Sorghum pest and disease damage whole-period management system based on machine learning

Abstract

The invention discloses a sorghum disease and insect pest whole-period management system based on machine learning, and relates to the technical field of disease and insect pest observation management, the final whole-observation period time is determined by obtaining the observation influence coefficient of a target sorghum planting area, and the final sub-observation period time and the observation number are obtained by combining the initial preset observation sub-period number; and observing the disease and pest treatment effect of the target sorghum planting area according to the final full observation period time, the final sub-observation period time and the observation number of the target sorghum planting area, and taking corresponding management measures according to the observation result. The observation full cycle time and the corresponding sub-cycle time of the sorghum disease and insect pest treatment effect can be flexibly changed according to the actual condition and state of a sorghum planting area, it is ensured that the sorghum disease and insect pest treatment effect can be observed in time, timeliness of disease and insect pest control measures is ensured, and emergencies of disease and insect pests can be coped with in time; the loss risk of sorghum crops is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of pest and disease observation and management, and in particular to a sorghum pest and disease full-cycle management system based on machine learning. Background Art

[0002] Sorghum pests and diseases refer to diseases and pests caused by various pathogenic microorganisms (such as fungi, bacteria and viruses) and pests (such as aphids, borers, locusts, etc.) during the sorghum planting process; common sorghum diseases include rust, smut, sheath blight and leaf spot, which can cause poor plant growth, leaf discoloration, withering and even death, seriously affecting the yield and quality of sorghum; therefore, in order to prevent the occurrence of sorghum pests and diseases, in general, various types of data related to sorghum planting are collected in real time in the sorghum planting area, and a preset machine learning model is used to predict the possible occurrence of pests and diseases in the sorghum planting area in the future, and remind staff to take corresponding preventive measures for the sorghum planting area.

[0003] In order to better observe the effects of treatment measures for sorghum diseases and insect pests in sorghum-growing areas, staff will set up a full observation cycle and several sub-cycles. According to the treatment effects of sorghum diseases and insect pests in the observed sorghum-growing areas, corresponding management measures will be taken in a timely manner to achieve the purpose of effectively controlling diseases and insect pests, thereby ensuring the healthy growth of sorghum and stable yield;

[0004] However, in actual situations, the conditions of sorghum planting areas are different. If the full-cycle observation time and corresponding sub-cycle time of the sorghum pest and disease treatment effect cannot be flexibly changed according to the actual situation and status, the pest and disease control measures may be delayed and unable to respond to sudden situations of pests and diseases in time, thereby increasing the risk of loss of sorghum crops. Summary of the invention

[0005] The purpose of the present invention is to solve the problem that pest and disease control measures are lagging behind and cannot respond to sudden pest and disease outbreaks in a timely manner, thereby increasing the risk of loss of sorghum crops, and proposes a sorghum pest and disease full-cycle management system based on machine learning.

[0006] The present invention proposes a full-cycle management system for sorghum pests and diseases based on machine learning, the system comprising:

[0007] Prevention module: Obtain sorghum planting data in the target sorghum planting area, and predict the type and severity of pests and diseases in the target sorghum planting area through a preset pest and disease prediction model, and collect corresponding treatment measures according to the type and severity of pests and diseases;

[0008] Regional area module: obtain the planting area and planting spacing of sorghum in the target sorghum planting area, and obtain the regional area coefficient based on the planting area and planting spacing;

[0009] Soil temperature and humidity module: obtains the temperature and humidity of the soil in the target sorghum planting area, and obtains the soil temperature and humidity deviation coefficient based on the soil temperature and humidity;

[0010] Weather impact module: obtains the light intensity of the target sorghum planting area and obtains the weather impact coefficient based on the light intensity;

[0011] Impact module: Determine the observation impact coefficient of the target sorghum planting area based on the regional area coefficient, soil temperature and humidity deviation coefficient and weather impact coefficient;

[0012] Calculation module: obtain the final full observation cycle time according to the observation influence coefficient of the target sorghum planting area and the initially preset full observation cycle time, and obtain the final time of each sub-observation cycle and the number of observations according to the observation influence coefficient, the initially preset number of observation sub-cycles, and the final full observation cycle time;

[0013] Observation management module: observe the treatment effect of pests and diseases in the target sorghum planting area according to the final full observation cycle time, the final observation cycle time of each sub-observation cycle and the number of observations, and take corresponding management measures according to the observation results.

[0014] Optionally, the regional area coefficient obtained according to the planting area and the planting spacing includes:

[0015] Obtain the planting area of ​​the target sorghum planting area and the total area of ​​the target sorghum planting area, and calculate the planting area ratio. The calculation formula is: DF = Ds / Dx, where DF is the planting area ratio, Ds is the planting area, and Dx is the total area of ​​the sorghum planting area;

[0016] The average horizontal spacing, average vertical spacing and total number of sorghum plants planted were obtained, and the effective planting density SW was calculated using the formula: SW = Dq / Ds, where Dq is the total number of plants planted;

[0017] The effective distance Qs between the planted sorghum crops was calculated using the following formula: Qs = Fg × Fb, where Fg and Fb are the average horizontal spacing and average vertical spacing, respectively;

[0018] Calculate the distance coefficient. The calculation formula is: In the formula, Cx is the distance coefficient, α is the preset soil nutrition influencing factor, and its value range is 0-1; β is the preset plant resistance influencing factor, and its value range is 0-1; Qz is the preset ideal planting spacing;

[0019] Calculate the regional area coefficient using the following formula: Dxc = a1 × DF + a2 × Cx, where Dxc is the regional area coefficient, and a1 and a2 are the preset weight coefficients for the planting area ratio and the distance coefficient, respectively.

[0020] Optionally, the soil temperature and humidity deviation coefficient is obtained according to the soil temperature and humidity, including:

[0021] Continuously obtain the soil temperature T of the target sorghum planting area i and soil moisture H i , get n data points, and record them as [T1, T2, ..., T n ] and [H1, H2, ..., H n ];

[0022] Calculate the average soil temperature and moisture using the formula:

[0023] Calculate the standard deviations σT and σH of temperature and humidity using the following formula:

[0024] The average soil temperature is compared with the preset average soil temperature Tr, and the deviation difference gh between the two is calculated. The calculation formula is:

[0025] The average soil moisture value is compared with the preset average soil moisture value Hr, and the deviation difference Kh between the two is calculated. The calculation formula is:

[0026] The standard deviation of the soil temperature is compared with the preset standard deviation of the soil temperature Tm, and the deviation difference sc between the two is calculated. The calculation formula is: sc = σT-Tm / Tm;

[0027] The standard deviation of soil moisture is compared with the preset standard deviation of soil moisture Hm, and the deviation difference bn between the two is calculated. The calculation formula is bn = σH-Hm / Hm;

[0028] Calculate the soil temperature and humidity deviation coefficient Bzc, the calculation formula is: Bzc = b1 × (gh + sc) + b2 × (Kh + bn), where b1 and b2 are preset weight coefficients.

[0029] Optionally, the weather influence coefficient obtained according to the light intensity includes:

[0030] Obtain the light intensity at different times in the target sorghum planting area within the preset time period, and mark the light intensity at each time as E a, a represents the order number of the light intensity at different times, a=1, 2, 3, 4, ..., v, v is a positive integer;

[0031] According to the light intensity E at different times a and the preset light intensity W at the corresponding moment a Get the weather influence coefficient Box, the calculation formula is:

[0032] Optionally, determining the observation influence coefficient of the target sorghum planting area according to the regional area coefficient, the soil temperature and humidity deviation coefficient and the weather influence coefficient includes:

[0033] The regional area coefficient, soil temperature and humidity deviation coefficient and weather influence coefficient are normalized, and the observation influence coefficient is calculated based on the normalized regional area coefficient, soil temperature and humidity deviation coefficient and weather influence coefficient. The calculation formula is:

[0034]

[0035] Where Qer is the observation influence coefficient, Dxc, Bzc, Box are the normalized regional area coefficient, soil temperature and humidity deviation coefficient, and weather influence coefficient, respectively, c1, c2, c3 are the preset proportional coefficients of the normalized regional area coefficient, soil temperature and humidity deviation coefficient, and weather influence coefficient, respectively, and c1, c2, c3 are all greater than 0.

[0036] Optionally, obtaining the final full observation cycle time according to the observation influence coefficient of the target sorghum planting area and the initially preset full observation cycle time includes:

[0037] FR=FG×(1+Qer)

[0038] Where FR is the final full observation cycle time, and FG is the initial preset full observation cycle time.

[0039] Optionally, obtaining the final time of each sub-observation cycle and the number of observations according to the observation influence coefficient, the number of initially preset observation sub-cycles, and the final full observation cycle time includes:

[0040] WS=WD×(1+3Qer)

[0041] YH=FR / WS

[0042] Where WS is the final number of sub-observation cycles, WD is the initial preset number of observation sub-cycles, and YH is the final time of each sub-observation cycle.

[0043] Beneficial effects of the present invention:

[0044] The present invention proposes a full-cycle management system for sorghum pests and diseases based on machine learning, which obtains the observation influence coefficient of the target sorghum planting area, and obtains the final full observation cycle time in combination with the initially preset full observation cycle time, and obtains the final time of each sub-observation cycle and the number of observations in combination with the initially preset number of observation sub-cycles and the final full observation cycle time; and observes the treatment effect of the pests and diseases in the target sorghum planting area according to the final full observation cycle time, the final time of each sub-observation cycle and the number of observations, and takes corresponding management measures according to the observation results. In this way, in actual situations, the full observation cycle time and the corresponding sub-cycle time of the treatment effect of sorghum pests and diseases can be flexibly changed according to the actual situation and status of the sorghum planting area, so as to ensure that the treatment effect of sorghum pests and diseases can be observed in time, ensure the timeliness of pest and disease control measures, and be able to respond to sudden situations of pests and diseases in time, thereby reducing the risk of loss of sorghum crops. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] The present invention will be further described below in conjunction with the accompanying drawings.

[0046] Figure 1 This is a framework diagram of the full-cycle management system for sorghum pests and diseases based on machine learning. DETAILED DESCRIPTION

[0047] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0048] Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in the field without making any creative work shall fall within the scope of protection of the present invention.

[0049] The embodiment of the present invention provides a full-cycle management system for sorghum pests and diseases based on machine learning. Figure 1 , Figure 1 A framework diagram of a full-cycle management system for sorghum pests and diseases based on machine learning provided in an embodiment of the present invention, the system comprising:

[0050] Prevention module: Obtain sorghum planting data in the target sorghum planting area, and predict the type and severity of pests and diseases in the target sorghum planting area through a preset pest and disease prediction model, and collect corresponding treatment measures according to the type and severity of pests and diseases;

[0051] Regional area module: obtain the planting area and planting spacing of sorghum in the target sorghum planting area, and obtain the regional area coefficient based on the planting area and planting spacing;

[0052] Soil temperature and humidity module: obtains the temperature and humidity of the soil in the target sorghum planting area, and obtains the soil temperature and humidity deviation coefficient based on the soil temperature and humidity;

[0053] Weather impact module: obtains the light intensity of the target sorghum planting area and obtains the weather impact coefficient based on the light intensity;

[0054] Impact module: Determine the observation impact coefficient of the target sorghum planting area based on the regional area coefficient, soil temperature and humidity deviation coefficient and weather impact coefficient;

[0055] Calculation module: obtain the final full observation cycle time according to the observation influence coefficient of the target sorghum planting area and the initially preset full observation cycle time, and obtain the final time of each sub-observation cycle and the number of observations according to the observation influence coefficient, the initially preset number of observation sub-cycles, and the final full observation cycle time;

[0056] Observation management module: observe the treatment effect of pests and diseases in the target sorghum planting area according to the final full observation cycle time, the final observation cycle time of each sub-observation cycle and the number of observations, and take corresponding management measures according to the observation results.

[0057] Based on the machine learning-based sorghum pest and disease full-cycle management system provided by the embodiment of the present invention, through the above-mentioned method, in actual situations, the full-cycle time and corresponding sub-cycle time for observing the treatment effect of sorghum pests and diseases can be flexibly changed according to the actual situation and status of the sorghum planting area, ensuring that the treatment effect of sorghum pests and diseases can be observed in time, ensuring the timeliness of pest and disease prevention and control measures, and being able to respond to sudden situations of pests and diseases in time, reducing the risk of loss of sorghum crops.

[0058] In one embodiment, sorghum planting data of a target sorghum planting area is obtained, and the type and severity level of the pests and diseases in the target sorghum planting area are predicted by a preset pest and disease prediction model, and corresponding treatment measures are collected according to the type and severity level of the pests and diseases;

[0059] It should be noted that the sorghum planting data in the target sorghum planting area refers to various types of data related to sorghum planting, such as leaf color, plant height, etc. The specific data are set by professionals based on the actual situation and are not limited or elaborated on. In addition, the preset machine learning model refers to an algorithm model used to analyze and predict future pest and disease risks in sorghum planting areas; these models can be trained based on historical data and environmental characteristics to identify potential pest and disease risks; the specific training method and model determination are based on the actual situation and are not limited or elaborated on.

[0060] The type and severity of pests and diseases in the target sorghum planting area are predicted by the preset pest and disease prediction model, and the corresponding treatment measures are collected according to the type and severity of pests and diseases. For example, if the model predicts that the target sorghum planting area may be infested by diamondback moth and its severity level is assessed to be high, the system will automatically generate a series of corresponding treatment measures. For high-level pests, the treatment measures may include the immediate application of specific pesticides for diamondback moth to quickly reduce the pest density and reduce its damage to sorghum. At the same time, the system will also recommend field inspections to ensure the coverage and effectiveness of pesticide application and monitor the actual changes in pests; in addition, for low-level pests such as aphids, the treatment measures may include strengthening physical control measures, such as setting up yellow board traps and spraying biological pesticides to reduce the impact on the environment and ensure that the growth of sorghum is not disturbed. Through these targeted management measures, the system can not only effectively respond to the threat of pests and diseases, but also provide farmers with scientific decision-making basis, optimize resource allocation, and achieve sustainable agricultural production. Ultimately, through continuous data feedback and observation, the system will also continuously adjust and optimize pest and disease control strategies to respond to possible changes and ensure the health of sorghum planting and a good harvest.

[0061] In one embodiment, obtaining the regional area coefficient according to the planting area and the planting spacing includes:

[0062] Obtain the planting area of ​​the target sorghum planting area and the total area of ​​the target sorghum planting area, and calculate the planting area ratio. The calculation formula is: DF = Ds / Dx, where DF is the planting area ratio, Ds is the planting area, and Dx is the total area of ​​the sorghum planting area;

[0063] The average horizontal spacing, average vertical spacing and total number of sorghum plants planted were obtained, and the effective planting density SW was calculated using the formula: SW = Dq / Ds, where Dq is the total number of plants planted;

[0064] The effective distance Qs between the planted sorghum crops was calculated using the following formula: Qs = Fg × Fb, where Fg and Fb are the average horizontal spacing and average vertical spacing, respectively;

[0065] Calculate the distance coefficient using the following formula: In the formula, Cx is the distance coefficient, α is the preset soil nutrition influencing factor, and its value range is 0-1; β is the preset plant resistance influencing factor, and its value range is 0-1; Qz is the preset ideal planting spacing;

[0066] Calculate the regional area coefficient using the following formula: Dxc = a1 × DF + a2 × Cx, where Dxc is the regional area coefficient, and a1 and a2 are the preset weight coefficients for the planting area ratio and the distance coefficient, respectively.

[0067] It should be noted that a1 and a2 are set by professionals according to actual conditions. Generally, the sum of a1 and a2 is 1. For example, a1 and a2 can be 0.5 and 0.5 respectively, or other numbers, which are not specifically limited.

[0068] It should be noted that the preset soil nutrition influencing factor, β is the preset plant resistance influencing factor, and Qz is the preset ideal planting spacing, which are all set by professionals based on actual conditions and are not limited or elaborated on. In addition, the planting area and total area data of the target sorghum planting area can be obtained through plot surveys and farmland management systems, and the horizontal and vertical spacing of planted crops can be obtained by remote sensing technology or field measurements. Other acquisition methods may also be used, and are not limited or elaborated on.

[0069] It should be noted that when the planting area of ​​the target sorghum planting area is larger and the distance coefficient is larger, that is, the area coefficient is larger, it means that the corresponding target sorghum planting area, after the pest prevention treatment, the corresponding subsequent observation full cycle should be increased, and the number of observation sub-cycles should be increased, and the time of each sub-cycle should be less than the previously preset sub-cycle observation time. The reason is that a larger planting area and distance coefficient usually means an increase in the risk of pests and diseases, which may lead to a faster spread of pests, thus requiring more frequent monitoring to timely evaluate the treatment effect. In addition, a larger planting area may make the impact of pests and diseases more dispersed, making it difficult to capture the comprehensive treatment effect through a single observation cycle. Therefore, appropriately extending the observation cycle and increasing the number of sub-cycles can ensure that the effectiveness of prevention and control measures is fully evaluated under different growth stages and environmental conditions, and at the same time, the response strategy can be adjusted more flexibly to ensure the healthy growth and stable yield of sorghum crops. At the same time, shortening the observation time of the sub-cycle can improve the response speed, timely discover potential problems, and quickly adjust management measures, thereby minimizing the impact of pests on sorghum yield and quality.

[0070] In one embodiment, obtaining the soil temperature and humidity deviation coefficient according to the soil temperature and humidity includes:

[0071] Continuously obtain the soil temperature T of the target sorghum planting area i and soil moisture Hi , get n data points, and record them as [T1, T2, ..., T n ] and [H1, H2, ..., H n ];

[0072] Calculate the average soil temperature and moisture using the formula:

[0073] Calculate the standard deviations σT and σH of temperature and humidity using the following formula:

[0074] The average soil temperature is compared with the preset average soil temperature Tr, and the deviation difference gh between the two is calculated. The calculation formula is:

[0075] The average soil moisture value is compared with the preset average soil moisture value Hr, and the deviation difference Kh between the two is calculated. The calculation formula is:

[0076] The standard deviation of the soil temperature is compared with the preset standard deviation of the soil temperature Tm, and the deviation difference sc between the two is calculated. The calculation formula is: sc = σT-Tm / Tm;

[0077] The standard deviation of soil moisture is compared with the preset standard deviation of soil moisture Hm, and the deviation difference bn between the two is calculated. The calculation formula is bn = σH-Hm / Hm;

[0078] Calculate the soil temperature and humidity deviation coefficient Bzc, the calculation formula is: Bzc = b1 × (gh + sc) + b2 × (Kh + bn), where b1 and b2 are preset weight coefficients.

[0079] It should be noted that b1 and b2 are set by professionals according to actual conditions. Generally, the sum of b1 and b2 is 1. For example, b1 and b2 can be 0.8, 0.3, or other numbers, without specific limitation.

[0080] It should be noted that the soil temperature and soil moisture of the target sorghum planting area can be directly obtained through the temperature sensor and humidity sensor of the target sorghum planting area, or other acquisition methods can be used. The specific acquisition method is set by professional staff according to the actual situation, and no specific limitation or elaboration is made; the preset average value of soil moisture, the preset standard deviation of soil moisture, the preset average value of soil temperature, and the preset standard deviation of soil temperature are all set by professional staff according to the actual situation, and no specific limitation or elaboration is made.

[0081] It should be noted that when the soil temperature and humidity deviation coefficient is larger, it means that the corresponding target sorghum planting area, after the pest prevention treatment, the corresponding subsequent observation full cycle should be increased, and the number of observation sub-cycles needs to be increased, and the time of each sub-cycle should be less than the previously preset sub-cycle observation time. The reason is that when the soil temperature and humidity deviation coefficient is large, it means that the soil conditions in the target sorghum planting area are relatively unstable, and the soil temperature and humidity fluctuate greatly. This instability may affect the health of sorghum plants, making them more susceptible to pests and diseases or affecting the effect of control agents. A large deviation coefficient means that the growth environment of the sorghum root system changes rapidly, especially when the temperature and humidity regulation effect is not ideal, which may lead to uneven absorption of the agent, thereby affecting the overall effect of pest treatment. Therefore, after pest prevention treatment, in order to ensure that new pests and diseases can be discovered and dealt with in time, it is very important to extend the observation time of the full cycle. By extending the observation period, the pest and disease dynamics of sorghum plants throughout the growth stage can be monitored in more detail to ensure that any subtle recurrence of pests and diseases or new symptoms can be discovered at the first time. At the same time, increasing the number of observation sub-cycles and shortening the observation time of each sub-cycle can make monitoring more frequent and detailed. This can timely capture the impact of temperature and humidity fluctuations on the growth of sorghum crops, especially when there are drastic changes in a short period of time, so that remedial measures can be taken quickly to ensure that the pest control effect is not weakened. In addition, frequent and detailed observations can also adjust the prevention and control strategies more accurately. If a large deviation in temperature and humidity is found in a sub-cycle, it means that the environmental fluctuations have had an impact, and the prevention and management measures can be adjusted immediately according to the actual situation, such as increasing irrigation, adjusting the frequency of fertilization, or repeating the spraying of control agents when appropriate. In this way, the potential risk of pests can be reduced while ensuring the healthy growth of sorghum, thereby optimizing the whole cycle management system of sorghum pests and diseases.

[0082] In one embodiment, obtaining the weather influence coefficient according to the light intensity includes:

[0083] Obtain the light intensity at different times in the target sorghum planting area within the preset time period, and mark the light intensity at each time as E a , a represents the order number of the light intensity at different times, a=1, 2, 3, 4, ..., v, v is a positive integer;

[0084] According to the light intensity E at different times a and the preset light intensity W at the corresponding moment a Get the weather influence coefficient Box, the calculation formula is:

[0085] It should be noted that the light intensity of the target sorghum planting area can be directly obtained through the light sensor of the target sorghum planting area, or other acquisition methods can be used. The specific acquisition method is set by professional staff according to the actual situation, and no specific limitation or elaboration is made; the preset light intensity and preset time period are set by professional staff according to the actual situation, and no specific limitation or elaboration is made.

[0086] It should be noted that when the weather influence coefficient is larger, that is, the greater the deviation of the light intensity of the target sorghum planting area from the preset light intensity, it means that the corresponding target sorghum planting area, after the pest prevention treatment, the corresponding subsequent observation full cycle should be increased, and the number of observation sub-cycles should be increased, and the time of each sub-cycle should be less than the previous preset sub-cycle observation time. The reason is that when the light intensity of the target sorghum planting area deviates from the preset light intensity, it means that the light conditions in the area are unstable, which may be too strong or too weak, thus affecting the growth of sorghum and its resistance to pests and diseases. Insufficient light will lead to weakened photosynthesis of sorghum, reduced nutrient accumulation of plants, reduced growth vitality of plants, weakened ability to resist pests and diseases, and make pests more likely to occur; on the contrary, excessive light may cause photoinhibition in plants, causing leaf burns, dryness, etc., thereby providing conditions for the invasion of pests and diseases. In this case, although pest prevention treatment has been carried out, due to the large changes in environmental light, there may be new risks of pests and diseases in the future. Therefore, it is necessary to extend the overall observation cycle to ensure that any possible re-emergence or mutation of pest risks can be captured; at the same time, increasing the number of observation sub-cycles and shortening the time of each sub-cycle can monitor plant health and pest conditions more frequently. In this way, any abnormalities can be detected in time at an early stage, so that countermeasures can be taken quickly to ensure the effectiveness of pest and disease control and the healthy growth of sorghum crops. This preventive intensive monitoring strategy helps reduce the risk of secondary pests caused by environmental fluctuations and provides better protection for the overall health and harvest of sorghum.

[0087] In one embodiment, determining the observation influence coefficient of the target sorghum planting area according to the regional area coefficient, the soil temperature and humidity deviation coefficient, and the weather influence coefficient includes:

[0088] The regional area coefficient, soil temperature and humidity deviation coefficient and weather influence coefficient are normalized, and the observation influence coefficient is calculated based on the normalized regional area coefficient, soil temperature and humidity deviation coefficient and weather influence coefficient. The calculation formula is:

[0089]

[0090] Where Qer is the observation influence coefficient, Dxc, Bzc, Box are the normalized regional area coefficient, soil temperature and humidity deviation coefficient, and weather influence coefficient, respectively, c1, c2, c3 are the preset proportional coefficients of the normalized regional area coefficient, soil temperature and humidity deviation coefficient, and weather influence coefficient, respectively, and c1, c2, c3 are all greater than 0.

[0091] It should be noted that c1, c2, and c3 are set by professionals according to actual conditions. Generally, the sum of c1, c2, and c3 is 1. For example, c1, c2, and c3 can be 0.4, 0.3, and 0.3, respectively, or other numbers without specific limitation.

[0092] In one embodiment, obtaining the final full observation cycle time according to the observation influence coefficient of the target sorghum planting area and the initially preset full observation cycle time includes:

[0093] FR=FG×(1+Qer)

[0094] Where FR is the final full observation cycle time, and FG is the initial preset full observation cycle time.

[0095] In one embodiment, obtaining the final time of each sub-observation cycle and the number of observations according to the observation influence coefficient, the number of initially preset observation sub-cycles, and the final full observation cycle time includes:

[0096] WS=WD×(1+3Qer)

[0097] YH==FR / QS

[0098] Where WS is the final number of sub-observation cycles, WD is the initial preset number of observation sub-cycles, and YH is the final time of each sub-observation cycle.

[0099] It should be noted that the initial preset full observation cycle time and the initial preset number of observation sub-cycles are set by professional staff according to actual conditions, and no specific restrictions or elaborations are made.

[0100] In one implementation method, through the above method, the treatment effect of pests and diseases in the target sorghum planting area is observed according to the final full observation cycle time, the final sub-observation cycle time and the number of observations in the target sorghum planting area, and corresponding management measures are taken according to the observation results. In this way, in actual situations, the full observation cycle time and the corresponding sub-cycle time of the treatment effect of sorghum pests and diseases can be flexibly changed according to the actual situation and status of the sorghum planting area, ensuring that the treatment effect of sorghum pests and diseases can be observed in time, ensuring the timeliness of pest and disease prevention and control measures, and being able to respond to sudden situations of pests and diseases in time to reduce the risk of loss of sorghum crops.

[0101] The above is a detailed description of an embodiment of the present invention, but the content is only a preferred embodiment of the present invention and cannot be considered to limit the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. A full-cycle management system for sorghum pests and diseases based on machine learning, characterized in that: include: Prevention module: Obtain sorghum planting data in the target sorghum planting area, and predict the type and severity of pests and diseases in the target sorghum planting area through a preset pest and disease prediction model, and collect corresponding treatment measures according to the type and severity of pests and diseases; Regional area module: obtain the planting area and planting spacing of sorghum in the target sorghum planting area, and obtain the regional area coefficient based on the planting area and planting spacing; Soil temperature and humidity module: obtains the temperature and humidity of the soil in the target sorghum planting area, and obtains the soil temperature and humidity deviation coefficient based on the soil temperature and humidity; Weather impact module: obtains the light intensity of the target sorghum planting area and obtains the weather impact coefficient based on the light intensity; Impact module: Determine the observation impact coefficient of the target sorghum planting area based on the regional area coefficient, soil temperature and humidity deviation coefficient and weather impact coefficient; Calculation module: obtain the final full observation cycle time according to the observation influence coefficient of the target sorghum planting area and the initially preset full observation cycle time, and obtain the final time of each sub-observation cycle and the number of observations according to the observation influence coefficient, the initially preset number of observation sub-cycles, and the final full observation cycle time; Observation management module: observe the treatment effect of pests and diseases in the target sorghum planting area according to the final full observation cycle time, the final observation cycle time of each sub-observation cycle and the number of observations, and take corresponding management measures according to the observation results.

2. The full-cycle management system for sorghum pests and diseases based on machine learning according to claim 1 is characterized in that: The regional area coefficients obtained based on the planting area and planting spacing include: Obtain the planting area of ​​the target sorghum planting area and the total area of ​​the target sorghum planting area, and calculate the planting area ratio. The calculation formula is: DF = Ds / Dx, where DF is the planting area ratio, Ds is the planting area, and Dx is the total area of ​​the sorghum planting area; The average horizontal spacing, average vertical spacing and total number of sorghum plants planted were obtained, and the effective planting density SW was calculated using the formula: SW = Dq / Ds, where Dq is the total number of plants planted; The effective distance Qs between the planted sorghum crops was calculated using the following formula: Qs = Fg × Fb, where Fg and Fb are the average horizontal spacing and average vertical spacing, respectively; Calculate the distance coefficient. The calculation formula is: In the formula, Cx is the distance coefficient, α is the preset soil nutrition influencing factor, and its value range is 0-1; β is the preset plant resistance influencing factor, and its value range is 0-1; Qz is the preset ideal planting spacing; Calculate the regional area coefficient using the following formula: Dxc = a1 × DF + a2 × Cx, where Dxc is the regional area coefficient, and a1 and a2 are the preset weight coefficients for the planting area ratio and the distance coefficient, respectively.

3. The full-cycle management system for sorghum pests and diseases based on machine learning according to claim 1 is characterized in that: According to the temperature and humidity of the soil, the soil temperature and humidity deviation coefficients include: Continuously obtain the soil temperature T of the target sorghum planting area i and soil moisture H i , get n data points, and record them as [T1,T2,…,T n ] and [H1,H2,…,H n ]; Calculate the average soil temperature and moisture using the formula: Calculate the standard deviations σT and σH of temperature and humidity using the following formula: The average soil temperature is compared with the preset average soil temperature Tr, and the deviation difference gh between the two is calculated. The calculation formula is: The average soil moisture value is compared with the preset average soil moisture value Hr, and the deviation difference Kh between the two is calculated. The calculation formula is: The standard deviation of the soil temperature is compared with the preset standard deviation of the soil temperature Tm, and the deviation difference sc between the two is calculated. The calculation formula is: sc = σT-Tm / Tm; The standard deviation of soil moisture is compared with the preset standard deviation of soil moisture Hm, and the deviation difference bn between the two is calculated. The calculation formula is bn = σH-Hm / Hm; Calculate the soil temperature and humidity deviation coefficient Bzc, the calculation formula is: Bzc = b1 × (gh + sc) + b2 × (Kh + bn), where b1 and b2 are preset weight coefficients.

4. The full-cycle management system for sorghum pests and diseases based on machine learning according to claim 1 is characterized in that: The weather influence coefficients obtained based on light intensity include: Obtain the light intensity at different times in the target sorghum planting area within the preset time period, and mark the light intensity at each time as E a , a represents the order number of the light intensity at different times, a=1, 2, 3, 4, ..., v, v is a positive integer; According to the light intensity E at different times a and the preset light intensity W at the corresponding moment a Get the weather influence coefficient Box, the calculation formula is:

5. The full-cycle management system for sorghum pests and diseases based on machine learning according to claim 1 is characterized in that: The observation influence coefficients for the target sorghum planting area are determined based on the regional area coefficient, soil temperature and humidity deviation coefficient, and weather influence coefficient. The regional area coefficient, soil temperature and humidity deviation coefficient and weather influence coefficient are normalized, and the observation influence coefficient is calculated based on the normalized regional area coefficient, soil temperature and humidity deviation coefficient and weather influence coefficient. The calculation formula is: Where Qer is the observation influence coefficient, Dxc, Bzc, Box are the normalized regional area coefficient, soil temperature and humidity deviation coefficient, and weather influence coefficient, respectively, c1, c2, c3 are the preset proportional coefficients of the normalized regional area coefficient, soil temperature and humidity deviation coefficient, and weather influence coefficient, respectively, and c1, c2, c3 are all greater than 0.

6. The full-cycle management system for sorghum pests and diseases based on machine learning according to claim 5 is characterized in that: The final full observation cycle time obtained based on the observation influence coefficient of the target sorghum planting area and the initially preset full observation cycle time includes: FR=FG×(1+Qer) Where FR is the final full observation cycle time, and FG is the initial preset full observation cycle time.

7. The full-cycle management system for sorghum pests and diseases based on machine learning according to claim 6 is characterized in that: According to the observation influence coefficient, the number of initially preset observation sub-cycles, and the final full observation cycle time, the final observation sub-cycle time and observation number are obtained, including: WS=WD×(1+3Qer) YH=FR / WS Where WS is the final number of sub-observation cycles, WD is the initial preset number of observation sub-cycles, and YH is the final time of each sub-observation cycle.

Images