Intelligent strawberry production management method and system

Abstract

The present invention discloses an intelligent strawberry production management method, which relates to the technical field of strawberry planting. The strawberry production management steps are as follows: S1, select fertile soil with strong water and fertilizer retention capacity, detect soil pH value, detect soil pH value, and construct a strawberry garden; S2, select suitable strawberry varieties, sow and plant the strawberry varieties, and breed them by stolon propagation; S3, install environmental sensors in the strawberry garden, monitor environmental data in the strawberry garden in real time, and upload the data to a management module. The sensors include temperature sensors, moisture sensors, and light sensors. The present invention also discloses an intelligent strawberry production management system. The present invention improves monitoring efficiency and accuracy by timely discovering pest conditions at the early stage of the disease, can take measures before the disease spreads on a large scale, reduce the amount of pesticides used, reduce prevention and control costs, ensure strawberry quality and safety, and intelligently manage strawberry growth.

Description

Technical Field

[0001] The invention relates to the technical field of strawberry planting, and in particular to an intelligent strawberry production management method and system. Background Art

[0002] As a fruit that is deeply loved by consumers, strawberry is widely planted around the world. The traditional strawberry production management model mainly relies on the experience and manual operation of the growers, which has many limitations. Growers can often only rely on subjective judgment and limited field observations to estimate whether the environmental factors such as soil moisture, temperature, and light intensity are suitable for strawberry growth. The judgment of soil moisture usually relies on hand feel or simple observation of the soil surface. This method is difficult to accurately determine the moisture content deep in the soil and the moisture difference in different regions, which can easily lead to over-watering or under-watering. Excessive moisture may cause root diseases, such as root rot, due to soil waterlogging. The roots will be deprived of oxygen and then infected by harmful microorganisms; too little water will stunt the growth of strawberry plants, wilt leaves, affect photosynthesis and fruit development, and ultimately lead to reduced yield and quality. In terms of pest and disease control, traditional pest and disease monitoring mainly relies on regular manual inspections, which makes it difficult to detect early signs of pests and diseases in a timely manner. Once pests and diseases break out on a large scale, a large amount of chemical pesticides are often required for emergency control, which not only increases the risk of pesticide residues and endangers consumer health, but may also damage the ecological environment and affect the beneficial biological communities in the strawberry garden. In this regard, we propose an intelligent strawberry production management method and system. Summary of the invention

[0003] In order to solve the above technical problems, an intelligent strawberry production management method and system are provided. This technical solution solves the limitations of the above manual detection and the problem of difficulty in timely detection of pest conditions.

[0004] To achieve the above purpose, the technical solution adopted by the present invention is: an intelligent strawberry production management method, the strawberry production management steps are: S1. Select fertile soil with strong water and fertilizer retention capacity, test the soil pH value, test the soil pH value, and construct a strawberry garden; S2. Select suitable strawberry varieties, sow and plant the strawberry varieties, and use runner propagation to breed them; S3, installing environmental sensors in the strawberry garden, monitoring the environmental data in the strawberry garden in real time, and uploading the data to the management module. The sensors include temperature sensors, moisture sensors and light sensors; S4, water irrigation, after the water sensor detects that the soil moisture is lower than the threshold of normal growth, intelligent water irrigation is carried out to ensure the normal growth of strawberries; intelligent management and control of the strawberry garden is carried out based on temperature sensors and light sensors; S5. Regularly use drones to take real-time photos of strawberry leaves in the strawberry garden, obtain leaf image data, upload the image data, analyze the leaf image data based on the image analysis module, determine whether the leaves have diseases based on the machine learning algorithm, establish a hazard threshold, and send an early warning message to the management module when the leaf disease reaches the threshold during monitoring, so that the management personnel can take preventive measures; S6. Regularly observe the strawberry garden to check the growth of strawberries and the fullness of the fruits, apply nitrogen fertilizer and potassium fertilizer to improve the yield and quality of strawberries, compile and upload the observation results to the suggestion generation module, and give subsequent intelligent management suggestions based on the current growth of strawberries.

[0005] Preferably, in step S1, the soil is selected in advance by the planting personnel observing the color of the soil surface, selecting soil with dark color, observing the activity of the soil, and judging by checking the number of earthworms in the soil. If there are many earthworms in the soil, the surface soil structure is excellent, and the air permeability and fertility are high. The soil pH value is detected by obtaining local soil samples and sending them to the laboratory for testing of pH and soil nutrients. If the pH value is detected to be too high or too low, the soil is adjusted by adding lime, and sulfur powder and acidic fertilizers are used to improve the soil pH.

[0006] Preferably, in step S2, a seedbed for sowing is prepared, the soil of the seedbed is deeply plowed to a depth of 20 cm-30 cm, and the pre-screened healthy and plump strawberry seeds are evenly sown on the seedbed. After sowing, a layer of 0.5 cm-1 cm of fine soil is covered, and water is sprayed with a fine mist nozzle, and the seeds are waited for germination. During the seed germination process, the temperature is controlled between 15° C. and 25° C.

[0007] Preferably, in step S2, the stolon propagation method is carried out by tidying up the mother plant, cutting off the sick, aging and poorly growing branches and leaves, and when the stolon grows and extends, guiding the stolon to extend toward the predetermined propagation area, using small bamboo sticks to fix the stolon on loose, moist and fertile soil, and fertilizing every 7d-10d.

[0008] Preferably, in step S3, temperature sensors are installed between plant rows, on the ground surface and at different heights in the strawberry garden, with an installation interval of 5m-10m between plant rows to capture the air temperature around the strawberry plants; temperature sensors are installed on the ground surface to capture changes in soil surface temperature; temperature sensors are installed at different heights in the strawberry garden to monitor temperature changes in the garden; moisture sensors are installed in an inserted manner to monitor changes in soil moisture content in different areas and obtain soil moisture data in real time; light sensors are installed on the top of the strawberry garden to collect light data of the strawberry garden; and the collected data is uploaded.

[0009] Preferably, in step S4, based on the expert analysis method, soil moisture content threshold, temperature threshold and light threshold suitable for strawberry growth are respectively established; the real-time collected data are compared with the thresholds to judge whether the current strawberry garden environment is suitable for normal strawberry growth; if the real-time collected data is lower than the threshold, it is judged that the current strawberry garden environment is not suitable for normal strawberry growth, and the irrigation water is controlled, the temperature change is adjusted and the light conditions are adjusted accordingly; the expert analysis method selects strawberry growth experts to collect existing strawberry planting record data, and obtains the threshold through expert analysis based on historical data.

[0010] Preferably, in step S5, drones are used to take photos regularly, and the photos of the leaves are randomly sampled and uploaded. The image analysis module preprocesses the acquired image data, including image grayscale processing, image filtering processing and image enhancement processing. The image grayscale is processed based on the component method. The image filtering processing adopts median filtering processing, sorts the pixel values ​​in the filter window, and takes the middle value as the value of the center pixel. The image enhancement processing is based on histogram equalization processing, and the histogram of the original image is stretched to make the distribution of pixel grayscale values ​​uniform and increase the contrast of the image; the characteristics of insect pests in the leaves are extracted, and the color, texture and wormhole characteristics of the leaves are extracted.

[0011] Preferably, the machine learning algorithm calculates the color feature vector extracted from the leaf image as C=(c1, c2, ..., c m ), where c i Represents the specific color feature value, and the texture feature vector is T=(t1, t2, ... t n ), t j Expressed as texture features, the wormhole feature vector is H = (h1, h2, ... h k ), h l The wormhole quantity feature is obtained by merging all feature vectors into a total feature vector X = (C, T, H) = (x1, x2, ..., x m+n+k ), let the weights be w=(w1,w2,…,w m+n+k ), the bias term is b, and the predicted value y of the disease degree is calculated using the following formula: During training, historical leaf data samples are used for training, and the weight vector w and the bias term b are adjusted based on the gradient descent method to make the predicted value y close to the disease label value. The calculation formula is: Where a is the number of training samples, ys is the true disease value of the sth sample, To predict the sth sample disease value, the weight vector w and the bias term b are updated based on minimizing the MSE, where the formula for updating w and b by minimizing the MSE is: Among them, α is the learning rate. After iterative training, the optimized weight vector w and bias term b are obtained to judge the disease of the leaf image; the leaf pest threshold is established based on the expert method, and the calculated result is compared with the threshold.

[0012] Preferably, after receiving the uploaded observation result data, the suggestion generation module in step S6 performs data analysis and processing based on IoT retrieval and gives subsequent intelligent management suggestions.

[0013] An intelligent strawberry production management system, the management system comprising: an environmental data acquisition module, an irrigation module, a lighting module, a temperature adjustment module, an image acquisition module, an image analysis module and a suggestion generation module; The environmental data acquisition module is used to collect environmental data of the strawberry garden, including temperature sensors, moisture sensors and light sensors; The irrigation module intelligently irrigates the strawberry garden based on the data collected by the environmental data acquisition module; The lighting module intelligently adjusts the lighting in the strawberry garden based on the data collected by the environmental data collection module; The temperature adjustment module intelligently adjusts the temperature in the strawberry garden environment based on the data collected by the environmental data collection module; The image acquisition module is used to collect image data of strawberry leaves, which is collected regularly based on drones; The image analysis module analyzes the leaf image data to determine whether the leaves are infested with insects; The suggestion generation module gives intelligent suggestions for subsequent management based on the current growth of strawberries.

[0014] Compared with the prior art, the present invention has the following beneficial effects: The present invention can accurately grasp environmental data in real time and intelligently adjust the environment in the strawberry garden by installing temperature, moisture and light sensors in the strawberry garden. It uses a drone to regularly take images of strawberry leaves and uses an image analysis module and a machine learning algorithm to judge the disease situation. The disease can be discovered in time at the early stage, which improves the monitoring efficiency and accuracy. Measures can be taken before the disease spreads on a large scale, thereby reducing the use of pesticides, reducing the cost of prevention and control, ensuring the quality and safety of strawberries, and intelligently managing the growth of strawberries. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 To invent a flowchart of strawberry management steps; Figure 2 This is a framework diagram of the invented strawberry management system. DETAILED DESCRIPTION

[0016] The following description is used to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments described below are only examples, and those skilled in the art may think of other obvious variations.

[0017] Reference Figure 1 and Figure 2 As shown, an intelligent strawberry production management method, the strawberry production management steps are: S1. Select fertile soil with strong water and fertilizer retention capacity, test the soil pH value, test the soil pH value, and construct a strawberry garden; S2. Select suitable strawberry varieties, sow and plant the strawberry varieties, and use runner propagation to breed them; S3, installing environmental sensors in the strawberry garden, monitoring the environmental data in the strawberry garden in real time, and uploading the data to the management module. The sensors include temperature sensors, moisture sensors and light sensors; S4, water irrigation, after the water sensor detects that the soil moisture is lower than the threshold of normal growth, intelligent water irrigation is carried out to ensure the normal growth of strawberries; intelligent management and control of the strawberry garden is carried out based on temperature sensors and light sensors; S5. Regularly use drones to take real-time photos of strawberry leaves in the strawberry garden, obtain leaf image data, upload the image data, analyze the leaf image data based on the image analysis module, determine whether the leaves have diseases based on the machine learning algorithm, establish a hazard threshold, and send an early warning message to the management module when the leaf disease reaches the threshold during monitoring, so that the management personnel can take preventive measures; S6. Regularly observe the strawberry garden to check the growth of strawberries and the fullness of the fruits, apply nitrogen fertilizer and potassium fertilizer to improve the yield and quality of strawberries, compile and upload the observation results to the suggestion generation module, and give subsequent intelligent management suggestions based on the current growth of strawberries.

[0018] By installing temperature, moisture and light sensors in the strawberry garden, the application can accurately grasp environmental data in real time. For example, the temperature sensor can detect temperature changes in time to prevent strawberries from suffering from high temperature heat damage or low temperature frost damage, and ensure that they grow in a suitable temperature range; the moisture sensor can accurately determine the soil moisture status to avoid root hypoxia and rot due to excessive moisture or poor plant growth due to too little moisture, and realize intelligent water irrigation, which not only saves water but also meets the water requirements of strawberries; the light sensor can ensure that strawberries obtain appropriate light intensity and duration at each growth stage, promote efficient photosynthesis, provide ideal environmental conditions for strawberry growth, and improve fruit quality and yield; Using drones to regularly take images of strawberry leaves and using image analysis modules and machine learning algorithms to determine disease conditions can help detect diseases in their early stages. Compared with traditional manual inspections, this greatly improves monitoring efficiency and accuracy, and can take measures before the disease spreads on a large scale, thereby reducing the use of pesticides, reducing prevention and control costs, ensuring strawberry quality and safety, and reducing yield losses caused by diseases. Upload various observation data to the corresponding module and generate intelligent suggestions to provide managers with a scientific decision-making basis. Whether it is environmental regulation, disease prevention or fertilizer management, accurate and reasonable operations can be made based on data and analysis results, reducing the errors and subjectivity of manual experience judgment, improving the overall management level and production efficiency of the strawberry garden, and reducing labor intensity and management costs.

[0019] In step S1, soil selection is done by the planting staff observing the color of the soil surface in advance, selecting soil with dark color, observing the activity of the soil, and judging by the number of earthworms in the soil. If there are more earthworms in the soil, the surface soil structure is excellent, and the air permeability and fertility are high. The soil pH value is tested by obtaining local soil samples and sending them to the laboratory for pH and soil nutrient testing. If the pH value is detected to be too high or too low, lime is added to the soil for adjustment, and sulfur powder and acidic fertilizers are used to improve the soil pH.

[0020] This application observes the color of the soil surface to preliminarily screen out potentially fertile soil. Dark-colored soil usually implies a higher organic matter content, because the decomposition and accumulation of organic matter in the soil will make the soil color darker. Strawberries have strict requirements on the pH value of the soil and are generally suitable for growing within a specific pH range. By sending soil samples to the laboratory for precise pH testing, the pH status of the soil can be accurately understood, thereby determining whether improvement is needed.

[0021] In step S2, the seedbed for sowing is prepared, the soil of the seedbed is deeply plowed to a depth of 20cm-30cm, and the pre-screened healthy and plump strawberry seeds are evenly sown on the seedbed. After sowing, a layer of 0.5cm-1cm fine soil is covered, and water is sprayed with a fine mist nozzle, waiting for the seeds to germinate. During the seed germination process, the temperature is controlled between 15℃-25℃.

[0022] The present application conducts 20cm-30cm deep tillage of the seedbed soil, which can break the plow bottom layer. The plow bottom layer is a relatively compact soil layer formed due to long-term shallow tillage or natural compaction of the soil. It will hinder the growth of strawberry roots and the circulation of soil moisture and air. Deep tillage can loosen this soil layer, increase the gaps between soil particles, and form a good aggregate structure. Pre-screening healthy and plump strawberry seeds is a key step to ensure a high germination rate and strong growth of seedlings. The temperature is controlled between 15℃-25℃ because this temperature range is very suitable for the germination of strawberry seeds.

[0023] In step S2, the runner propagation method is to tidy up the mother plant, cut off the sick, aging and poorly growing branches and leaves, and when the runners grow and extend, guide the runners to extend toward the predetermined propagation area, use small bamboo sticks to fix the runners on loose, moist and fertile soil, and fertilize every 7d-10d.

[0024] The present application guides the runners to extend toward the predetermined propagation area, so that the spatial distribution of the strawberry seedlings can be more reasonable, which can make full use of land resources, avoid crowding of plants caused by the random growth of runners, provide sufficient growth space for each new seedling, and facilitate subsequent management operations, such as irrigation, fertilization, and pest and disease control.

[0025] In step S3, temperature sensors are installed between plant rows, on the ground surface and at different heights in the strawberry garden. The intervals between plant rows are 5m-10m to capture the air temperature around the strawberry plants. Temperature sensors are installed on the ground surface to capture changes in soil surface temperature. Temperature sensors are installed at different heights in the strawberry garden to monitor temperature changes in the garden. The moisture sensor is installed in an inserted manner to monitor changes in soil moisture content in different areas and obtain soil moisture data in real time. The light sensor is installed on the top of the strawberry garden to collect light data of the strawberry garden. The collected data is uploaded.

[0026] This application uploads the collected various environmental data to the management module, which can build a data foundation for the intelligent management of the strawberry garden. The management module can integrate, analyze and store these data, and automatically generate corresponding management suggestions based on data changes through preset algorithms and models, such as automatically controlling the opening and closing of the irrigation system, and adjusting the shading and ventilation equipment of the greenhouse. At the same time, growers can also view the data remotely and grasp the environmental conditions of the strawberry garden at any time. Even if they are not on site in the park, they can make reasonable management decisions in time, which greatly improves the efficiency, accuracy and convenience of strawberry garden management.

[0027] In step S4, based on the expert analysis method, the soil moisture content threshold, temperature threshold and light threshold suitable for strawberry growth are established respectively; the real-time collected data are compared with the thresholds to determine whether the current strawberry garden environment is suitable for normal strawberry growth. If the real-time collected data is lower than the threshold, it is determined that the current strawberry garden environment is not suitable for normal strawberry growth, and the irrigation water is controlled accordingly, the temperature changes are adjusted, and the light conditions are adjusted; the expert analysis method selects strawberry growth experts to collect existing strawberry planting record data, and based on historical data, the threshold is obtained through expert analysis.

[0028] This application uses expert analysis to establish the soil moisture content threshold for strawberry growth, providing a precise target range for water management in strawberry gardens. Collecting existing strawberry planting record data and analyzing based on historical data is a very effective method. Historical data contains various environmental parameters, strawberry growth conditions, and corresponding management measures and results in the past strawberry planting process. By mining these data, the potential laws between strawberry growth and environmental factors can be discovered.

[0029] In step S5, drones are used to take photos regularly. The leaves taken are randomly sampled and uploaded. The image analysis module preprocesses the acquired image data, including image grayscale processing, image filtering processing and image enhancement processing. Image grayscale processing is based on component method. Image filtering processing adopts median filtering processing. The pixel values ​​in the filter window are sorted and the middle value is taken as the value of the center pixel. Image enhancement processing is based on histogram equalization processing. The histogram of the original image is stretched to make the distribution of pixel grayscale values ​​uniform and increase the contrast of the image. The characteristics of insect pests in the leaves are extracted, and the color, texture and wormhole characteristics of the leaves are extracted.

[0030] The component method of this application is a color space conversion method in image processing technology, which is used to convert a color image into a grayscale image. In the common RGB color image model, each pixel of the color image is composed of three color channels: red (R), green (G), and blue (B). The basic principle of the component method is to directly select the value of one of the color channels as the grayscale value corresponding to the grayscale image; Converting color images into grayscale images can reduce the amount of data and computational complexity. Grayscale images eliminate the interference of color information to a certain extent, making the texture and shape characteristics of the leaves more prominent. In the actual shooting process, the image may be interfered by various factors and produce noise, such as changes in lighting and vibrations of drone flights. Based on the extracted pest characteristics, accurate prevention and control strategies can be formulated. If it is found that the leaves are mainly early symptoms of pests with color changes, biological control or physical control methods can be adopted, such as releasing natural enemies of pests or using insect-attracting lamps; if there are many wormholes and the pest is serious, it is necessary to consider the use of chemical pesticides for prevention and control, and determine the key areas for spraying pesticides based on the distribution of wormholes.

[0031] Machine learning algorithm calculation Assume that the color feature vector extracted from the leaf image is C = (c1, c2, ..., c m ), where c i Represents the specific color feature value, and the texture feature vector is T=(t1, t2, ... t n ), t j Expressed as texture features, the wormhole feature vector is H = (h1, h2, ... h k ), h l The wormhole quantity feature is obtained by merging all feature vectors into a total feature vector X = (C, T, H) = (x1, x2, ..., x m+n+k ), let the weights be w=(w1,w2,…,w m+n+k ), the bias term is b, and the predicted value y of the disease degree is calculated using the following formula: During training, historical leaf data samples are used for training, and the weight vector w and the bias term b are adjusted based on the gradient descent method to make the predicted value y close to the disease label value. The calculation formula is: Where a is the number of training samples, ys is the true disease value of the sth sample, To predict the sth sample disease value, the weight vector w and the bias term b are updated based on minimizing the MSE, where the formula for updating w and b by minimizing the MSE is: Among them, α is the learning rate. After iterative training, the optimized weight vector w and bias term b are obtained to judge the disease of the leaf image; the leaf pest threshold is established based on the expert method, and the calculated result is compared with the threshold.

[0032] After receiving the uploaded observation result data, the suggestion generation module in step S6 conducts data analysis and processing based on IoT retrieval and gives intelligent suggestions for subsequent management.

[0033] The intelligent suggestions in this application provide scientific decision-making basis for strawberry garden managers. These suggestions are generated based on data analysis and processing results, combined with professional agricultural knowledge and best practices. Strawberry garden management based on scientific data analysis and intelligent suggestions will help achieve sustainable development. Accurate fertilization and irrigation suggestions can reduce the waste of chemical fertilizers and water resources and reduce negative impacts on the environment.

[0034] An intelligent strawberry production management system, the management system comprising: an environmental data acquisition module, an irrigation module, a lighting module, a temperature adjustment module, an image acquisition module, an image analysis module and a suggestion generation module; The environmental data acquisition module is used to collect environmental data of the strawberry garden, including temperature sensors, moisture sensors and light sensors; The irrigation module intelligently irrigates the strawberry garden based on the data collected by the environmental data acquisition module; The lighting module intelligently adjusts the lighting in the strawberry garden based on the data collected by the environmental data collection module; The temperature adjustment module intelligently adjusts the temperature in the strawberry garden environment based on the data collected by the environmental data collection module; The image acquisition module is used to collect image data of strawberry leaves, which is collected regularly based on drones; The image analysis module analyzes the leaf image data to determine whether the leaves are infested with insects; The suggestion generation module gives intelligent suggestions for subsequent management based on the current growth of strawberries.

[0035] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments, and the above embodiments and descriptions only describe the principles of the present invention. The present invention may be subject to various changes and improvements without departing from the spirit and scope of the present invention, and these changes and improvements fall within the scope of the present invention claimed.

Claims

1. An intelligent strawberry production management method, characterized in that: The steps for strawberry production management are: S1. Select fertile soil with strong water and fertilizer retention capacity, test the soil pH value, test the soil pH value, and construct a strawberry garden; S2. Select suitable strawberry varieties, sow and plant the strawberry varieties, and use runner propagation to breed them; S3, installing environmental sensors in the strawberry garden, monitoring the environmental data in the strawberry garden in real time, and uploading the data to the management module. The sensors include temperature sensors, moisture sensors and light sensors; S4, water irrigation, after the water sensor detects that the soil moisture is lower than the threshold of normal growth, intelligent water irrigation is carried out to ensure the normal growth of strawberries; intelligent management and control of the strawberry garden is carried out based on temperature sensors and light sensors; S5. Regularly use drones to take real-time photos of strawberry leaves in the strawberry garden, obtain leaf image data, upload the image data, analyze the leaf image data based on the image analysis module, determine whether the leaves have diseases based on the machine learning algorithm, establish a hazard threshold, and send an early warning message to the management module when the leaf disease reaches the threshold during monitoring, so that the management personnel can take preventive measures; S6. Regularly observe the strawberry garden to check the growth of strawberries and the fullness of the fruits, apply nitrogen fertilizer and potassium fertilizer to improve the yield and quality of strawberries, compile and upload the observation results to the suggestion generation module, and give subsequent intelligent management suggestions based on the current growth of strawberries.

2. An intelligent strawberry production management method according to claim 1, characterized in that: In step S1, soil selection is done by the planting staff observing the color of the soil surface in advance, selecting soil with dark color, observing the activity of the soil, and judging by the number of earthworms in the soil. If there are more earthworms in the soil, the surface soil structure is excellent, and the air permeability and fertility are high. The soil pH value is tested by obtaining local soil samples and sending them to the laboratory for pH and soil nutrient testing. If the pH value is detected to be too high or too low, lime is added to the soil for adjustment, and sulfur powder and acidic fertilizers are used to improve the soil pH.

3. An intelligent strawberry production management method according to claim 1, characterized in that: In step S2, the seedbed for sowing is prepared, the soil of the seedbed is deeply plowed to a depth of 20cm-30cm, and the pre-screened healthy and plump strawberry seeds are evenly sown on the seedbed. After sowing, a layer of 0.5cm-1cm fine soil is covered, and water is sprayed with a fine mist nozzle, waiting for the seeds to germinate. During the seed germination process, the temperature is controlled between 15℃-25℃.

4. An intelligent strawberry production management method according to claim 1, characterized in that: In step S2, the runner propagation method is to tidy up the mother plant, cut off the sick, aging and poorly growing branches and leaves, and when the runners grow and extend, guide the runners to extend toward the predetermined propagation area, use small bamboo sticks to fix the runners on loose, moist and fertile soil, and fertilize every 7d-10d.

5. An intelligent strawberry production management method according to claim 1, characterized in that: In step S3, temperature sensors are installed between plant rows, on the ground surface and at different heights in the strawberry garden. The intervals between plant rows are 5m-10m to capture the air temperature around the strawberry plants. Temperature sensors are installed on the ground surface to capture changes in soil surface temperature. Temperature sensors are installed at different heights in the strawberry garden to monitor temperature changes in the garden. The moisture sensor is installed in an inserted manner to monitor changes in soil moisture content in different areas and obtain soil moisture data in real time. The light sensor is installed on the top of the strawberry garden to collect light data of the strawberry garden. The collected data is uploaded.

6. An intelligent strawberry production management method according to claim 1, characterized in that: In step S4, based on the expert analysis method, the soil moisture content threshold, temperature threshold and light threshold suitable for strawberry growth are established respectively; the real-time collected data are compared with the thresholds to determine whether the current strawberry garden environment is suitable for normal strawberry growth. If the real-time collected data is lower than the threshold, it is determined that the current strawberry garden environment is not suitable for normal strawberry growth, and the irrigation water is controlled accordingly, the temperature changes are adjusted, and the light conditions are adjusted; the expert analysis method selects strawberry growth experts to collect existing strawberry planting record data, and based on historical data, the threshold is obtained through expert analysis.

7. An intelligent strawberry production management method according to claim 1, characterized in that: In step S5, drones are used to take photos regularly. The leaves taken are randomly sampled and uploaded. The image analysis module preprocesses the acquired image data, including image grayscale processing, image filtering processing and image enhancement processing. Image grayscale processing is based on component method. Image filtering processing adopts median filtering processing. The pixel values ​​in the filter window are sorted and the middle value is taken as the value of the center pixel. Image enhancement processing is based on histogram equalization processing. The histogram of the original image is stretched to make the distribution of pixel grayscale values ​​uniform and increase the contrast of the image. The characteristics of insect pests in the leaves are extracted, and the color, texture and wormhole characteristics of the leaves are extracted.

8. An intelligent strawberry production management method according to claim 7, characterized in that: Machine learning algorithm calculation Assume that the color feature vector extracted from the leaf image is C = (c1, c2, ..., c m ), where c i Represents the specific color feature value, and the texture feature vector is T=(t1, t2, ... t n ), t j Expressed as texture features, the wormhole feature vector is H = (h1, h2, ... h k ), h l The wormhole quantity feature is obtained by merging all feature vectors into a total feature vector X = (C, T, H) = (x1, x2, ..., x m+n+k ), let the weights be w=(w1,w2,…,w m+n+k ), the bias term is b, and the predicted value y of the disease degree is calculated using the following formula: During training, historical leaf data samples are used for training, and the weight vector w and the bias term b are adjusted based on the gradient descent method to make the predicted value y close to the disease label value. The calculation formula is: Where a is the number of training samples, ys is the true disease value of the sth sample, To predict the sth sample disease value, the weight vector w and the bias term b are updated based on minimizing the MSE, where the formula for updating w and b by minimizing the MSE is: Where α is the learning rate. After iterative training, the optimized weight vector w and bias term b are obtained to judge the disease of the leaf image. The leaf pest threshold was established based on the expert method, and the calculated results were compared with the threshold.

9. An intelligent strawberry production management method according to claim 1, characterized in that: After receiving the uploaded observation result data, the suggestion generation module in step S6 conducts data analysis and processing based on IoT retrieval and gives intelligent suggestions for subsequent management.

10. An intelligent strawberry production management system, characterized in that: The management system includes: environmental data acquisition module, irrigation module, lighting module, temperature regulation module, image acquisition module, image analysis module and suggestion generation module; The environmental data acquisition module is used to collect environmental data of the strawberry garden, including temperature sensors, moisture sensors and light sensors; The irrigation module intelligently irrigates the strawberry garden based on the data collected by the environmental data acquisition module; The lighting module intelligently adjusts the lighting in the strawberry garden based on the data collected by the environmental data collection module; The temperature adjustment module intelligently adjusts the temperature in the strawberry garden environment based on the data collected by the environmental data collection module; The image acquisition module is used to collect image data of strawberry leaves, which is collected regularly based on drones; The image analysis module analyzes the leaf image data to determine whether the leaves are infested with insects; The suggestion generation module gives intelligent suggestions for subsequent management based on the current growth of strawberries.

Images