A mulberry disease and pest analysis method and system based on unmanned aerial vehicle monitoring

Abstract

The application discloses a mulberry disease and pest analysis method and system based on unmanned aerial vehicle monitoring, which comprises the following steps: an unmanned aerial vehicle collects multispectral and RGB high-definition images of multiple mulberry planting areas according to a preset flight route; image data is preprocessed to generate multi-dimensional spectral indexes, and image texture and color features are analyzed to evaluate diseases and pests, a planting area fusion feature vector is constructed, feature differences are calculated through the Euclidean distance, and associated areas with feature deviations meeting a preset range are marked; images of multiple periods of associated areas are collected, preset indexes and disease and pest indexes are calculated by weighting, disease and pest trends are evaluated, and disease source areas are screened; disease and pest control priorities are set according to disease and pest characteristics, and a disease and pest control scheme of the unmanned aerial vehicle is planned. The application realizes accurate monitoring, disease source positioning and scientific prevention and control of mulberry diseases and pests, improves monitoring efficiency and prevention and control pertinence, is suitable for multiple planting scenes, and effectively improves the intelligent management level of mulberry planting.

Description

Technical Field

[0001] This invention relates to the field of unmanned aerial vehicle (UAV) monitoring and agricultural planting analysis, and more specifically, to a method and system for analyzing mulberry pests and diseases based on UAV monitoring. Background Technology

[0002] Mulberries, a unique fruit with both edible and medicinal value, have seen continuous expansion in planting scale in recent years, especially in many southern provinces where large-scale planting bases and smart planting demonstration parks have been established. However, frequent outbreaks of pests and diseases such as mulberry borer, mulberry thrips, sclerotinia stem rot, and brown spot disease have severely affected the yield and quality of mulberries. According to statistics, the yield reduction caused by pests and diseases can reach 20%-50%, and some severely affected areas have even experienced total crop failure.

[0003] Traditional mulberry pest and disease monitoring mainly relies on manual inspections, which has many limitations: firstly, the monitoring efficiency is extremely low; secondly, the identification is highly subjective; and thirdly, the spatial coverage is insufficient. Furthermore, traditional mulberry planting monitoring techniques are difficult to locate the source of disease, deduce the transmission path, and achieve accurate trend prediction. They lack predictions of disease trends, pest and disease impact trends and ranges, and lack comprehensive analysis methods using multispectral imaging and high-definition images. As a result, they rely on manual experience for monitoring and pest and disease control, making them impractical. Summary of the Invention

[0004] This invention overcomes the shortcomings of existing technologies and proposes a method and system for analyzing mulberry diseases and pests based on UAV monitoring.

[0005] The first aspect of this invention provides a method for analyzing mulberry pests and diseases based on unmanned aerial vehicle (UAV) monitoring, comprising: S1: The drone will cruise and collect multispectral and RGB high-definition images of multiple mulberry planting areas according to the preset route, and obtain multispectral image data and high-definition image data. S2: Preprocess the image data, generate multi-dimensional spectral indices from multispectral image data, extract texture and color features and assess pests and diseases from high-definition image data to obtain multi-dimensional image pest and disease indices, and construct a fusion feature vector for each planting area based on the spectral indices and image pest and disease indices. S3: For each planting area, calculate the difference of the corresponding fused feature vector using Euclidean distance, analyze the feature deviation in the preset neighboring areas of the planting area, and mark the planting areas with feature deviations below the preset range as associated areas; S4: Using drones, collect image data of related areas over multiple periods, and perform weighted calculations on preset spectral indices and preset image pest and disease indices. Based on the calculation results over multiple periods, evaluate the trend of pest and disease characteristics and screen out the source areas of pests and diseases. S5: Assess the characteristics of pests and diseases in the source areas and planting areas, set regional prevention and control priorities, and plan pest and disease control solutions based on drones.

[0006] In this solution, S1 specifically refers to: Based on the target mulberry area, multiple mulberry planting areas are divided, and the mulberry planting density in each mulberry planting area is within a preset range; Based on the geographical information, planting density and topographic features of the mulberry planting area, a grid-based flight route planning method is adopted to set preset flight routes, and the spacing between the flight routes is adaptively adjusted according to the planting density. In each mulberry planting area, multispectral and high-resolution image data were acquired using drone data acquisition units.

[0007] In this solution, S2 specifically refers to: A radiometric correction algorithm based on the dark target method is used to extract pixels and perform radiometric correction on multispectral images. Based on the geographical coordinates of the mulberry planting area, a quadratic polynomial fitting algorithm is used to correct the geometric distortion of the image. A Gaussian filtering algorithm is used to denoise the multispectral image. A histogram equalization algorithm is used to adjust the brightness and contrast of high-definition image data. The Sobel operator was used for background segmentation to separate the mulberry plant area from the background area, and finally, image enhancement preprocessing was performed.

[0008] In this solution, S2 further includes: Multiband calculations and reflectance analysis were performed on the preprocessed multispectral image data, and multispectral index inversion analysis was conducted to obtain multidimensional spectral indices. Multidimensional spectral indices include normalized vegetation index, normalized red edge index, and leaf chlorophyll content inversion value; The preprocessed high-definition image data is converted to grayscale, and the texture features of mulberry plants are extracted based on the LBP operator. Color histograms are introduced to statistically analyze the color features of the high-definition image data. The texture features and color features are compared with the preset features for similarity, and the texture feature score and color feature score are obtained based on the similarity analysis results. Image disease and pest index includes texture feature degree and color feature degree; A fused feature vector is generated based on multiple dimensions, including normalized vegetation index, normalized red edge index, leaf chlorophyll content inversion value, texture feature degree, and color feature degree. A fused feature vector for each planting area is constructed based on spectral indices and image pest and disease indices.

[0009] In this solution, S3 specifically refers to: Mark a current planting area and calculate the feature deviation between the current planting area and multiple neighboring areas within a one-layer proximity range; The feature bias is obtained by calculating the average Euclidean distance between the fused feature vectors of the current planting area and the neighboring areas, and the first bias P1 is obtained. Based on the second-level proximity range, calculate the second deviation P2 of the current planting area; If both P1 and P2 are below the preset deviation, and P1 ≥ P2, then the current planting area is marked as an associated area; Analysis of all mulberry planting areas yielded multiple related regions.

[0010] In this solution, S4 specifically refers to: A multi-cycle UAV image acquisition scheme is set up based on the associated area, and the preset spectral index and the image pest and disease index are weighted and calculated based on the multispectral image data and high-definition image data under the multi-cycle, to obtain the result G; The result G is subjected to univariate linear fitting over multiple periods, specifically with the period time T as the independent variable and the value of G as the dependent variable, to obtain the fitting curve and fitting coefficient. Based on the fitted curves, the correlation trends of pest and disease conditions in multiple related regions were assessed over multiple periods, and the average error between the fitted curves and the actual values ​​was calculated over multiple periods. ; like If the preset error is less than 0 and the fitting coefficient is less than 0, the corresponding associated region will be marked as a source area of ​​pests and diseases.

[0011] In this solution, S5 specifically refers to: By weighting the preset spectral index and the image pest and disease index in the associated region, the characteristics and impact trends of pests and diseases are evaluated, and the prevention and control priorities of the associated region are set. Based on the pest and disease source areas and related areas, pesticide plans were formulated, with general mulberry planting areas given the lowest priority and pest and disease source areas given the highest priority. A pest and disease control plan based on drones was also planned.

[0012] In this solution, the pest and disease control plan is sent to the display unit.

[0013] A second aspect of the present invention also provides a mulberry pest and disease analysis system based on UAV monitoring. The system includes: a memory, a processor, and a data interface. The memory includes a UAV-based mulberry pest and disease analysis program. When executed by the processor, the UAV-based mulberry pest and disease analysis program performs the following steps: S1: The drone will cruise and collect multispectral and RGB high-definition images of multiple mulberry planting areas according to the preset route, and obtain multispectral image data and high-definition image data. S2: Preprocess the image data, generate multi-dimensional spectral indices from multispectral image data, extract texture and color features and assess pests and diseases from high-definition image data to obtain multi-dimensional image pest and disease indices, and construct a fusion feature vector for each planting area based on the spectral indices and image pest and disease indices. S3: For each planting area, calculate the difference of the corresponding fused feature vector using Euclidean distance, analyze the feature deviation in the preset neighboring areas of the planting area, and mark the planting areas with feature deviations below the preset range as associated areas; S4: Using drones, collect image data of related areas over multiple periods, and perform weighted calculations on preset spectral indices and preset image pest and disease indices. Based on the calculation results over multiple periods, evaluate the trend of pest and disease characteristics and screen out the source areas of pests and diseases. S5: Assess the characteristics of pests and diseases in the source areas and planting areas, set regional prevention and control priorities, and plan pest and disease control solutions based on drones.

[0014] A third aspect of the present invention also provides a computer-readable storage medium comprising a mulberry pest and disease analysis program based on UAV monitoring, wherein when the UAV-based mulberry pest and disease analysis program is executed by a processor, it implements the steps of the mulberry pest and disease analysis method based on UAV monitoring as described in any of the preceding claims.

[0015] This invention discloses a method and system for analyzing mulberry pests and diseases based on drone monitoring. The method includes: collecting multispectral and RGB high-definition images of multiple mulberry planting areas via drones along preset flight paths; preprocessing the image data to generate multi-dimensional spectral indices, analyzing image texture and color features to assess pests and diseases, constructing a fusion feature vector for the planting areas, calculating feature differences using Euclidean distance, and marking associated areas with feature deviations within a preset range; collecting multi-period associated area images, weighting and calculating preset indices and pest and disease indices, assessing pest and disease trends, and screening source areas; setting control priorities based on pest and disease characteristics, and planning drone-based pest and disease control schemes. This invention achieves precise monitoring, source location, and scientific control of mulberry pests and diseases, improving monitoring efficiency and control targeting, adapting to various planting scenarios, and effectively improving the level of intelligent management in mulberry planting. Attached Figure Description

[0016] Figure 1 A flowchart of a method for analyzing mulberry pests and diseases based on UAV monitoring according to the present invention is shown; Figure 2 The flowchart of the fusion feature vector construction process of the present invention is shown; Figure 3 A block diagram of a mulberry pest and disease analysis system based on UAV monitoring according to the present invention is shown. Detailed Implementation

[0017] To better understand the above-mentioned objects, features, and advantages of the present invention, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. In the embodiments, unless otherwise specified, the embodiments and features described herein can be combined with each other.

[0018] 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 therefore the scope of protection of the invention is not limited to the specific embodiments disclosed below.

[0019] Figure 1 The flowchart of a method for analyzing mulberry pests and diseases based on UAV monitoring according to the present invention is shown.

[0020] like Figure 1 As shown, the first aspect of the present invention provides a method for analyzing mulberry pests and diseases based on unmanned aerial vehicle (UAV) monitoring, comprising: S1: The drone will cruise and collect multispectral and RGB high-definition images of multiple mulberry planting areas according to the preset route, and obtain multispectral image data and high-definition image data. S2: Preprocess the image data, generate multi-dimensional spectral indices from multispectral image data, extract texture and color features and assess pests and diseases from high-definition image data to obtain multi-dimensional image pest and disease indices, and construct a fusion feature vector for each planting area based on the spectral indices and image pest and disease indices. S3: For each planting area, calculate the difference of the corresponding fused feature vector using Euclidean distance, analyze the feature deviation in the preset neighboring areas of the planting area, and mark the planting areas with feature deviations below the preset range as associated areas; S4: Using drones, collect image data of related areas over multiple periods, and perform weighted calculations on preset spectral indices and preset image pest and disease indices. Based on the calculation results over multiple periods, evaluate the trend of pest and disease characteristics and screen out the source areas of pests and diseases. S5: Assess the characteristics of pests and diseases in the source areas and planting areas, set regional prevention and control priorities, and plan pest and disease control solutions based on drones.

[0021] According to an embodiment of the present invention, S1 specifically includes: Based on the target mulberry area, multiple mulberry planting areas are divided, and the mulberry planting density in each mulberry planting area is within a preset range; Based on the geographical information, planting density and topographic features of the mulberry planting area, a grid-based flight route planning method is adopted to set preset flight routes, and the spacing between the flight routes is adaptively adjusted according to the planting density. In each mulberry planting area, multispectral and high-resolution image data were acquired using drone data acquisition units.

[0022] In this embodiment, the preset flight path may include a single drone flight path or multiple drone flight paths. The multispectral camera may include visible light band, near-infrared band, red-edge band, etc., with a spectral range of 450-950nm and a spectral resolution of ≤10nm. The RGB high-definition camera has a resolution of ≥20 million pixels. The drone's cruising speed is set to 1-4 meters per second, and image data is acquired for each planting area.

[0023] According to an embodiment of the present invention, step S2 specifically includes: A radiometric correction algorithm based on the dark target method is used to extract pixels and perform radiometric correction on multispectral images. Based on the geographical coordinates of the mulberry planting area, a quadratic polynomial fitting algorithm is used to correct the geometric distortion of the image. A Gaussian filtering algorithm is used to denoise the multispectral image. A histogram equalization algorithm is used to adjust the brightness and contrast of high-definition image data. The Sobel operator was used for background segmentation to separate the mulberry plant area from the background area, and finally, image enhancement preprocessing was performed.

[0024] In this embodiment, geometric distortion of the image is corrected to ensure that each image pixel corresponds one-to-one with the actual geographical location. Image data preprocessing includes processing of multispectral and high-resolution image data. The filtering process can be configured with a 3×3 filter kernel size to remove Gaussian and random noise from the image while preserving spectral details and avoiding the loss of crucial information during denoising.

[0025] Adjust the brightness and contrast of the high-definition image data to make the color distribution of the image more uniform, highlight the color changes of mulberry leaves and fruits, and facilitate subsequent color feature extraction; the segmented region includes the mulberry plant (leaves and fruits) region and the background region (soil, weeds, gravel).

[0026] Figure 2 A flowchart illustrating the construction process of the fused feature vector of the present invention is shown.

[0027] According to an embodiment of the present invention, step S2 further includes: Multiband calculations and reflectance analysis were performed on the preprocessed multispectral image data, and multispectral index inversion analysis was conducted to obtain multidimensional spectral indices. Multidimensional spectral indices include normalized vegetation index, normalized red edge index, and leaf chlorophyll content inversion value; The preprocessed high-definition image data is converted to grayscale, and the texture features of mulberry plants are extracted based on the LBP operator. Color histograms are introduced to statistically analyze the color features of the high-definition image data. The texture features and color features are compared with the preset features for similarity, and the texture feature score and color feature score are obtained based on the similarity analysis results. Image disease and pest index includes texture feature degree and color feature degree; A fused feature vector is generated based on multiple dimensions, including normalized vegetation index, normalized red edge index, leaf chlorophyll content inversion value, texture feature degree, and color feature degree. A fused feature vector for each planting area is constructed based on spectral indices and image pest and disease indices.

[0028] In this embodiment, during the similarity comparison between texture features, color features, and preset features, the features are vectorized, and the similarity is calculated using Euclidean distance. The preset features include standardized texture features and color features, i.e., image feature data under ideal planting conditions, used to calculate the similarity. Two similarity scores (i.e., texture feature score and color feature score) are obtained based on the texture and color features respectively. The texture feature score and color feature score can be used to directly reflect the pest and disease index of the image; the higher the similarity score, the lower the level of pests and diseases.

[0029] Here, the Normalized Difference Vegetation Index (NDVI) is calculated using reflectance in the near-infrared and red bands, reflecting the growth status of mulberry plants and the degree of pest and disease stress. The Normalized Red Edge Index (NDRE) is calculated using reflectance in the red edge and near-infrared bands, used for early identification of pest and disease infection in mulberry plants, especially suitable for monitoring the latent stage of pests and diseases. The chlorophyll content inversion value is calculated based on the reflectance of multispectral bands by constructing an inversion model, reflecting the chlorophyll content of mulberry leaves. Pest and disease infection leads to a decrease in chlorophyll content, so this index can be used as one of the core indicators for pest and disease identification. The multidimensional spectral parameters here are merely illustrative. Based on the needs of mulberry research, other spectral parameters can be set to measure pest and disease characteristics to meet different analytical needs.

[0030] In this invention, by combining spectral features and high-definition image features to analyze the characteristics of pests and diseases in different planting areas, it is possible to effectively achieve multi-dimensional assessment of pest and disease characteristics. Furthermore, by introducing fused features to screen related regions, it is possible to avoid the high error rate caused by single-factor assessment methods.

[0031] According to an embodiment of the present invention, step S3 specifically includes: Mark a current planting area and calculate the feature deviation between the current planting area and multiple neighboring areas within a one-layer proximity range; The feature bias is obtained by calculating the average Euclidean distance between the fused feature vectors of the current planting area and the neighboring areas, and the first bias P1 is obtained. Based on the second-level proximity range, calculate the second deviation P2 of the current planting area; If both P1 and P2 are below the preset deviation, and P1 ≥ P2, then the current planting area is marked as an associated area; Analysis of all mulberry planting areas yielded multiple related regions.

[0032] In this embodiment, the adjacent ranges of the first and second layers are specifically marked by the adjacent areas of the current planting area that are separated by a certain range. For example, in the grid-divided planting area, the current planting area has 8 adjacent areas in the first layer, separated by 0 areas, and the adjacent areas in the first layer can be 16, separated by 1 area.

[0033] According to an embodiment of the present invention, step S4 specifically includes: A multi-cycle UAV image acquisition scheme is set up based on the associated area, and the preset spectral index and the image pest and disease index are weighted and calculated based on the multispectral image data and high-definition image data under the multi-cycle, to obtain the result G; The result G is subjected to univariate linear fitting over multiple periods, specifically with the period time T as the independent variable and the value of G as the dependent variable, to obtain the fitting curve and fitting coefficient. Based on the fitted curves, the correlation trends of pest and disease conditions in multiple related regions were assessed over multiple periods, and the average error between the fitted curves and the actual values ​​was calculated over multiple periods. ; like If the preset error is less than 0 and the fitting coefficient is less than 0, the corresponding associated region will be marked as a source area of ​​pests and diseases.

[0034] In this embodiment, image acquisition includes the acquisition of multispectral image data and high-resolution image data. The fitting coefficient is k in the univariate fitting equation y=kx+b. A fitting coefficient less than 0 indicates that the result G shows a linear decrease, and this region is more likely to be a pathogenic area. The preset error can be set to within 5%.

[0035] The weighted calculation is as follows: G=K1×N1+K2×N2+K3×W+K4×S; Wherein, G represents the weighted pest and disease characteristic value. The larger the value, the closer the pest and disease situation is to the expected situation. K1-K4 are the weight values. N1 and N2 represent preset spectral indices, such as the normalized vegetation index and leaf chlorophyll content. W and S are texture characteristic degree and color characteristic degree, respectively.

[0036] In this invention, the selected associated regions are those with a certain trend of pest and disease influence, with a certain range of pest and disease influence trends in neighboring regions, and whose pest and disease feature information based on spectral and high-definition images is similar to that of neighboring regions. Furthermore, based on the feature deviation of neighboring regions, associated regions can be classified. Through association classification, priority settings for pesticide spraying and other prevention and control schemes can be set for associated regions with different influence trends.

[0037] According to an embodiment of the present invention, step S5 specifically includes: By weighting the preset spectral index and the image pest and disease index in the associated region, the characteristics and impact trends of pests and diseases are evaluated, and the prevention and control priorities of the associated region are set. Based on the pest and disease source areas and related areas, pesticide plans were formulated, with general mulberry planting areas given the lowest priority and pest and disease source areas given the highest priority. A pest and disease control plan based on drones was also planned.

[0038] In this embodiment, typical mulberry planting areas are considered as unrelated areas and non-pest / disease-prone areas. Pest / disease control plans include application schemes based on control priorities (pesticide selection, application parameter settings, application timing), and route plans.

[0039] In this invention, the screening of pathogen areas based on associated regions can reduce unnecessary regional analysis processes, improve the system's ability to accurately predict pathogen areas, reduce data collection and monitoring processes, and reduce drone resource consumption. It has high practicality for the analysis of disease trends and the prediction of the impact range of pests and diseases in large-scale mulberry planting.

[0040] Finally, the pest and disease control plan is sent to the display unit. The display unit includes computer equipment and mobile devices.

[0041] According to an embodiment of the present invention, it further includes: Based on the pest and disease control plan, control and monitor the planting area, obtain image data of the current multi-cycle during the control process, calculate the fitting curve of the pest and disease source area, and use the fitting curve to predict the data of N cycles to obtain N prediction result values. Obtain a set of environmental parameters for the current multi-cycle pest and disease source area; Construct a multi-dimensional environmental suitability evaluation system based on soil moisture, light intensity, temperature, and wind force, and evaluate the environmental suitability for each current period through a set of environmental parameters; The overall rate of change of environmental suitability is analyzed based on the current multi-cycle environmental suitability analysis. The N predicted values ​​are corrected based on the comprehensive rate of change, and the control effect and the impact trend of the pest and disease source area are evaluated based on the corrected predicted values ​​over the current multi-cycle period.

[0042] It is worth mentioning that pests and diseases are quite sensitive to environmental changes. The prediction of the characteristics of the source area through comprehensive assessment often needs to be modified by combining environmental suitability in order to improve the accuracy of the impact trend prediction and conform to the future trend.

[0043] This invention constructs an environmental suitability evaluation system for pest and disease reproduction. It allows setting suitable ranges and scoring standards for each environmental parameter. The suitable temperature range is 20-28℃, the suitable light intensity range is 5000-8000 lux, the suitable soil moisture content range is 15%-25%, and the suitable wind force range is 0-3. Each environmental parameter has a scoring range of 0-100 points. Within the suitable range, a score ≥80 points is awarded; outside the suitable range, the score is reduced based on the degree of deviation, with greater deviation resulting in a lower score. A weighted summation algorithm is used to calculate environmental suitability. Different parameter weights can be set for different planting environments. For example, temperature has a weight of 0.3, light intensity a weight of 0.25, soil moisture content a weight of 0.25, and wind force a weight of 0.2. Environmental suitability = 0.3 × temperature score + 0.25 × light intensity score + 0.25 × soil moisture content score + 0.2 × wind force score. A higher environmental suitability indicates that the area is more suitable for pest and disease growth and reproduction. Calculation of the comprehensive change rate of environmental suitability over multiple periods: Calculate the rate of change of environmental suitability over multiple periods: Rate of change = (Current period environmental suitability - Previous period environmental suitability) / Previous period environmental suitability × 100%. Calculate the average rate of change over the current multiple periods as the comprehensive rate of change. If the absolute value of the comprehensive rate of change is ≥10%, it is considered that the environmental change is drastic, and the predicted value can be dynamically adjusted based on the change in the comprehensive rate of change. Corrected predicted value = Original predicted value × (1 - Environmental suitability change rate × 0.5). If the comprehensive rate of change is positive, it means that the suitability is increasing overall, and the predicted value should be decreased accordingly. If the comprehensive rate of change is negative, it means that the suitability is decreasing overall, and the predicted value should be increased accordingly.

[0044] In this embodiment, the fitting curve is a fitting curve based on the weighted result of a preset spectral index and an image pest and disease index.

[0045] Figure 3 A block diagram of a mulberry pest and disease analysis system based on UAV monitoring according to the present invention is shown.

[0046] A second aspect of the present invention also provides a mulberry pest and disease analysis system based on UAV monitoring. The system includes: a memory, a processor, and a data interface. The memory includes a UAV-based mulberry pest and disease analysis program. When executed by the processor, the UAV-based mulberry pest and disease analysis program performs the following steps: S1: The drone will cruise and collect multispectral and RGB high-definition images of multiple mulberry planting areas according to the preset route, and obtain multispectral image data and high-definition image data. S2: Preprocess the image data, generate multi-dimensional spectral indices from multispectral image data, extract texture and color features and assess pests and diseases from high-definition image data to obtain multi-dimensional image pest and disease indices, and construct a fusion feature vector for each planting area based on the spectral indices and image pest and disease indices. S3: For each planting area, calculate the difference of the corresponding fused feature vector using Euclidean distance, analyze the feature deviation in the preset neighboring areas of the planting area, and mark the planting areas with feature deviations below the preset range as associated areas; S4: Using drones, collect image data of related areas over multiple periods, and perform weighted calculations on preset spectral indices and preset image pest and disease indices. Based on the calculation results over multiple periods, evaluate the trend of pest and disease characteristics and screen out the source areas of pests and diseases. S5: Assess the characteristics of pests and diseases in the source areas and planting areas, set regional prevention and control priorities, and plan pest and disease control solutions based on drones.

[0047] The system is capable of performing one or more steps in the aforementioned method for analyzing mulberry pests and diseases based on UAV monitoring. The data interface is used for data interaction with the acquisition and display unit.

[0048] A third aspect of the present invention also provides a computer-readable storage medium comprising a mulberry pest and disease analysis program based on UAV monitoring, wherein when the UAV-based mulberry pest and disease analysis program is executed by a processor, it implements the steps of the mulberry pest and disease analysis method based on UAV monitoring as described in any of the preceding claims.

[0049] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods, such as: multiple units or components can be combined, or integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the various components shown or discussed can be through some interfaces, and the indirect coupling or communication connection between devices or units can be electrical, mechanical, or other forms.

[0050] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units. They may be located in one place or distributed across multiple network units. Some or all of the units may be selected to achieve the purpose of this embodiment according to actual needs.

[0051] In addition, in the various embodiments of the present invention, each functional unit can be integrated into one processing unit, or each unit can be a separate unit, or two or more units can be integrated into one unit; the integrated unit can be implemented in hardware or in the form of hardware plus software functional units.

[0052] Those skilled in the art will understand that all or part of the steps of the above method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it performs the steps of the above method embodiments. The aforementioned storage medium includes various media capable of storing program code, such as mobile storage devices, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0053] Alternatively, if the integrated units of this invention are implemented as software functional modules and sold or used as independent products, they can also be stored in a computer-readable storage medium. Based on this understanding, the technical solutions of the embodiments of this invention, or the parts that contribute to the prior art, 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 methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as mobile storage devices, ROM, RAM, magnetic disks, or optical disks.

[0054] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for analyzing mulberry pests and diseases based on unmanned aerial vehicle (UAV) monitoring, characterized in that, include: S1: The drone will cruise and collect multispectral and RGB high-definition images of multiple mulberry planting areas according to the preset route, and obtain multispectral image data and high-definition image data. S2: Preprocess the image data, generate multi-dimensional spectral indices from multispectral image data, extract texture and color features and assess pests and diseases from high-definition image data to obtain multi-dimensional image pest and disease indices, and construct a fusion feature vector for each planting area based on the spectral indices and image pest and disease indices. S3: For each planting area, calculate the difference of the corresponding fused feature vector using Euclidean distance, analyze the feature deviation in the preset neighboring areas of the planting area, and mark the planting areas with feature deviations below the preset range as associated areas; S4: Using drones, collect image data of related areas over multiple periods, and perform weighted calculations on preset spectral indices and preset image pest and disease indices. Based on the calculation results over multiple periods, evaluate the trend of pest and disease characteristics and screen out the source areas of pests and diseases. S5: Assess the characteristics of pests and diseases in the source areas and planting areas, set regional prevention and control priorities, and plan pest and disease control solutions based on drones. Specifically, S3 is: Mark a current planting area and calculate the feature deviation between the current planting area and multiple neighboring areas within a one-layer proximity range; The feature bias is obtained by calculating the average Euclidean distance between the fused feature vectors of the current planting area and the neighboring areas, and the first bias P1 is obtained. Based on the second-level proximity range, calculate the second deviation P2 of the current planting area; If both P1 and P2 are below the preset deviation, and P1 ≥ P2, then the current planting area is marked as an associated area; Analysis of all mulberry planting areas yielded multiple related regions; Specifically, S4 is: A multi-cycle UAV image acquisition scheme is set up based on the associated area, and the preset spectral index and the image pest and disease index are weighted and calculated based on the multispectral image data and high-definition image data under the multi-cycle, to obtain the result G; A univariate linear fit is performed on the result G over multiple periods, specifically with the period time T as the independent variable and G as the dependent variable, to obtain the fitted curve and the fitted coefficient. Based on the fitted curves, the correlation trends of pest and disease conditions in multiple related regions were assessed over multiple periods, and the average error between the fitted curves and the actual values ​​was calculated over multiple periods. ; like If the preset error is less than 0 and the fitting coefficient is less than 0, the corresponding associated region will be marked as a source area of ​​pests and diseases.

2. The method for analyzing mulberry pests and diseases based on UAV monitoring according to claim 1, characterized in that, Specifically, S1 is: Based on the target mulberry area, multiple mulberry planting areas are divided, and the mulberry planting density in each mulberry planting area is within a preset range; Based on the geographical information, planting density and topographic features of the mulberry planting area, a grid-based flight route planning method is adopted to set preset flight routes, and the spacing between the flight routes is adaptively adjusted according to the planting density. In each mulberry planting area, multispectral and high-resolution image data were acquired using drone data acquisition units.

3. The method for analyzing mulberry pests and diseases based on UAV monitoring according to claim 2, characterized in that, Specifically, S2 is: A radiometric correction algorithm based on the dark target method is used to extract pixels and perform radiometric correction on multispectral images. Based on the geographical coordinates of the mulberry planting area, a quadratic polynomial fitting algorithm is used to correct the geometric distortion of the image. A Gaussian filtering algorithm is used to denoise the multispectral image. A histogram equalization algorithm is used to adjust the brightness and contrast of high-definition image data. The Sobel operator was used for background segmentation to separate the mulberry plant area from the background area, and finally, image enhancement preprocessing was performed.

4. The method for analyzing mulberry pests and diseases based on UAV monitoring according to claim 3, characterized in that, The S2 further includes: Multiband calculations and reflectance analysis were performed on the preprocessed multispectral image data, and multispectral index inversion analysis was conducted to obtain multidimensional spectral indices. Multidimensional spectral indices include normalized vegetation index, normalized red edge index, and leaf chlorophyll content inversion value; The preprocessed high-definition image data is converted to grayscale, and the texture features of mulberry plants are extracted based on the LBP operator. Color histograms are introduced to statistically analyze the color features of the high-definition image data. The texture features, color features and preset features are compared for similarity, and the texture feature score and color feature score are obtained based on the similarity analysis results. Image disease and pest index includes texture feature degree and color feature degree; The normalized vegetation index, normalized red edge index, leaf chlorophyll content inversion value, texture feature degree and color feature degree are used as multiple dimensions to generate a fused feature vector. A fused feature vector for each planting area is constructed based on spectral indices and image pest and disease indices.

5. The method for analyzing mulberry pests and diseases based on UAV monitoring according to claim 1, characterized in that, Specifically, S5 is: By weighting the preset spectral index and the image pest and disease index in the associated region, the characteristics and impact trends of pests and diseases are evaluated, and the prevention and control priorities of the associated region are set. Based on the pest and disease source areas and related areas, pesticide plans were formulated, with general mulberry planting areas given the lowest priority and pest and disease source areas given the highest priority. A pest and disease control plan based on drones was also planned.

6. The method for analyzing mulberry pests and diseases based on UAV monitoring according to claim 1, characterized in that, include: Send the pest and disease control plan to the display unit.

7. A mulberry pest and disease analysis system based on unmanned aerial vehicle (UAV) monitoring, characterized in that, The system includes: a memory, a processor, and a data interface. The memory includes a mulberry pest and disease analysis program based on UAV monitoring. When the UAV-based mulberry pest and disease analysis program is executed by the processor, it implements the steps of the mulberry pest and disease analysis method based on UAV monitoring as described in claim 1.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a mulberry pest and disease analysis program based on UAV monitoring. When the UAV-based mulberry pest and disease analysis program is executed by a processor, it implements the steps of the mulberry pest and disease analysis method based on UAV monitoring as described in any one of claims 1 to 6.

Images