An Internet of Things-based grape disease prediction method, device, terminal and medium

Abstract

The present application relates to a grape disease prediction method, device, terminal and medium based on the Internet of Things, belonging to the technical field of grape disease prediction. The grape disease prediction method includes: collecting the actual image information of grapes and the actual environmental data in the detection area; determining the grape variety, the growth stage and the actual surface characteristics of the current grapes according to the actual image information; determining the standard surface characteristics and standard environmental data corresponding to the growth stage according to the grape variety; comparing the standard surface characteristics with the actual surface characteristics and the standard environmental data with the actual environmental data; if both the surface characteristics and the environmental data do not match, then retrieving the disease statistics table matching the growth stage, and screening the first predicted disease matching the actual environmental data and the actual surface characteristics, and sending the first warning information. The present application has the beneficial effects of effectively improving the disease prevention ability in the grape planting process and reducing the risk of economic losses caused by diseases.

Description

Technical Field

[0001] The present application relates to the technical field of grape disease prediction, and particularly to a grape disease prediction method, device, terminal and medium based on the Internet of Things. Background Art

[0002] The grape cultivation industry is an important part of modern agriculture. With the continuous improvement of the level of agricultural modernization, precision agriculture technology has gradually become one of the key factors promoting the improvement of agricultural production efficiency. By introducing advanced monitoring and prediction means, the level of crop health management can be effectively improved, thereby reducing the incidence of pests and diseases, reducing the amount of pesticide use, ensuring the quality of agricultural products and food safety, while optimizing resource allocation and promoting sustainable development. At present, grape cultivation industries have been formed in many regions, and the research on grape cultivation is also relatively in-depth.

[0003] However, in terms of grape disease management, traditional methods still rely on manual observation and empirical judgment. Specifically, it usually adopts the method of regularly inspecting the fields, and combines the visual observation of the changes in parts such as leaves and fruits to discover potential disease problems. In addition, methods such as using a microscope to examine infected samples or using chemical reagents to test soil components are also used to assist in diagnosis. In recent years, some new tools and technologies have also been applied to this field. For example, portable spectrometers can be used to quickly detect changes in the content of plant pigments; unmanned aerial vehicles equipped with multi-spectral cameras can cover large areas for remote sensing imaging analysis, etc. Although these methods have their own characteristics and make up for the deficiencies of relying solely on manual operations to a certain extent, there are still many limitations.

[0004] The above-mentioned various conventional means generally have the following defects: they cannot comprehensively and accurately capture the early subtle lesion signs, so that they cannot play a good role in disease prevention, thus delaying the best prevention and control time and resulting in a relatively high risk of economic losses. Summary of the Invention

[0005] In order to overcome the above technical problems, the present application provides a grape disease prediction method, device, terminal and medium based on the Internet of Things.

[0006] In the first aspect, the present application provides a grape disease prediction method based on the Internet of Things, adopting the following technical solution:

[0007] A grape disease prediction method based on the Internet of Things includes:

[0008] Collect the actual image information of grapes in the detection area and the actual environmental data of the grapes;

[0009] Determine the grape variety, growth stage and actual surface characteristics of the current grapes according to the actual image information;

[0010] Determine the standard surface features and standard environmental data corresponding to the growth stage according to the grape variety;

[0011] Compare the standard surface features with the actual surface features and the standard environmental data with the actual environmental data;

[0012] If the actual surface features do not match the standard surface features and the actual environmental data do not match the standard environmental data, retrieve the disease statistics table that matches the growth stage;

[0013] Screen the first predicted disease that matches the actual environmental data and the actual surface features from the disease statistics table, and send the first warning message.

[0014] By adopting the above technical solution, a grape disease prediction method based on the Internet of Things is realized. This method comprehensively collects and analyzes the actual image information and environmental data of grapes in the detection area, and combines the grape variety, growth stage and their corresponding standardized data for accurate comparison. It can quickly locate potential disease types when the surface features and environmental data are both abnormal, and send warning messages in time, thus effectively improving the disease prevention ability in the grape planting process and reducing the risk of economic losses caused by diseases.

[0015] Optionally, the step of screening the first predicted disease that matches the actual environmental data and the actual surface features from the disease statistics table includes:

[0016] Determine the recognition framework according to the disease statistics table;

[0017] Construct a basic probability assignment according to the actual environmental data and the actual surface features;

[0018] Use the Dempster combination rule to combine the final combined probability;

[0019] Calculate the belief function and the likelihood function according to the final combined probability;

[0020] Obtain the predicted disease type according to the belief function and the likelihood function.

[0021] By adopting the above technical solutions, a disease prediction method based on Dempster-Shafer evidence theory is realized, which improves the accuracy and reliability of grape disease prediction. The specific effects are as follows: 1) Determine the recognition framework, clarify the basic elements and scope of disease prediction, and ensure the consistency of subsequent calculations. 2) Construct the basic probability assignment, fully consider the uncertainty of actual environmental data and actual surface features, and enhance the data adaptability of the model. 3) Use Dempster's combination rule to synthesize the final combined probability, effectively integrate multi-source information, and enhance the credibility of the prediction results. 4) Calculate the belief function and the likelihood function, further quantify the confidence level of the prediction results, and provide a scientific basis for decision-making.

[0022] Optionally, before obtaining the predicted disease type according to the belief function and the likelihood function, it includes:

[0023] Obtain the difference between the belief function and the likelihood function;

[0024] Judge whether the absolute value of the difference exceeds the difference threshold;

[0025] If not, obtain the predicted disease type according to the belief function and the likelihood function.

[0026] By adopting the above technical solutions, through the evaluation of the difference between the belief function and the likelihood function, the accuracy of the disease prediction result is further improved. Specifically, when the difference does not exceed the set threshold, it indicates that the prediction model has a high confidence level. At this time, the disease can be directly searched according to the final prediction value, thereby reducing the misjudgment rate and improving the prediction efficiency.

[0027] Optionally, the judgment of whether the difference exceeds the difference threshold further includes:

[0028] If so, define the current detection point as an abnormal point, and increase the acquisition frequency and acquisition density of the surface features and environmental data corresponding to the current grape variety;

[0029] If the number of occurrences of abnormal points exceeds the first number threshold within the same detection time, define the detection time as an abnormal detection time;

[0030] If the number of occurrences of the abnormal detection time exceeds the second number threshold within a detection period, select the predicted disease type with the most occurrences among all abnormal points as the final predicted disease type.

[0031] By adopting the above technical solution, when the difference between the belief function and the likelihood function exceeds the difference threshold, abnormal situations can be identified in a timely manner and the current detection point can be defined as an abnormal point, thereby improving the reliability of the prediction model. At the same time, by increasing the acquisition frequency and acquisition density of the surface features and environmental data corresponding to the current grape variety, the accuracy and comprehensiveness of the data are further enhanced. During the same detection time, if the number of occurrences of abnormal points exceeds the set first number threshold, then this time period is marked as an abnormal detection time, which helps to focus on the high-incidence periods of problems. If the abnormal detection time occurs frequently and exceeds the second number threshold, then through statistical analysis of the final prediction value with the most occurrences among all abnormal points, it is ensured that the prediction result is more representative. Finally, according to the selected prediction value, the disease type is accurately matched, effectively improving the accuracy and response speed of grape disease prediction.

[0032] Optionally, after comparing the standard surface features with the actual surface features and the standard environmental data with the actual environmental data, it further includes:

[0033] If only the actual surface features do not match the standard surface features, then determine whether there is abnormal fluctuation in the actual environmental data within the first time window;

[0034] If so, obtain the detection points with fluctuations greater than the standard fluctuation curve;

[0035] Screen similar environmental data and calculate the average environmental data;

[0036] Screen the second predicted disease that matches the average environmental data and the actual surface features from the disease statistics table and send a second warning message;

[0037] If not, screen the first predicted disease that matches the actual environmental data and the actual surface features from the disease statistics table and send a first warning message.

[0038] By adopting the above technical solution, when the actual surface features do not match the standard surface features, further determine the abnormal fluctuation situation of the actual environmental data within the first time window. If there is abnormal fluctuation, by screening similar environmental data and calculating the average environmental data, the accuracy of the data is improved, thereby screening out a more accurate second predicted disease and sending a second warning message; if there is no abnormal fluctuation, directly screen the first predicted disease and send a first warning message. This method effectively reduces the misjudgment rate and improves the accuracy and reliability of grape disease prediction.

[0039] Optionally, the step of increasing the acquisition frequency and acquisition density of the surface features and environmental data corresponding to the current grape variety includes:

[0040] Obtain the offset value of the absolute value of the difference and the difference threshold ; , is the difference, is the difference threshold;

[0041] According to the offset value and the preset hierarchical adjustment rule, determine the frequency coefficient and the density coefficient ;

[0042] Obtain the reference sampling frequency and the reference sampling density ; ， N is the number of initially activated image acquisition devices, S is the area of the detection region, and K is the regional complexity coefficient; ;

[0043] Input the reference sampling frequency , the frequency coefficient and the offset value into the sampling frequency adjustment model to obtain the adjusted new sampling frequency ; The sampling frequency adjustment model is ;

[0044] According to the new sampling frequency , determine the acquisition interval;

[0045] Input the reference sampling density , the density coefficient and the offset value into the sampling density adjustment model to obtain the adjusted new sampling density ; The sampling density adjustment model is , is the area compensation coefficient, is the critical area, i.e., the area of the entire grape planting region;

[0046] According to the new sampling density , obtain the number of newly activated image acquisition devices ; ;

[0047] According to the detection region area S, obtain the total layout number of the sensor group ; , m is the redundancy coefficient, is the number of spare groups for dealing with sudden failures, is the reference acquisition density of the sensor group;

[0048] According to the offset value and the total layout quantity to obtain the newly added number of opening groups , .

[0049] Optionally, after comparing the standard surface features with the actual surface features and the standard environmental data with the actual environmental data, it further includes:

[0050] If only the actual environmental data does not match the standard environmental data, it is determined whether the actual environmental data continues to not match within the second time window;

[0051] If so, screen the third predicted disease that matches the actual environmental data from the disease statistics table and send a third warning message;

[0052] If not, screen the first predicted disease that matches the actual environmental data and the actual surface features from the disease statistics table and send a first warning message.

[0053] By adopting the above technical solution, when the actual environmental data does not match the standard environmental data, it can be further determined whether this mismatch is persistent. If the actual environmental data continues to not match within the second time window, a more accurate third predicted disease can be screened from the disease statistics table and a third warning message can be sent, thereby improving the accuracy of disease prediction. If the condition of continuous mismatch is not met, as a fallback, the first predicted disease is matched based on the actual environmental data and the actual surface features and a first warning message is sent, ensuring the robustness and comprehensiveness of the prediction system.

[0054] In a second aspect, the present application provides an Internet of Things-based grape disease prediction device, adopting the following technical solution:

[0055] An Internet of Things-based grape disease prediction device, comprising:

[0056] An image acquisition module, configured to acquire the actual image information of grapes in the detection area;

[0057] A data acquisition module, configured to acquire the actual environmental data of the grapes in the detection area;

[0058] An image analysis module, configured to determine the grape variety, the growth stage, and the actual surface features of the current grapes according to the actual image information;

[0059] A screening and matching module, configured to determine the standard surface features and standard environmental data corresponding to the growth stage according to the grape variety;

[0060] A comparison processing module for comparing the standard surface features with the actual surface features and the standard environmental data with the actual environmental data; if the actual surface features do not match the standard surface features and the actual environmental data do not match the standard environmental data, a disease statistics table matching the growth stage is retrieved.

[0061] The screening and matching module is used to screen the first predicted disease matching the actual environmental data and the actual surface features from the disease statistics table and send a first warning message.

[0062] In a third aspect, the present application provides a terminal, adopting the following technical solution:

[0063] A terminal, comprising:

[0064] A memory storing a grape disease prediction program based on the Internet of Things;

[0065] A processor for executing the program stored on the memory to implement the steps of the above-mentioned grape disease prediction method based on the Internet of Things.

[0066] In a fourth aspect, the present application provides a computer-readable storage medium, adopting the following technical solution:

[0067] A computer-readable storage medium storing a computer program capable of being loaded and executed by a processor to implement the above-mentioned grape disease prediction method based on the Internet of Things.

[0068] In summary, the present application includes at least one of the following beneficial technical effects:

[0069] By collecting the actual image information and environmental data of grapes and comparing them with the standard surface features and standard environmental data, the growth state and potential diseases of grapes can be accurately identified, significantly improving the accuracy of disease prediction and solving the problem that it is difficult for traditional methods to comprehensively capture early lesions.

[0070] Based on a multi-level warning mechanism under different matching conditions, such as abnormal fluctuation judgment and time window analysis, the reliability of disease prediction is further improved, the risk of misjudgment and missed diagnosis is reduced, and timely prevention and control measures are ensured.

[0071] Combining the Dempster synthesis rule and the disease statistics table to achieve scientific and quantitative prediction of disease types not only enhances the credibility of the prediction results but also provides strong support for subsequent precise prevention and control. BRIEF DESCRIPTION OF THE DRAWINGS

[0072] Figure 1 is a flowchart of an implementation manner of the first embodiment of the method of the present application;

[0073] Figure 2 It is a flowchart of a specific step of S160;

[0074] Figure 3 It is a schematic diagram of the combined probability of four pieces of evidence;

[0075] Figure 4 It is a flowchart of steps that can be executed after the difference between the belief function and the likelihood function exceeds the difference threshold;

[0076] Figure 5 It is a flowchart of improving the acquisition frequency and acquisition density;

[0077] Figure 6 It is an implementation manner of steps that can be executed after S140;

[0078] Figure 7 It is another implementation manner of steps that can be executed after S140;

[0079] Figure 8 It is a structural block diagram of an implementation manner of the first embodiment of the system of the present application. Specific implementation manner

[0080] To make the objectives, technical solutions, and advantages of the embodiments of the present application clearer, the following will combine the appended Figure 1 - appended Figure 8 , and clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the scope of protection of the present invention.

[0081] The first embodiment of the present application discloses a grape disease prediction method based on the Internet of Things. Referring to Figure 1 , as an implementation manner of the grape disease prediction method, the grape disease prediction method may include S110 - S160:

[0082] S110, collecting actual image information of grapes in the detection area and actual environmental data of the grapes;

[0083] S120, determining the grape variety, growth stage, and actual surface characteristics of the current grapes according to the actual image information;

[0084] S130, determining the standard surface characteristics and standard environmental data corresponding to the growth stage according to the grape variety;

[0085] S140, comparing the standard surface characteristics with the actual surface characteristics and the standard environmental data with the actual environmental data;

[0086] S150. If both the actual surface features do not match the standard surface features and the actual environmental data do not match the standard environmental data, retrieve the disease statistics table that matches the growth stage.

[0087] S160. Screen the first predicted disease that matches the actual environmental data and the actual surface features from the disease statistics table, and send the first warning message.

[0088] Specifically, the surface features of grapes include the leaf spot area and the degree of fruit discoloration; the environmental data includes temperature data and humidity data. Taking grapes planted in a greenhouse as an example, high-resolution digital cameras or cameras and other image acquisition devices are installed in the greenhouse. The area covered by the image acquisition devices is defined as the detection area. It can be understood that a detection area contains a certain number of image acquisition devices. In addition, a certain number of temperature sensors and humidity sensors are also installed in the detection area. The temperature sensors and humidity sensors can be installed at fixed positions or on mobile positions to collect data at different positions by moving.

[0089] Use an image recognition model based on deep learning (such as YOLOv4 or YOLOv5s) to classify the collected grape images to identify the grape varieties. The image recognition model based on deep learning is trained with a large amount of historical data of grape varieties. Analyze the color, shape, and texture features of grape leaves to determine the growth stage of the grapes. Use image processing algorithms (such as edge detection, color histogram analysis) to extract the spot area on the grape surface and the degree of fruit discoloration; the image processing algorithms are conventional recognition algorithms.

[0090] Based on historical data and expert experience, establish a standard surface feature library for different grape varieties at different growth stages, including leaf spot area, fruit discoloration degree, etc. Collect the standard environmental data (temperature, humidity range) corresponding to different growth stages, and associate it with the standard surface features to generate a grape comparison library.

[0091] Use an image processing algorithm (such as a convolutional neural network) to compare the collected actual surface features with the standard surface features and calculate the similarity; since one growth stage may correspond to multiple standard surface features, the actual surface features are compared with each standard surface feature. If the similarity after each comparison is below the similarity threshold, it is determined that the actual surface features do not match the standard surface features.

[0092] One growth stage corresponds to multiple standard temperature ranges and multiple standard humidity ranges; if the actual temperature is not within the multiple standard temperature ranges and the actual humidity is not within the multiple standard humidity ranges, it is determined that the actual environmental data does not match the standard environmental data.

[0093] The growth stages of grapes usually include the budding stage, the shoot growth stage, the flowering stage, the berry growth stage, the berry ripening stage, the defoliation stage, and the dormancy stage, etc. The sensitivities of different growth stages of different grape varieties to environmental conditions and diseases vary. Common grape disease types include downy mildew, powdery mildew, anthracnose, gray mold, etc. The occurrence of these diseases can be obtained through means such as field surveys, laboratory tests, and historical grape disease occurrence data, and the grape variety, growth stage, surface characteristics, environmental data, disease type, disease severity, etc. are recorded. A disease value range is assigned to different disease severities of disease types. For example, for downy mildew, the disease value range for level 0 can be set as 0 - 5; for level 1 it is 6 - 20; for level 2 it is 21 - 40; for level 3 it is 41 - 70; for level 4 it is 71 - 100; these data are associated into a disease information.

[0094] According to the grape variety and growth stage, retrieve the disease statistics table that matches the current growth stage from the disease database. The disease statistics table refers to multiple groups of disease information for the current grape variety at the current growth stage. The disease database records multiple groups of disease information corresponding to each growth stage for each grape variety.

[0095] After matching the first predicted disease based on the actual surface characteristics and actual environmental data, a first warning message will be sent, for example, sending the first warning message to the background monitoring room, etc.

[0096] Refer to Figure 2 Furthermore, the steps of screening the first predicted disease that matches the actual environmental data and actual surface characteristics from the disease statistics table can specifically include S161 - S165:

[0097] S161, determine the disease recognition framework according to the disease statistics table;

[0098] S162, construct the basic probability assignment according to the actual environmental data and actual surface characteristics;

[0099] S163, use the Dempster combination rule to combine the final combined probability;

[0100] S164, calculate the belief function and the likelihood function according to the final combined probability;

[0101] S165, obtain the predicted disease type according to the belief function and the likelihood function.

[0102] Specifically, assuming that the disease statistics table includes three types: downy mildew, powdery mildew, and anthracnose, then the disease recognition framework can be ; where A = downy mildew, B = powdery mildew, C = anthracnose.

[0103] The combined probability of four evidences refers to Figure 3as shown

[0104] Assume that a discoloration area of 35% may support B and C, a leaf spot coverage of 25% may support B and C, a high temperature of 32 °C may support C, and a high humidity of 85% may support C.

[0105] Step 1: Combine the first two pieces of evidence ( with )

[0106] Calculate the conflict coefficient ;

[0107] Synthesize the combined probability ( ):

[0108]

[0109] Step 2: Combine the result with the temperature evidence

[0110] Calculate the conflict coefficient ；

[0111] Synthesize the combined probability ( ):

[0112]

[0113] Step 3: Combine the result with the humidity evidence

[0114] Calculate the conflict coefficient ；

[0115] Synthesize the final combined probability ( ):

[0116]

[0117] The belief function (Bel) is , and the plausibility function (Pls) is , that is, to determine that the disease type is anthracnose; then, based on the deviation value between the actual environmental data and the standard environmental data and the deviation value between the actual surface characteristics and the standard surface characteristics, the severity is determined. For example, if the deviation values of both are greater than the maximum deviation threshold, the severity is level 4; if only the deviation value corresponding to the environmental data is greater than the maximum deviation threshold, the severity is level 3; if only the deviation value corresponding to the surface characteristics is greater than the maximum deviation threshold, the severity is level 2; if any one of the deviation values of the environmental data and the surface characteristics is greater than the minimum deviation threshold and less than the maximum deviation threshold, the severity is level 1; if the deviation values of the environmental data and the surface characteristics are both less than the minimum deviation threshold, the severity is level 0.

[0118] Refer to Figure 4 , further, after S164 and before S165, the difference between the belief function and the likelihood function can be obtained first, and then it is judged whether the absolute value of the difference exceeds the difference threshold. If not, S165 is executed; if so, S201 - S203 can be executed:

[0119] S201, define the current detection point as an abnormal point, and increase the acquisition frequency and acquisition density of the surface characteristics and environmental data corresponding to the current grape variety;

[0120] S202, if the number of times the abnormal point appears exceeds the first number threshold within the same detection time, then define this detection time as an abnormal detection time;

[0121] S203, if the number of times the abnormal detection time appears exceeds the second number threshold within one detection cycle, then select the predicted disease type with the most occurrences among all abnormal points as the final predicted disease type.

[0122] Specifically, the current detection point refers to the acquisition time point of the current grape surface characteristics and environmental data. If the absolute value of the difference between the belief function and the likelihood function corresponding to the current acquisition time point exceeds the preset difference threshold, then the current acquisition time point can be defined as an abnormal point, and the acquisition frequency and acquisition density of the surface characteristics and environmental data corresponding to the current grape variety are increased.

[0123] Refer to Figure 5 , the steps of increasing the acquisition frequency and acquisition density of the surface characteristics and environmental data can include S301 - S309:

[0124] S301, obtain the offset value of the absolute value of the difference and the difference threshold ; , is the difference, is the difference threshold;

[0125] S302, according to the offset value and a preset hierarchical adjustment rule to determine the frequency coefficient and the density coefficient ;

[0126] S303. Obtain the reference sampling frequency and the reference sampling density ; ， N is the number of initially turned-on image acquisition devices, S is the area of the detection region, and K is the regional complexity coefficient; ;

[0127] S304. Input the reference sampling frequency , the frequency coefficient and the offset value into the sampling frequency adjustment model to obtain the adjusted new sampling frequency ; The sampling frequency adjustment model is ;

[0128] S305. Determine the acquisition interval according to the new sampling frequency ;

[0129] S306. Input the reference sampling density , the density coefficient and the offset value into the sampling density adjustment model to obtain the adjusted new sampling density ; The sampling density adjustment model is , is the area compensation coefficient, is the critical area, i.e., the area of the entire grape planting region;

[0130] S307. Obtain the number of newly turned-on image acquisition devices according to the new sampling density ; ;

[0131] S308. Obtain the total layout number of the sensor group according to the detection region area S ; , m is the redundancy coefficient, is the number of standby groups for dealing with sudden failures, is the reference acquisition density of the sensor group;

[0132] S309. Obtain the newly added number of turned-on groups according to the offset value and the total layout number , .

[0133] For example, the hierarchical adjustment rule is:

[0134] When it is a mild deviation, the corresponding frequency coefficient = 0.5, and the density coefficient = 0.3;

[0135] When it is a moderate deviation, the corresponding frequency coefficient = 1.2, and the density coefficient = 0.8;

[0136] When it is a severe deviation, the corresponding frequency coefficient = 2.0, and the density coefficient = 1.5.

[0137] If a certain difference is 0.35 and the difference threshold is 0.2, then = 0.75, which belongs to a moderate deviation. Therefore, the frequency coefficient is determined to be 1.2, and the density coefficient = 0.8.

[0138] If the number N of initially opened image acquisition devices is 20 and the area S of the detection region is 80 , and the area of the entire grape planting area is 100 , is 0.3; then , then , then , so 38 new image acquisition devices need to be opened on the original basis, that is, a total of 58.

[0139] , ; then the acquisition interval can be , and then rounding up is 3. Therefore, the acquisition interval can be once every 3 hours. That is, the sensor group also acquires once every 3 hours.

[0140] For the sensor group (including temperature sensors and humidity sensors): If , , then groups. If , then , then 201 new sensor groups need to be opened on the original basis.

[0141] If the environmental data and surface features are both collected once every 3 hours; if the current image acquisition device is turned on at 58 and the number of sensor groups is 300 groups; assuming that according to historical detection data, one image acquisition device can cover the monitoring range of 6 sensor components; when the differences between 6 environmental data and the surface features collected by the camera all exceed the difference threshold, it indicates that this detection point is an abnormal point; assuming the first number threshold is 40, if there are 45 abnormal points, exceeding the first number threshold, so this moment is defined as the abnormal detection moment; if within a detection cycle, for example, within 21 hours, the number of abnormal detection moments appears 15 times, exceeding the second number threshold of 12 times, then the disease type with the most occurrences among all abnormal points can be selected as the final predicted disease type.

[0142] In addition, if within the same detection time, the number of abnormal points does not exceed the first number threshold, then S165 can be directly executed.

[0143] Refer to Figure 6 , further, after S140, if only the actual surface feature does not match the standard surface feature, then S401 - S405 can be executed:

[0144] S401, determine whether there is abnormal fluctuation in the actual environmental data within the first time window;

[0145] S402, if so, obtain the detection points with fluctuations greater than the standard fluctuation curve;

[0146] S403, screen similar environmental data and calculate the average environmental data;

[0147] S404, screen the second predicted disease that matches the average environmental data and the actual surface feature from the disease statistics table and send a second warning message;

[0148] S405, if not, screen the first predicted disease that matches the actual environmental data and the actual surface feature from the disease statistics table and send a first warning message.

[0149] Specifically, the first time window represents starting from the current detection time and monitoring the actual environmental data every half hour for one day. Construct an actual monitoring curve based on the monitoring data and monitoring time, and compare the actual monitoring curve with the standard fluctuation curve. The standard fluctuation curve represents the monitoring curve of historical environmental data that matches the grape type and growth stage within one day under normal circumstances. When comparing, screen out the detection points greater than the standard fluctuation curve, screen similar environmental data, and calculate the average environmental data; similar environmental data refers to the difference between two actual environmental data being less than the difference threshold.

[0150] The execution logics of steps S304 and S305 are the same as those of S161 - S165.

[0151] Referring to Figure 7 , further, after S140, if only the actual environmental data does not match the standard environmental data, then S501 - S503 can be executed:

[0152] S501, determine whether the actual environmental data continuously does not match within the second time window;

[0153] S502, if so, screen the third predicted disease that matches the actual environmental data from the disease statistics table and send a third warning message;

[0154] S503, if not, screen the first predicted disease that matches the actual environmental data and the actual surface characteristics from the disease statistics table and send a first warning message.

[0155] Specifically, only when both the actual environmental data and the standard environmental data do not match, it indicates that the actual surface characteristics are okay, so only the actual environmental data needs to be considered. The second time window represents that the environmental data is monitored every 1 hour within 12 hours. If the number of abnormal environmental data is greater than 1 / 4, it means that the actual environmental data continuously does not match within the second time window; abnormal means the environmental data is greater than the standard environmental data and the difference between the monitored environmental data and the actual environmental data is greater than the difference threshold.

[0156] The execution logics of S502 and S503 are the same as those of S161 - S165.

[0157] Based on the above method embodiments, the second embodiment of the present application discloses a grape disease prediction device based on the Internet of Things. Referring to Figure 8 , as an implementation manner of the grape disease prediction device, the grape disease prediction device may include:

[0158] An image acquisition module, configured to acquire the actual image information of grapes in the detection area;

[0159] A data acquisition module, configured to acquire the actual environmental data where the grapes in the detection area are located;

[0160] An image analysis module, configured to determine the grape variety, the growth stage where it is located, and the actual surface characteristics of the current grapes according to the actual image information;

[0161] A screening and matching module, configured to determine the standard surface characteristics and standard environmental data corresponding to the growth stage according to the grape variety;

[0162] A comparison processing module is used to compare the standard surface features with the actual surface features and the standard environmental data with the actual environmental data; if the actual surface features do not match the standard surface features and the actual environmental data do not match the standard environmental data, a disease statistics table matching the growth stage is retrieved.

[0163] A screening and matching module is used to screen the first predicted diseases that match the actual environmental data and the actual surface features from the disease statistics table and send a first warning message.

[0164] The modules of the grape disease prediction device based on the Internet of Things correspond one by one to the grape disease prediction method based on the Internet of Things, and will not be elaborated here.

[0165] The third embodiment of this application provides a terminal. As an implementation of this terminal, the terminal may include: a memory and a processor; wherein,

[0166] The memory is used to store the above-mentioned grape disease prediction program based on the Internet of Things;

[0167] The processor is used to execute the program stored on the memory to implement the steps of the above-mentioned grape disease prediction method based on the Internet of Things.

[0168] Among them, the memory can be communicatively connected to the processor through a communication bus, and the communication bus can be an address bus, a data bus, a control bus, etc.

[0169] In addition, the memory may include a random access memory (RAM), and may also include a non-volatile memory (NVM), such as at least one disk memory.

[0170] And the processor can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc.

[0171] The fourth embodiment of this application provides a computer-readable storage medium, storing a computer program that can be loaded and executed by a processor to implement the above-mentioned grape disease prediction method based on the Internet of Things.

[0172] The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or a data center that integrates one or more available media. Among them, the available medium can be a magnetic medium (for example, a floppy disk, a hard disk, a magnetic tape), an optical medium (for example, a DVD), or a semiconductor medium (for example, a solid-state drive), etc.

[0173] The above are all preferred embodiments of the present application, which do not successively limit the protection scope of the present application. Any feature disclosed in this specification (including the abstract and drawings), unless specifically described, can be replaced by other equivalent or similar-purpose alternative features. That is, unless specifically described, each feature is only an example of a series of equivalent or similar features.

Claims

1. A grape disease prediction method based on the Internet of Things, characterized in that, Including: Collecting the actual image information of grapes in the detection area and the actual environmental data where the grapes are located; Determining the grape variety, growth stage and actual surface characteristics of the current grapes according to the actual image information; Determining the standard surface characteristics and standard environmental data corresponding to the growth stage according to the grape variety; Comparing the standard surface characteristics with the actual surface characteristics and the standard environmental data with the actual environmental data; If the actual surface characteristics do not match the standard surface characteristics and the actual environmental data do not match the standard environmental data, then retrieve the disease statistics table matching the growth stage; Screening the first predicted disease that matches the actual environmental data and the actual surface characteristics from the disease statistics table, and sending the first warning message; If only the actual surface characteristics do not match the standard surface characteristics, then judge whether there is an abnormal fluctuation in the actual environmental data within the first time window; If so, obtain the detection points with fluctuations greater than the standard fluctuation curve; Screening similar environmental data and calculating the average environmental data; Screening the second predicted disease that matches the average environmental data and the actual surface characteristics from the disease statistics table, and sending the second warning message; If not, then screening the first predicted disease that matches the actual environmental data and the actual surface characteristics from the disease statistics table, and sending the first warning message; If only the actual environmental data do not match the standard environmental data, then judge whether the actual environmental data continues to not match within the second time window; If so, screening the third predicted disease that matches the actual environmental data from the disease statistics table, and sending the third warning message; If not, then screening the first predicted disease that matches the actual environmental data and the actual surface characteristics from the disease statistics table, and sending the first warning message.

2. The method for predicting grape diseases based on the Internet of Things according to claim 1, wherein, The steps of screening the first predicted disease that matches the actual environmental data and the actual surface characteristics from the disease statistics table include: determining the recognition framework according to the disease statistics table; Constructing the basic probability assignment according to the actual environmental data and the actual surface characteristics; Using the Dempster combination rule to combine the final combined probability; Calculating the belief function and the likelihood function according to the final combined probability; Obtaining the predicted disease type according to the belief function and the likelihood function.

3. The method for predicting grape diseases based on the Internet of Things according to claim 2, wherein, Before obtaining the predicted disease type according to the belief function and the likelihood function, it includes: Obtaining the difference between the belief function and the likelihood function; Judging whether the absolute value of the difference exceeds the difference threshold; If not, then obtaining the predicted disease type according to the belief function and the likelihood function.

4. A grape disease prediction method based on the Internet of Things according to claim 3, characterized in that, The judgment of whether the absolute value of the difference exceeds the difference threshold further includes: If so, defining the current detection point as an abnormal point, and increasing the acquisition frequency and acquisition density of the surface characteristics and environmental data corresponding to the current grape variety; If the number of abnormal points exceeds the first number threshold within the same detection time, then defining the detection time as an abnormal detection time; If the occurrence times of the abnormal detection time exceed the second number threshold within a detection period, the predicted disease type with the most occurrence times among all abnormal points is selected as the final predicted disease type.

5. The method for predicting grape diseases based on the Internet of Things according to claim 4, characterized in that, The steps of improving the acquisition frequency and acquisition density of the corresponding surface features and environmental data of the current grape variety include: Obtain the offset value δ between the absolute value of the difference and the difference threshold; Δ is the difference, Δ 阈 is the difference threshold; Determine the frequency coefficient α and the density coefficient β according to the offset value δ and the preset hierarchical adjustment rule; Obtain the reference sampling frequency f base and the reference sampling density d base ; N is the number of initially turned-on image acquisition devices, S is the area of the detection region, and K is the regional complexity coefficient; Input the reference sampling frequency f base , the frequency coefficient α, and the offset value δ into the sampling frequency adjustment model to obtain the adjusted new sampling frequency f new ; the sampling frequency adjustment model is According to the new sampling frequency f new , determine the acquisition interval; Input the reference sampling density d base , the density coefficient β, and the offset value δ into the sampling density adjustment model to obtain the adjusted new sampling density d new ; the sampling density adjustment model is γ is the area compensation coefficient, and S 临界 is the critical area, i.e., the area of the entire grape planting area; According to the new sampling density d new , obtain the number N of newly started image acquisition devices new ; N new = S·d new ; According to the area S of the detection region, obtain the total arrangement number N of the sensor groups bass ; m is the redundancy coefficient, ε is the number of standby groups for coping with sudden failures, and n base is the reference acquisition density of the sensor group; Based on the offset value δ and the total layout quantity N bass , the newly added number of opened groups ΔN is obtained, 6. An IoT-based grape disease prediction device, characterized in that, Implementing the Internet of Things-based grape disease prediction method according to any one of claims 1-5, including: An image acquisition module for acquiring the actual image information of grapes in the detection area; A data acquisition module for acquiring the actual environmental data of the grapes in the detection area; An image analysis module for determining the grape variety, growth stage, and actual surface features of the current grapes according to the actual image information; A screening and matching module for determining the standard surface features and standard environmental data corresponding to the growth stage according to the grape variety; A comparison and processing module for comparing the standard surface features with the actual surface features and the standard environmental data with the actual environmental data; if the actual surface features do not match the standard surface features and the actual environmental data do not match the standard environmental data, a disease statistics table matching the growth stage is retrieved; The screening and matching module is used to screen the first predicted disease that matches the actual environmental data and the actual surface features from the disease statistics table and send a first warning message.

7. A terminal, characterized in that, Including: A memory storing an Internet of Things-based grape disease prediction program; A processor for executing the program stored on the memory to implement the steps of the Internet of Things-based grape disease prediction method according to any one of claims 1-5.

8. A computer-readable storage medium, characterized in that, A computer program capable of being loaded and executed by a processor to implement the Internet of Things-based grape disease prediction method according to any one of claims 1-5.

Images