A planting management method and system for smart agriculture

Abstract

The present invention relates to the field of smart agricultural planting technology, and provides a smart agricultural planting management method and system. By acquiring monitoring data of a planting area, a planting database is established. Targeted storage data is acquired from the planting database through retrieval analysis, and the data is substituted into a preset training model to acquire initial control parameters and later control parameters corresponding to the plants before and after they emerge from the soil, respectively. Then, target parameters that best fit the real-time state of the current plants can be acquired, and then, by comparing with the real-time monitoring data of the planted plants, it is determined whether the planting parameters need to be regulated. Based on the database, the most suitable planting parameters are acquired through big data analysis and calculation in the cloud, which can effectively improve the level of intelligent agricultural planting, and thus meet the needs of open-air planting.

Description

Technical Field

[0001] The present invention relates to the field of smart agriculture planting technology, and in particular to a smart agriculture planting management method and system. Background Art

[0002] What follows is the planting method of smart agriculture, which can effectively improve the agricultural ecological environment. It regards farmland, livestock farms, aquaculture bases and other production units and the surrounding ecological environment as a whole, and through systematic and precise calculation of their material exchange and energy cycle relationship, ensures that the ecological environment of agricultural production is within an acceptable range. For example, quantitative fertilization will not cause soil compaction, and the treated livestock and poultry manure will not cause water and air pollution, but can improve soil fertility.

[0003] Smart agriculture is essentially the use of sensors and software to control agricultural production through mobile platforms or computer platforms, and to monitor and manage operations such as sowing and irrigation of crops through manual control in the form of monitoring. However, in the actual management process, decisions are still made based on manual experience, with too many subjective factors, and cannot meet the needs of intelligence.

[0004] At the same time, smart agricultural planting methods are mostly suitable for greenhouse planting, which promotes plant growth by changing the temperature, humidity, light and other data in the greenhouse. However, this method cannot effectively control and manage crops planted in open-air farmland. Summary of the invention

[0005] The present invention provides a planting management method and system for smart agriculture, which solves the technical problems that the existing smart agriculture planting management method relies on manual processing, is not intelligent enough, and cannot be applied to open-air planting management.

[0006] In order to solve the above technical problems, the present invention provides a planting management method of smart agriculture, comprising the steps of:

[0007] S1. Acquire monitoring data of the planting area and store it using XML data storage architecture to obtain a planting database;

[0008] S2. Retrieve relevant stored data from the planting database according to the plant category, substitute it into the preset training model, and obtain initial control parameters and later control parameters;

[0009] S3, obtaining the growth image of the current plant and identifying it, determining whether it is in a growth state, if so, calling the corresponding later control parameter as the target parameter, otherwise calling the corresponding initial control parameter as the target parameter;

[0010] S4. Acquire real-time monitoring data of the planted plants and compare the data with the target parameters. If the comparison is inconsistent, adjust the planting parameters of the planted plants according to the target parameters.

[0011] This basic solution obtains monitoring data from the planting area, establishes a planting database, obtains targeted storage data from the planting database through retrieval and analysis, substitutes it into the preset training model, and obtains the initial control parameters and later control parameters corresponding to the plants before and after they emerge from the ground, thereby obtaining the target parameters that best fit the current real-time status of the plants, and then compares them with the real-time monitoring data of the planted plants to determine whether the planting parameters need to be adjusted. Based on the database, the most appropriate planting parameters are obtained through big data analysis and calculation in the cloud, which can effectively improve the level of intelligent agricultural planting and meet the needs of open-air planting.

[0012] In a further embodiment, step S1 comprises the steps of:

[0013] S11, collecting data from the open-air planting area of ​​each type of plant through a sensor network to obtain monitoring data;

[0014] S12, marking the corresponding area label and monitoring time for the monitoring data;

[0015] S13, uploading the labeled monitoring data to the cloud, and storing it in an XML data storage architecture to obtain a planting database for each type of plant;

[0016] The monitoring data includes at least one or more of soil data, temperature data, moisture data, image data and weather data.

[0017] This solution classifies and stores the monitoring data, regional labels and monitoring time of each type of plant in a targeted manner, which is conducive to the subsequent targeted planting management. At the same time, data storage is performed in an XML data storage architecture, which can improve the orderliness and effectiveness of data storage and facilitate the calculation of subsequent initial control parameters and later control parameters.

[0018] In a further embodiment, step S2 comprises the steps of:

[0019] S21, based on the current planted plant, obtaining the corresponding stored data from the planting database, screening the stored data of the approximate growth environment through keyword retrieval, performing multiple regression analysis to obtain a relevant regression function, and calculating the corresponding initial control parameters according to the relevant regression function and the current environmental parameters;

[0020] S22. Based on the current plant, the corresponding stored data is obtained from the planting database, the stored data of the approximate growth environment is obtained through keyword retrieval, and the corresponding later control parameters are obtained by weighted average calculation.

[0021] In a further embodiment, step S21 comprises the steps of:

[0022] A1. Based on the current planting plant, match the corresponding planting database;

[0023] A2. Perform keyword search on the planting database according to the first search condition to obtain the stored data that is most similar to the growth environment of the area where the current plant belongs;

[0024] A3. Obtain the stored data in the cultivation stage and remove outliers;

[0025] A4, using the soil data and moisture data in the stored data as dependent variables, and the weather data and temperature data as independent variables to perform multiple regression analysis to obtain a relevant regression function;

[0026] A5. Substituting the current environmental parameters of the region into the relevant regression function to obtain initial control parameters;

[0027] Among them, the first search conditions are in the order of similar areas, different years, and close time periods; the current environmental parameters include weather data and temperature data, and the initial control parameters include target soil data and target moisture data.

[0028] This solution sets targeted retrieval conditions for the actual planting situation of plant seeds from sowing to germination, "sequentially similar areas, different years, and close time periods", and then selects the storage data corresponding to the optimal germination conditions for the plants; after outlier removal processing, the validity of the data is further guaranteed and the germination probability of the plants is improved; through multivariate regression analysis, the relevant regression function is obtained to establish the relationship between controllable variables (i.e., dependent variables, soil data and moisture data) and uncontrollable variables (independent variables, weather data and temperature data), and then after substituting the current environmental parameters of the area to which it belongs, the soil data and moisture data that are most suitable for the current state of the plants can be obtained. The high accuracy of the control data can improve the efficiency of plant cultivation.

[0029] In a further embodiment, step S22 comprises the steps of:

[0030] B1. Based on the current planting plants, match the corresponding planting database;

[0031] B2. Perform keyword search on the planting database according to the second search condition to obtain the stored data that is most similar to the growth environment of the area where the currently planted plant belongs;

[0032] B3, performing image recognition on each growth image in the stored data and scoring the growth status;

[0033] B4, obtaining several groups of stored data with higher scores, performing weighted average calculation, and obtaining predicted data;

[0034] B5, obtaining weather data and temperature data in a time period close to the predicted data, determining whether there is abnormal weather, and if so, discarding the stored data corresponding to the abnormal weather, and returning to step B2 to regenerate the predicted data as the later control parameter;

[0035] Among them, the second search conditions are in the order of similar areas, the same year, and close time periods; the later control parameters include target soil data and target moisture data.

[0036] This solution sets targeted retrieval conditions of "similar areas in sequence, same year, and close time periods" according to the actual planting conditions of the plants during their growth stages, providing the plants with the closest and most reasonable control data. It uses image recognition to identify the growth images in each stored data and scores the growth conditions. It can control the plants to enter similar growth environments through stored data with better growth conditions, thereby ensuring that the growth environment of the plants tends to be optimal. It uses weighted average calculation for data prediction to ensure the balanced stability of the later control parameters. At the same time, it eliminates data from abnormal weather to prevent abnormal weather data from interfering with the analysis of the data parameters of the normal growth environment of the plants (i.e., the later control parameters), thereby improving the data accuracy and ensuring the effectiveness of the later control data.

[0037] In a further implementation scheme, in step B5, the judgment of abnormal weather is specifically as follows: obtaining weather data and temperature data, and drawing a corresponding change curve, performing differential calculation on the change curve, if the differential calculation value is greater than a preset threshold, it is considered that the external environment has changed drastically and abnormal weather exists in the corresponding time period.

[0038] In a further embodiment, step S4 comprises the steps of:

[0039] S41, dividing the range by taking the target parameter as the midpoint value of the interval and 10% of the midpoint value as the interval length to generate a control interval;

[0040] S42. Obtain real-time monitoring data of the planted plants to determine whether they are within the control range. If so, do nothing. Otherwise, calculate the difference between the real-time monitoring data and the closest boundary value of the control range, and generate a control signal based on the difference to control the corresponding control equipment to perform environmental control on the planting area of ​​the planted plants.

[0041] This scheme uses the target parameter to set the median point of the interval and the control interval to adapt to the impact of the actual environmental change range on plants and is more in line with the actual plant growth. In the comparative analysis of the control interval and the real-time monitoring data, the difference growth control signal between the real-time monitoring data and the closest boundary value of the control interval is used to avoid excessive changes in the plant environment, but to achieve a relatively good growth environment to ensure the normal growth of the plant.

[0042] The present invention also provides a planting management system for smart agriculture, which is used to implement the above-mentioned planting management method for smart agriculture, including a processing module and a sensor network, a data terminal, and a control device connected to the processing module;

[0043] The sensor network is used to collect monitoring data of open-air planting areas;

[0044] The processing module is used to upload the monitoring data to the data terminal;

[0045] The data terminal is used to obtain monitoring data of the planting area and store it in an XML data storage architecture to obtain a planting database; it is also used to retrieve relevant stored data from the planting database according to plant categories, substitute it into a preset training model, and obtain initial control parameters and later control parameters;

[0046] The processing module is used to obtain the growth image of the current plant and identify it, determine whether it is in a growth state, and if so, call the corresponding later control parameter as the target parameter, otherwise call the corresponding initial control parameter as the target parameter; and also obtain the real-time monitoring data of the plant and compare it with the target parameter, and if the comparison is inconsistent, generate a control signal according to the target parameter;

[0047] The control device is used to perform environmental control on the planting area according to the control signal.

[0048] In a further embodiment, the regulating equipment includes one or more of drip irrigation equipment, drainage equipment and fertilization equipment.

[0049] In a further implementation scheme, the sensor network adopts a distributed Internet of Things, which includes data-connected sensor nodes and gateway devices; the processing module is connected to a number of the gateway devices, and the gateway device is connected to a number of the sensor nodes.

[0050] This solution applies Internet of Things technology to traditional agriculture, using sensors and software to control agricultural production through mobile platforms or computer platforms, making traditional agriculture more "intelligent". BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 This is a workflow diagram of a planting management method for smart agriculture provided by an embodiment of the present invention;

[0052] Figure 2 It is a system framework diagram of a smart agriculture planting management system provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0053] The following specifically illustrates the implementation mode of the present invention in conjunction with the accompanying drawings. The embodiments are provided for illustrative purposes only and cannot be understood as limiting the present invention. The accompanying drawings are provided for reference and illustration only and do not constitute a limitation on the scope of patent protection of the present invention, because many changes can be made to the present invention without departing from the spirit and scope of the present invention.

[0054] Example 1

[0055] An embodiment of the present invention provides a planting management method for smart agriculture, such as Figure 1 As shown, in this embodiment, the steps include:

[0056] S1, obtaining monitoring data of the planting area, and storing it in an XML data storage architecture to obtain a planting database, including steps S11 to S13:

[0057] S11, collecting data from the open-air planting area of ​​each type of plant through a sensor network to obtain monitoring data;

[0058] S12, marking the corresponding area labels and monitoring time for the monitoring data;

[0059] S13, uploading the marked monitoring data to the cloud, and storing it in an XML data storage architecture to obtain a planting database for each type of plant;

[0060] Specifically, the XML data storage architecture stores data in a progressive order of region, time, and monitoring data, and uses the region as a constraint, the time as a detailed description under the region, the weather data in the monitoring data as a detailed description under the time, the temperature data as a detailed description under the weather data, and other monitoring data as detailed descriptions under the temperature data. At the same time, different monitoring data are arranged and constrained in a set order, and the set order can be soil data, moisture data, and image data. The XML architecture is constructed through the above content. After the architecture is constructed, the received data is stored in the corresponding position of the architecture to form a storage file. Among them, in other embodiments, the order and type of factors such as region, time weather, and temperature can be selectively adjusted according to the impact on plant growth, and this embodiment does not limit it.

[0061] In this embodiment, the monitoring data includes at least one or more of soil data, temperature data, moisture data, image data and weather data. Soil data refers to the content of a certain substance in the soil, and light and wind are used as weather data.

[0062] Alternatively, it also includes S14, supplementing the artificial planting data in the planting database.

[0063] The supplement of artificial planting data can ensure the adequacy of the previous data. Since the above data are arranged according to a certain structure, effective information can be extracted from the data, providing a data basis for subsequent data analysis, which effectively saves the time for data sorting.

[0064] This embodiment classifies and stores the monitoring data, area labels and monitoring time of each type of plants in a targeted manner, which is beneficial to the subsequent targeted planting management. At the same time, data is stored in an XML data storage architecture, which can improve the orderliness and effectiveness of data storage and facilitate the subsequent calculation of initial control parameters and later control parameters.

[0065] S2, according to the plant category, retrieve the relevant stored data from the planting database, substitute it into the preset training model, and obtain the initial control parameters and the later control parameters, including steps S21~S22:

[0066] S21, based on the current plant, obtain the corresponding stored data from the planting database, filter the stored data of the approximate growth environment by keyword search, perform multivariate regression analysis to obtain the relevant regression function, and calculate the corresponding initial control parameters according to the relevant regression function and the current environmental parameters, including steps A1 to A5:

[0067] A1. Based on the current planting plant, match the corresponding planting database;

[0068] A2. Perform keyword search on the planting database according to the first search condition to obtain the stored data that is most similar to the growth environment of the area where the current plant belongs;

[0069] Among them, the first search conditions are similar areas, different years, and close time periods in order.

[0070] For example, the specific retrieval process is as follows: extract an area within a certain range of the region, such as an area within the shortest distance of 100 meters from the border, and extract a detailed description of the time before and after a period of time in different years under the above region, such as 30 days before and after, extract weather data within a certain threshold of the difference before and after the time description, such as a wind speed difference within 1.5 meters per second, and extract temperature data within a certain range of temperature differences under the weather data, such as an upper and lower difference of 4°C, to complete the extraction of data related to the region (i.e., the stored data in step A2).

[0071] A3. Obtain the stored data in the cultivation phase and remove outliers.

[0072] Specifically, the principles for selecting stored data in the cultivation phase are as follows:

[0073] Image recognition is performed on the image data under the temperature data in the storage data in step A2 to identify whether the image is a state where growth can be observed and the specific growth state score. In this recognition process, the convolutional neural network consisting of three convolutional layers and two fully connected layers constructed by the python component is recognized and completed. After the construction is completed, it needs to be trained, and it can be used after the training is effective. If the growth state of the area cannot be observed, it is judged to be an image at a certain time after sowing, that is, an image of the cultivation stage, and the corresponding data is also the storage data of the cultivation stage. The image recognition training model is a commonly used technical means in this field, and this embodiment is not limited and will not be repeated.

[0074] At this time, the soil data and moisture data corresponding to the cultivation stage image are retained, and outliers are removed from the soil data and moisture data.

[0075] A4. Perform multiple regression analysis using soil data and moisture data in the stored data as dependent variables and weather data and temperature data as independent variables to obtain the relevant regression function.

[0076] For example, multiple regression analysis is performed with nitrogen, phosphorus, potassium content and water data in soil data as dependent variables, weather data and temperature data as independent variables, and the relevant regression functions of nitrogen, phosphorus, potassium content and water data in soil data are generated respectively. The formula of the relevant regression function is as follows:

[0077]

[0078] Among them, x1 is light data, x2 is wind data, and x3 is temperature data. In the above formula, when y represents different dependent variables, its light weight β1, wind weight β2, temperature weight β3, and constant term c are set with different constant values ​​according to the different dependent variables represented by y.

[0079] A5. Substitute the current environmental parameters of the region into the relevant regression function to obtain the initial control parameters;

[0080] Specifically, by substituting the weather data and temperature data in the current environmental parameters of the region into the relevant regression function, the initial control parameters of the soil data and the moisture data are generated.

[0081] Among them, the current environmental parameters include weather data and temperature data, and the initial control parameters include target soil data and target moisture data.

[0082] In open-air farmland, because weather temperature is an uncontrollable factor, that is, as an independent variable, soil data and moisture data are predicted under different weather and temperatures, and then soil data and moisture data suitable for the above conditions are generated, and the control equipment is used to control according to the above data to achieve effective control and management of open-air farmland sowing.

[0083] This embodiment sets targeted search conditions of "sequentially similar regions, different years, and close time periods" for the actual planting conditions of plant seeds from sowing to germination, and then selects the storage data corresponding to the optimal germination conditions for the plants; after outlier removal processing, the validity of the data is further guaranteed and the germination probability of the plants is improved; the relevant regression function is obtained through multivariate regression analysis, and the relationship between the controllable variables (i.e., the dependent variable, soil data and moisture data) and the uncontrollable variables (independent variables, weather data and temperature data) is established, and then after substituting the current environmental parameters of the area to which it belongs, the soil data and moisture data that are most suitable for the current state of the plants can be obtained, and the high accuracy of the control data can improve the efficiency of plant cultivation.

[0084] S22, based on the current plant, obtain the corresponding stored data from the planting database, obtain the stored data of the approximate growth environment through keyword retrieval, and perform weighted average calculation to obtain the corresponding later control parameters, including steps B1 to B5:

[0085] B1. Based on the current planting plants, match the corresponding planting database;

[0086] B2. Perform keyword search on the planting database according to the second search condition to obtain the stored data that is most similar to the growth environment of the area where the current plant belongs;

[0087] For example, the specific retrieval process is as follows:

[0088] First, a keyword search of the area to which the user belongs is performed in the storage files in the memory, and an area within a larger range of the area to which the user belongs, such as an area within the shortest distance of 300 meters from the border, is extracted. A detailed description of the time in the previous shorter period of time in the same year in the above-mentioned area, such as the previous three days, is also extracted to complete the extraction of data related to the area to which the user belongs (i.e., the extraction of the data stored in step B2).

[0089] B3, performing image recognition on each growth image in the stored data and scoring the growth status;

[0090] B4. Obtain several groups of stored data with higher scores and perform weighted average calculation to obtain predicted data.

[0091] In this embodiment, the higher the score, the greater the weight assigned, and the weights are distributed in equal intervals and the sum is 1, such as the difference between the weights is 0.005.

[0092] B5. Obtain weather data and temperature data in a time period close to the predicted data, determine whether there is abnormal weather, and if so, remove the stored data corresponding to the abnormal weather, and return to step B2 to regenerate the predicted data as a later control parameter;

[0093] Among them, the second search conditions are similar areas, the same year, and close time periods in order; the later control parameters include target soil data and target moisture data.

[0094] In this embodiment, the judgment of abnormal weather is specifically as follows: obtaining weather data and temperature data, drawing a corresponding change curve, and performing differential calculation on the change curve. If the differential calculation value is greater than a preset threshold, it is considered that the external environment has changed drastically and abnormal weather exists in the corresponding time period.

[0095] The specific elimination of the stored data corresponding to abnormal weather is as follows: according to the time that exceeds the preset threshold, the termination time of the node is extracted based on the day before the time, and a detailed description of the time in the time period before the termination time is extracted.

[0096] This embodiment sets targeted search conditions of "similar areas in sequence, same year, and close time periods" according to the actual planting conditions of the plant growth stage to provide the closest and most reasonable control data for the plants; uses image recognition to identify the growth images in each stored data to score the growth status, and can control the plants to enter similar growth environments through stored data with better growth conditions, thereby ensuring that the growth environment of the plants tends to be optimal; uses weighted average calculation for data prediction to ensure the balanced stability of later control parameters; at the same time, data from abnormal weather is eliminated to prevent abnormal weather data from interfering with the analysis of data parameters of the normal growth environment of the plants (i.e., later control parameters), thereby improving data accuracy and ensuring the effectiveness of later control data.

[0097] That is, the initial control parameters are the parameters for controlling the period from sowing to the time when the growth of the plants can be observed through images, and the later control parameters are the parameters for controlling the process of the growth of the plants can be observed through images.

[0098] S3, obtaining the growth image of the current plant and identifying it, judging whether it is in a growth state, if so, calling the corresponding later control parameter as the target parameter, otherwise calling the corresponding initial control parameter as the target parameter;

[0099] S4, obtaining real-time monitoring data of the planted plants and comparing them with the target parameters. If the comparison is inconsistent, adjusting the planting parameters of the planted plants according to the target parameters, including steps S41 to S42:

[0100] S41, taking the target parameter as the midpoint value of the interval and 10% of the midpoint value as the interval length to divide the range and generate a control interval;

[0101] S42. Obtain real-time monitoring data of the planted plants to determine whether they are within the control range. If so, do nothing. Otherwise, calculate the difference between the real-time monitoring data and the closest boundary value of the control range, and generate a control signal based on the difference to control the corresponding control equipment to perform environmental control on the planting area of ​​the planted plants.

[0102] The difference has positive and negative signs, with a positive sign indicating a decrease and a negative sign indicating an increase.

[0103] This embodiment uses the target parameter to set the interval median point and the control interval to adapt to the impact of the actual environmental change range on the plants, which is more in line with the actual plant growth; in the comparative analysis of the control interval and the real-time monitoring data, the difference growth control signal between the real-time monitoring data and the closest boundary value of the control interval is used to avoid excessive changes in the plant environment, but to achieve a relatively good growth environment to ensure the normal growth of the plant.

[0104] The embodiment of the present invention establishes a planting database by acquiring monitoring data of the planting area, obtains targeted storage data from the planting database through retrieval and analysis, substitutes the data into a preset training model, and obtains initial control parameters and later control parameters corresponding to the plants before and after they emerge from the soil, respectively. Then, the target parameters that best fit the real-time state of the current plants can be obtained, and then, by comparing with the real-time monitoring data of the planted plants, it can be determined whether the planting parameters need to be adjusted. Based on the database, the most suitable planting parameters are obtained through big data analysis and calculation in the cloud, which can effectively improve the level of intelligent agricultural planting and meet the needs of open-air planting.

[0105] Example 2

[0106] The reference numerals appearing in the drawings of the embodiments of the present invention include: a processing module 1 , a sensor network 2 , a data terminal 3 , and a control device 4 .

[0107] The embodiment of the present invention also provides a planting management system for smart agriculture, which is used to implement a planting management method for smart agriculture provided in the above embodiment. Figure 2 , including a processing module 1 and a sensor network 2, a data terminal 3, and a control device 4 connected to the processing module 1;

[0108] Sensor network 2 is used to collect monitoring data of open-air planting areas;

[0109] In this embodiment, the sensor network 2 adopts a distributed Internet of Things, which includes data-connected sensor nodes and gateway devices; the processing module 1 is connected to a number of gateway devices, and the gateway device is connected to a number of sensor nodes.

[0110] Specifically, the sensor nodes include soil nitrogen, phosphorus and potassium sensors, temperature sensors, moisture sensors, drone-mounted cameras, fixed cameras, light sensors and wind sensors. According to a certain area, such as 10 mu, one or more of the above sensors are set up in the area. The drone-mounted camera can collect images of multiple areas at the same time due to its mobile monitoring capability. It is not necessary to configure a drone for each area. After the node is set up, the sensors in the area are connected through a gateway device. The gateway device is responsible for transmitting the monitoring data in the area and marking the monitoring data during the transmission process, marking the area label and monitoring time.

[0111] This embodiment applies the Internet of Things technology to traditional agriculture, uses sensors and software to control agricultural production through a mobile platform or a computer platform, and makes traditional agriculture more "intelligent".

[0112] The processing module 1 is used to upload the monitoring data to the data terminal 3 .

[0113] After the sensor network 2 is set up, the monitoring data is transmitted to the processing module 1. The processing module 1 uses a single chip microcomputer, on which a wireless communication network interface is set to wirelessly receive data from the gateway device and transmit the data from the gateway device to the data terminal 3.

[0114] The data terminal 3 is used to obtain the monitoring data of the planting area and store it in the XML data storage architecture to obtain a planting database; it is also used to retrieve relevant stored data from the planting database according to the plant category, substitute it into the preset training model, and obtain the initial control parameters and the later control parameters.

[0115] At the same time, the data terminal 3 can also be connected to a mobile terminal, through which the stored data can be viewed in terms of area, time and monitoring data, thereby realizing the display and traceability of data and facilitating analysis and remote monitoring and traceability by relevant personnel.

[0116] Processing module 1 is used to obtain the growth image of the current plant and identify it to determine whether it is in a growth state. If so, the corresponding later control parameters are called as target parameters. Otherwise, the corresponding initial control parameters are called as target parameters. It also obtains real-time monitoring data of the plant and compares it with the target parameters. If the comparison is inconsistent, a control signal is generated according to the target parameters.

[0117] The control signal in the processing module 1 can be provided periodically, for example, once every three days, and when the monitoring data reaches the interval, the signal control is stopped. At the same time, the control signal can be manually modified.

[0118] It should be noted that in the control signal, when the soil fertility is higher than the control interval, a soil fertility reduction control signal is generated. The soil fertility reduction control signal has the highest priority. After the soil fertility reduction is completed, the drainage equipment is kept turned on. After an interval of 1 day, the external environment is regulated through the control device 4 through other control signals.

[0119] The control device 4 is used to control the environment of the planting area according to the control signal.

[0120] In this embodiment, the regulating device 4 includes one or more of drip irrigation equipment, drainage equipment and fertilization equipment.

[0121] The planting management system provided in this embodiment adopts various modules to implement various steps in the planting management method, providing a hardware foundation for the planting management method and facilitating the implementation of the method.

[0122] The above embodiments are preferred implementation modes of the present invention, but the implementation modes of the present invention are not limited to the above embodiments. Any other changes, modifications, substitutions, combinations, and simplifications that do not deviate from the spirit and principles of the present invention should be equivalent replacement methods and are included in the protection scope of the present invention.

Claims

1. A planting management method for smart agriculture, characterized in that: Includes steps: S1. Acquire monitoring data of the planting area and store it using XML data storage architecture to obtain a planting database; S2. Retrieve relevant stored data from the planting database according to the plant category, substitute it into the preset training model, and obtain initial control parameters and later control parameters; S3, obtaining the growth image of the current plant and identifying it, determining whether it is in a growth state, if so, calling the corresponding later control parameter as the target parameter, otherwise calling the corresponding initial control parameter as the target parameter; S4, obtaining real-time monitoring data of the plant, and comparing the data with the target parameters, and if the comparison is inconsistent, adjusting the planting parameters of the plant according to the target parameters; The step S2 comprises the steps of: S21, based on the current planted plant, obtaining the corresponding stored data from the planting database, screening the stored data of the approximate growth environment through keyword retrieval, performing multiple regression analysis to obtain a relevant regression function, and calculating the corresponding initial control parameters according to the relevant regression function and the current environmental parameters; S22. Based on the current plant, the corresponding stored data is obtained from the planting database, the stored data of the approximate growth environment is obtained through keyword retrieval, and the corresponding later control parameters are obtained by weighted average calculation.

2. A planting management method for smart agriculture as claimed in claim 1, characterized in that: The step S1 comprises the steps of: S11, collecting data from the open-air planting area of ​​each type of plant through a sensor network to obtain monitoring data; S12, marking the corresponding area label and monitoring time for the monitoring data; S13, uploading the labeled monitoring data to the cloud, and storing it in an XML data storage architecture to obtain a planting database for each type of plant; The monitoring data includes at least one or more of soil data, temperature data, moisture data, image data and weather data.

3. A planting management method for smart agriculture as claimed in claim 2, characterized in that: The step S21 comprises the steps of: A1. Based on the current planting plant, match the corresponding planting database; A2. Perform keyword search on the planting database according to the first search condition to obtain the stored data that is most similar to the growth environment of the area where the current plant belongs; A3. Obtain the stored data in the cultivation stage and remove outliers; A4, using the soil data and moisture data in the stored data as dependent variables, and the weather data and temperature data as independent variables to perform multiple regression analysis to obtain a relevant regression function; A5. Substituting the current environmental parameters of the region into the relevant regression function to obtain initial control parameters; Among them, the first search conditions are in the order of similar areas, different years, and close time periods; the current environmental parameters include weather data and temperature data, and the initial control parameters include target soil data and target moisture data.

4. A planting management method for smart agriculture as claimed in claim 2, characterized in that: The step S22 comprises the steps of: B1. Based on the current planting plants, match the corresponding planting database; B2. Perform keyword search on the planting database according to the second search condition to obtain the stored data that is most similar to the growth environment of the area where the currently planted plant belongs; B3, performing image recognition on each growth image in the stored data and scoring the growth status; B4, obtaining several groups of stored data with higher scores, performing weighted average calculation, and obtaining predicted data; B5, obtaining weather data and temperature data in a time period close to the predicted data, determining whether there is abnormal weather, and if so, discarding the stored data corresponding to the abnormal weather, and returning to step B2 to regenerate the predicted data as the later control parameter; The second search conditions are, in order, similar regions, same year, and similar time periods; The post-regulation parameters include target soil data and target moisture data.

5. A planting management method for smart agriculture as claimed in claim 4, characterized in that: In step B5, the abnormal weather is determined by obtaining weather data and temperature data, drawing a corresponding change curve, and performing differential calculation on the change curve. If the differential calculation value is greater than a preset threshold, it is considered that the external environment has changed dramatically and abnormal weather exists in the corresponding time period.

6. The planting management method of smart agriculture according to claim 1, characterized in that: The step S4 comprises the steps of: S41, dividing the range by taking the target parameter as the midpoint value of the interval and 10% of the midpoint value as the interval length to generate a control interval; S42. Obtain real-time monitoring data of the planted plants to determine whether they are within the control range. If so, do nothing. Otherwise, calculate the difference between the real-time monitoring data and the closest boundary value of the control range, and generate a control signal based on the difference to control the corresponding control equipment to perform environmental control on the planting area of ​​the planted plants.

7. A planting management system for smart agriculture, used to implement a planting management method for smart agriculture as claimed in any one of claims 1 to 6, characterized in that: It includes processing modules and sensor networks, data terminals, and control equipment connected to the modules; The sensor network is used to collect monitoring data of open-air planting areas; The processing module is used to upload the monitoring data to the data terminal; The data terminal is used to obtain monitoring data of the planting area and store it in an XML data storage architecture to obtain a planting database; It is also used to retrieve relevant stored data from the planting database according to the plant category, substitute it into the preset training model, and obtain initial control parameters and later control parameters; The processing module is used to obtain the growth image of the current plant and identify it, determine whether it is in a growth state, and if so, call the corresponding later control parameter as the target parameter, otherwise call the corresponding initial control parameter as the target parameter; and also obtain the real-time monitoring data of the plant and compare it with the target parameter, and if the comparison is inconsistent, generate a control signal according to the target parameter; The control device is used to perform environmental control on the planting area according to the control signal.

8. A planting management method for smart agriculture as claimed in claim 7, characterized in that: The regulating equipment includes one or more of drip irrigation equipment, drainage equipment and fertilization equipment.

9. The planting management system for smart agriculture according to claim 7, characterized in that: The sensor network adopts a distributed Internet of Things, which includes data-connected sensor nodes and gateway devices; the processing module is connected to a number of the gateway devices, and the gateway device is connected to a number of the sensor nodes.

Images