small-grain sclerotinia stem rot.
7. The mulberry sclerotinia stem rot detection and grading system based on near-infrared spectroscopy and gene analysis according to claim 6, characterized in that, The preset α threshold range is as follows: α < 1% is identified as "healthy fruit"; 1% ≤ α ≤ 3% is identified as "early suspected"; α > 3% is identified as "infected fruit"; among which, samples identified as "infected fruit" are further distinguished into different degrees of severity according to the range of α values.
8. The mulberry sclerotinia stem rot detection and grading system based on near-infrared spectroscopy and gene analysis according to claim 7, characterized in that, For fruits identified as "early suspected", the processing method is further determined based on the number of fruits identified as "early suspected" within a set range around them: if the number is less than a preset threshold, they are marked and short-term retested; if the number reaches or exceeds the preset threshold, they are directly sampled and sent to the gene analysis verification module.
9. The mulberry sclerotinia stem rot detection and grading system based on near-infrared spectroscopy and gene analysis according to claim 8, characterized in that, The gene analysis verification module uses molecular detection technology with the sensitivity to detect trace amounts of pathogenic bacterial nucleic acid.
10. A method for detecting sclerotinia stem rot in mulberry based on near-infrared spectroscopy and gene analysis, characterized in that, The method, using the detection system as described in any one of claims 1-9, comprises the steps of: The mulberry fruit was imaged and preliminarily identified using a near-infrared non-destructive testing module. Minimally invasive sampling was performed on fruit samples marked as requiring verification in the initial identification, and the pathogens were accurately identified and quantified through the gene analysis verification module. By integrating preliminary judgment results and gene verification results through the data fusion and classification decision center, the final comprehensive disease classification information is output, thereby providing a basis for the formulation of disease prevention and control strategies.