A method, system, device and medium for automatically monitoring farmland

Abstract

The present application discloses a method, system, device and medium for automatic monitoring of farmland, which mainly relates to the field of farmland monitoring technology, and is used to solve the problem that the existing solutions cannot effectively analyze the growth of crops in the farmland, and cannot provide data analysis support for the current farmland cultivation. It includes: determining whether there is abnormal data through the relationship between the real-time farmland data and the preset threshold; obtaining the weather information corresponding to the monitored farmland, and determining the growth of the crops in the current monitored farmland through the first collected picture and the second collected picture; inputting the weather information, soil moisture, soil nitrogen, phosphorus and potassium content and growth status into the trained neural network to obtain output recommendation information; wherein the training data of the neural network is weather information, soil moisture, soil nitrogen, phosphorus and potassium content, growth status and preset recommendation information; when it is checked that the actual farmland picture corresponds to the crop harvest, generating recommended crops for planting according to the current farmland real-time data.

Description

Technical Field

[0001] The present application relates to the field of farmland monitoring technology, and in particular to a method, system, equipment and medium for automatic monitoring of farmland. Background Art

[0002] In traditional agricultural production, the condition of crops in the farmland requires farmers to go down to the farmland for real-time observation in order to obtain the actual status of the crops. However, when farmers are observing, they can only observe the condition of crops around the farmland, and cannot observe the condition of crops in the middle of the farmland. This is mainly because the crops fill the entire farmland and farmers cannot smoothly enter the farmland for observation. At the same time, with the promotion of large-scale integrated modern agriculture and the promotion of large-scale contracted farmland, the traditional method of farmers going down to the fields for observation requires a waste of a lot of manpower and material resources, and the operation is cumbersome, and it is not convenient to record and uniformly manage.

[0003] Existing automatic farmland monitoring solutions are mainly for monitoring the growth process. For example, a method for monitoring the growth of farmland crops based on drones mainly monitors the growth of crops through directional cruising of drones and fixed-point shooting of imaging equipment carried by drones.

[0004] However, existing monitoring solutions mainly monitor the growth process of crops in farmland, and are unable to effectively analyze the growth conditions of crops in farmland, and are unable to provide data analysis support for current farmland cultivation. Summary of the invention

[0005] In view of the above-mentioned deficiencies in the prior art, the present application provides a method, system, equipment and medium for automatic monitoring of farmland to solve the problem that the existing solutions cannot effectively analyze the growth conditions of crops in the farmland and cannot provide data analysis support for current farmland cultivation.

[0006] In a first aspect, the present application provides a method for automatic monitoring of farmland, the method comprising: Based on the preset monitoring equipment placed in the monitored farmland, real-time farmland data is collected; wherein the preset monitoring equipment includes at least: a pest and disease detector, an insect trap, a soil pH meter, a soil moisture sensor, a soil nitrogen, phosphorus and potassium content tester, and a monitoring camera; the real-time farmland data includes at least: whether there are pests and diseases, insect trapping increment, soil pH value, soil moisture, soil nitrogen, phosphorus and potassium content, and a first collected picture; the second collected picture is collected by a drone; whether there is abnormal data is determined based on the relationship between the real-time farmland data and the preset threshold; when abnormal data exists, the corresponding processing terminal is determined according to the specific content corresponding to the abnormal data; the weather information corresponding to the monitored farmland is obtained, and the growth of the crops in the current monitored farmland is determined through the first collected picture and the second collected picture; the weather information, soil moisture, soil nitrogen, phosphorus and potassium content and growth status are input into the trained neural network to obtain output recommendation information; wherein the training data of the neural network is weather information, soil moisture, soil nitrogen, phosphorus and potassium content, growth status and preset recommendation information; when it is checked that the actual picture of the farmland corresponds to the harvest of crops, the recommended crops are generated according to the current real-time farmland data.

[0007] The automatic monitoring method for farmland provided in the embodiment of the present application integrates a variety of monitoring equipment (such as pest and disease detectors, soil pH meters, etc.), and the scheme can collect farmland real-time data from all directions and angles to ensure the comprehensiveness and accuracy of the data. The second collected picture is also collected by drone, combined with the first collected picture, and the real-time data is compared with the preset threshold to determine whether there is abnormal data. It is helpful to timely discover abnormal conditions in farmland and provide a basis for timely measures. By inputting data such as weather information, soil moisture, soil nitrogen, phosphorus and potassium content and growth conditions into a trained neural network, it is possible to output suggestion information and provide intelligent decision support for farmland management. It helps farmers adjust management measures according to the current farmland conditions. When it is detected that the actual picture of the farmland corresponds to the harvest of crops, the present application can generate recommended crops for planting according to the current real-time data of the farmland. It helps farmers to formulate more reasonable planting plans according to the actual situation of the farmland to avoid waste of resources and yield reduction caused by blind planting. It solves the problem that the existing scheme cannot effectively analyze the growth of crops in the farmland and cannot provide data analysis support for the current farmland cultivation.

[0008] In one implementation of the present application, the method also includes: equipping the four corners of the monitored farmland with monitoring cameras, and obtaining multiple groups of first collected pictures according to the offset of the monitoring camera angles; obtaining second collected pictures through a camera device carried on a drone; using computer vision technology to calculate and obtain three-dimensional data of crops in the first collected pictures and the second collected pictures; wherein the three-dimensional data includes at least: the specific position, height, color, and shape of the crops; applying digital twin technology to the three-dimensional data to present a three-dimensional image and realize remote modeling.

[0009] In one implementation of the present application, the growth status of crops in the currently monitored farmland is determined by using the first collected picture and the second collected picture, specifically including: Identifying the variety of crops in the currently monitored farmland through the first collected picture or the second collected picture; Obtain a growth cycle change table corresponding to the current crop variety; wherein the growth cycle change table includes at least: the corresponding relationship between the growth condition, the height and the color of the crop; use computer vision technology to calculate the height and color of the crop in the first collected image and the second collected image; determine the growth condition of the crops in the currently monitored farmland based on the growth cycle change table, height and color; wherein the growth condition includes poor development, normal development and excessive growth.

[0010] In one implementation of the present application, determining whether there is abnormal data is based on the relationship between the real-time farmland data and the preset threshold, specifically including: When it is determined that there are pests and diseases, or the insect trapping increment exceeds the increment threshold, or the soil pH value is not within the preset pH value range, or the soil moisture is not within the preset moisture range, or the soil nitrogen, phosphorus and potassium content is not within the preset normal range, it is determined that there is abnormal data; Identify the variety and growth cycle of the crops in the first collected image; determine a corresponding trained pest and disease recognition algorithm according to the variety and growth cycle of the crops, and then use the first collected image as an input of the corresponding trained pest and disease recognition algorithm to obtain a recognition result; The training process of the pest and disease identification algorithm is as follows: Collect sample images of several crops, and divide the sample images into several sets according to the varieties and growth cycles of the crops; The sample images and preset pest and disease identifications in the set are used as sample data to train the preset image recognition algorithm to obtain the trained pest and disease identification algorithms corresponding to each set.

[0011] In one implementation of the present application, before using the real-time farmland data based on the preset monitoring equipment placed in the monitoring farmland, the method further includes: Obtain the total area of ​​the monitored farmland; divide the total area into several sub-areas based on the monitoring range of the preset monitoring equipment, determine the central geographical location of each sub-area as the placement point of the current preset monitoring equipment, and complete the placement of the preset monitoring equipment through the drone; wherein the preset monitoring equipment is divided into at least: pest and disease detector, insect trap, soil pH meter, soil moisture sensor, soil nitrogen, phosphorus and potassium content tester.

[0012] In a second aspect, the present application provides a farmland automatic monitoring system, the system comprising: The collection module is used to collect real-time data of farmland based on preset monitoring equipment placed in the monitored farmland; wherein the preset monitoring equipment at least includes: a pest detector, an insect trap, a soil pH meter, a soil moisture sensor, a soil nitrogen, phosphorus and potassium content tester, and a monitoring camera; the real-time data of farmland at least includes: whether there are pests and diseases, insect trap increment, soil pH value, soil moisture, soil nitrogen, phosphorus and potassium content, and a first collected picture; the second collected picture is collected by a drone; The central processing module is used to determine whether there is abnormal data through the relationship between the real-time farmland data and the preset threshold; when there is abnormal data, determine the corresponding processing terminal according to the specific content corresponding to the abnormal data; obtain the weather information corresponding to the monitored farmland, and determine the growth status of the crops in the current monitored farmland through the first collected picture and the second collected picture; input the weather information, soil moisture, soil nitrogen, phosphorus and potassium content and growth status into the trained neural network to obtain output recommendation information; wherein the training data of the neural network is weather information, soil moisture, soil nitrogen, phosphorus and potassium content, growth status and preset recommendation information; when it is checked that the actual picture of the farmland corresponds to the crop harvest, generate recommended crops for planting according to the current real-time data of the farmland.

[0013] In one implementation of the present application, the central processing module includes a growth identification unit, which is used to identify the variety of crops in the currently monitored farmland through the first collected picture or the second collected picture; obtain the growth cycle change table corresponding to the current crop variety; wherein the growth cycle change table at least includes: the correspondence between the growth condition, the height and the color of the crop; use computer vision technology to calculate the height and color of the crop in the first collected picture and the second collected picture; determine the growth condition of the crops in the currently monitored farmland based on the growth cycle change table, height, and color; wherein the growth condition includes poor development, normal development, and excessive growth.

[0014] In one implementation of the present application, the system also includes a location determination module for obtaining the total area of ​​the monitored farmland; based on the monitoring range of the preset monitoring equipment, the total area is divided into several sub-areas, and the central geographical location of each sub-area is determined as the placement point of the current preset monitoring equipment, and the placement of the preset monitoring equipment is completed by a drone; wherein the preset monitoring equipment is divided into at least: pest and disease detector, insect trap, soil pH meter, soil moisture sensor, soil nitrogen, phosphorus and potassium content tester.

[0015] In a third aspect, the present application provides an automatic farmland monitoring device, the device comprising: processor; and a memory having executable codes stored thereon, which, when executed, causes the processor to execute any of the above-mentioned methods for automatically monitoring farmland.

[0016] In a fourth aspect, the present application provides a non-volatile computer storage medium having computer instructions stored thereon, which when executed implement an automatic farmland monitoring method as described above.

[0017] Those skilled in the art can understand that the present application has at least the following beneficial effects: The present application discloses a method, system, equipment and medium for automatic monitoring of farmland. By integrating a variety of monitoring equipment (such as pest and disease detectors, soil pH meters, etc.), the scheme can collect real-time farmland data from all directions and angles to ensure the comprehensiveness and accuracy of the data. The second collected picture is also collected by drone, combined with the first collected picture, and the real-time data is compared with the preset threshold to determine whether there is abnormal data. It is helpful to timely discover abnormal conditions in farmland and provide a basis for timely measures. By inputting data such as weather information, soil moisture, soil nitrogen, phosphorus and potassium content and growth conditions into a trained neural network, it is possible to output suggestion information and provide intelligent decision support for farmland management. It helps farmers adjust management measures according to the current farmland conditions. When it is detected that the actual picture of the farmland corresponds to the harvest of crops, the present application can generate recommended crops for planting according to the current real-time data of the farmland. It helps farmers to formulate more reasonable planting plans according to the actual situation of the farmland to avoid waste of resources and yield reduction caused by blind planting. It solves the problem that the existing scheme cannot effectively analyze the growth of crops in the farmland and cannot provide data analysis support for current farmland cultivation. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Some embodiments of the present disclosure are described below with reference to the accompanying drawings, in which: Figure 1 This is a flow chart of an automatic farmland monitoring method provided in an embodiment of the present application.

[0019] Figure 2 It is a schematic diagram of the internal structure of an automatic farmland monitoring system provided in an embodiment of the present application.

[0020] Figure 3 It is a schematic diagram of the internal structure of an automatic farmland monitoring device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0021] It should be understood by those skilled in the art that the embodiments described below are only preferred embodiments of the present disclosure, and do not mean that the present disclosure can only be implemented through the preferred embodiments. The preferred embodiments are only used to explain the technical principles of the present disclosure, and are not used to limit the protection scope of the present disclosure. Based on the preferred embodiments provided by the present disclosure, all other embodiments obtained by ordinary technicians in this field without creative work should still fall within the protection scope of the present disclosure.

[0022] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0023] 1. In order to clearly describe the technical solutions of the embodiments of the present application, some of the terms and technologies involved in the embodiments of the present application are briefly introduced below: Machine Learning: Machine learning is a common research hotspot in the fields of artificial intelligence and pattern recognition. Its theories and methods have been widely used to solve complex problems in engineering applications and scientific fields. Machine learning enables computers to learn from research data and statistical information.

[0024] Insect Trap: A product that uses sex pheromones to trap and control pests.

[0025] Pest and disease detector: Pest and disease detector is a pest and disease detection instrument that helps people quickly and accurately diagnose the source of plant disease in the early stage of plant disease. It can detect dozens of diseases caused by bacteria, fungi and viruses, thereby strengthening the use of drugs for diseased plants, reducing drug waste and pollution, and providing guidance for drug dispensing. It is a convenient and practical instrument for preventing and controlling pests and diseases and assisting in scientific drug use.

[0026] Digital twin: Digital twin is to make full use of physical models, sensors, operation history and other data, integrate multi-disciplinary, multi-physical quantity, multi-scale, multi-probability simulation processes, complete mapping in virtual space, and thus reflect the entire life cycle of the corresponding physical equipment. Digital twin is a concept that transcends reality and can be regarded as a digital mapping system of one or more important, interdependent equipment systems.

[0027] Computer vision: Computer vision is a science that studies how to make machines "see". It refers to using cameras and computers to replace human eyes to identify, track and measure targets, and further perform graphic processing so that the computer processes the images into images that are more suitable for human eye observation or transmission to instruments for detection.

[0028] The technical solution proposed in the embodiments of the present application is described in detail below with reference to the accompanying drawings.

[0029] The embodiment provides a method for automatic monitoring of farmland, such as Figure 1 As shown, the method provided in the embodiment of the present application mainly includes the following steps: Step 110: Based on the preset monitoring equipment placed in the monitored farmland, real-time data of the farmland is collected; and a second collected picture is collected by a drone.

[0030] It should be noted that the preset monitoring equipment includes at least: pest and disease detector, insect trap, soil pH meter, soil moisture sensor, soil nitrogen, phosphorus and potassium content tester, monitoring camera, and the real-time data of farmland includes at least: whether there are pests and diseases, insect trap increment, soil pH value, soil moisture, soil nitrogen, phosphorus and potassium content, and the first collected picture.

[0031] In addition, the present application can determine the specific placement location based on the monitoring range of the preset monitoring device itself. Furthermore, due to the large farmland, the placement location has the problem of inaccuracy. The present application can use drones to place the preset monitoring device at a specific location, and then manually adjust it. The specific implementation process can be: Obtain the total area of ​​the monitored farmland; divide the total area into several sub-areas based on the monitoring range of the preset monitoring equipment, determine the central geographical location of each sub-area as the placement point of the current preset monitoring equipment, and complete the placement of the preset monitoring equipment through the drone; wherein the preset monitoring equipment is divided into at least: pest and disease detector, insect trap, soil pH meter, soil moisture sensor, soil nitrogen, phosphorus and potassium content tester.

[0032] As an example, the monitoring range of the pest detector is the first sub-range: 2r*π (where r is the monitoring length), and the total area is divided into several sub-areas r*r. In addition, the center point of this area is determined as the placement point of the pest detector. According to the relationship between the total area and the sub-area, the longitude and latitude of each center point can be obtained, and then the longitude and latitude are transmitted to the drone. Through the GPS positioning device in the drone, the pest detector can be placed more accurately at the placement point.

[0033] Step 120: determine whether there is abnormal data based on the relationship between the real-time farmland data and the preset threshold; when there is abnormal data, determine the corresponding processing terminal according to the specific content corresponding to the abnormal data.

[0034] It should be noted that different types of farmland real-time data correspond to different preset thresholds. Those skilled in the art can determine the specific content corresponding to the preset threshold according to actual needs. In addition, if one of the data exceeds the preset threshold, the host will issue a red warning; if one of the data is close to the preset threshold or reaches the preset threshold, the host will issue a yellow warning. In addition, nutrients can be analyzed based on the current crops planted, and when the soil nutrients are insufficient, corresponding fertilization suggestions can be given; when the soil pH is abnormal, preset suggestions for adjusting the abnormal soil pH are given.

[0035] Among them, whether there is abnormal data is determined by the relationship between the real-time farmland data and the preset threshold, which can be specifically: When it is determined that there are pests and diseases, or the insect trapping increment exceeds the increment threshold, or the soil pH value is not within the preset pH value range, or the soil moisture is not within the preset moisture range, or the soil nitrogen, phosphorus and potassium content is not within the preset normal range, it is determined that there is abnormal data; the variety and growth cycle of the crops in the first collected image are identified; according to the variety and growth cycle of the crops, a corresponding trained pest and disease recognition algorithm is determined, and the first collected image is used as the input of the corresponding trained pest and disease recognition algorithm to obtain a recognition result.

[0036] The training process of the above pest and disease identification algorithm can be specifically as follows: Sample images of several crops are collected, and the sample images are divided into several sets according to the varieties and growth cycles of the crops. The sample images in the sets and preset pest and disease identifications are used as sample data to train a preset image recognition algorithm to obtain a trained pest and disease identification algorithm corresponding to each set.

[0037] Those skilled in the art will appreciate that by carefully classifying sample images, the trained pest and disease identification algorithm can further improve processing accuracy.

[0038] Step 130: Acquire weather information corresponding to the monitored farmland, and determine the growth conditions of crops in the currently monitored farmland through the first collected picture and the second collected picture.

[0039] It should be noted that weather information may at least include: rainfall, wind speed, temperature, air humidity, etc. Weather forecast information (weather information) may be obtained through the local meteorological bureau.

[0040] The growth conditions of the crops in the currently monitored farmland are determined by using the first collected picture and the second collected picture, which may be specifically: Identify the variety of crops in the currently monitored farmland through the first collected picture or the second collected picture; obtain the growth cycle change table corresponding to the current crop variety; wherein the growth cycle change table at least includes: the corresponding relationship between the growth condition, the height and the color of the crop; use computer vision technology to calculate the height and color of the crops in the first collected picture and the second collected picture; determine the growth condition of the crops in the currently monitored farmland based on the growth cycle change table, height and color; wherein the growth condition includes poor development, normal development and excessive growth.

[0041] Step 140: input weather information, soil moisture, soil nitrogen, phosphorus and potassium content and growth conditions into the trained neural network to obtain output recommendation information; when it is checked that the actual picture of the farmland corresponds to the crop harvest, generate recommended crops for planting based on the current real-time data of the farmland.

[0042] It should be noted that the training data of the neural network are weather information, soil moisture, soil nitrogen, phosphorus and potassium content, growth conditions and preset recommended information. In addition, different crops correspond to different farmland real-time data requirements. By comparing the similarity between the current farmland real-time data and the preset farmland real-time data of the crop, the crop with the greatest similarity can be determined as the recommended crop.

[0043] In addition, the present application can convert the current crop state into a three-dimensional image. The specific conversion process can be: Monitoring cameras are installed at the four corners of the monitored farmland, and multiple sets of first collected pictures are obtained according to the offset of the monitoring camera angles; and second collected pictures are obtained through the camera device carried by the drone; Computer vision technology is used to calculate and obtain three-dimensional data of crops in the first collected picture and the second collected picture; wherein the three-dimensional data at least includes: the specific position, height, color, and shape of the crops; the three-dimensional data is applied with digital twin technology to present a three-dimensional image and realize remote modeling.

[0044] In addition, this application Figure 2 An automatic farmland monitoring system is provided in the embodiment of the present application. Figure 2 As shown, the system provided in the embodiment of the present application mainly includes: The collection module 210 is used to collect real-time data of the farmland based on the preset monitoring equipment placed in the monitored farmland; wherein the preset monitoring equipment includes at least: a pest and disease detector, an insect trap, a soil pH meter, a soil moisture sensor, a soil nitrogen, phosphorus and potassium content tester, and a monitoring camera; the real-time data of the farmland includes at least: whether there are pests and diseases, insect trapping increments, soil pH value, soil moisture, soil nitrogen, phosphorus and potassium content, and a first collected picture; the second collected picture is collected by a drone.

[0045] The central processing module 220 is used to determine whether there is abnormal data through the relationship between the real-time farmland data and the preset threshold; when abnormal data exists, determine the corresponding processing terminal according to the specific content corresponding to the abnormal data; obtain the weather information corresponding to the monitored farmland, and determine the growth status of the crops in the current monitored farmland through the first collected picture and the second collected picture; input the weather information, soil moisture, soil nitrogen, phosphorus and potassium content and growth status into the trained neural network to obtain output recommendation information; wherein the training data of the neural network is weather information, soil moisture, soil nitrogen, phosphorus and potassium content, growth status and preset recommendation information; when it is checked that the actual picture of the farmland corresponds to the crop harvest, generate recommended crops for planting according to the current real-time data of the farmland.

[0046] The central processing module 220 includes a growth identification unit, which is used to identify the variety of crops in the currently monitored farmland through the first collected picture or the second collected picture; obtain the growth cycle change table corresponding to the current crop variety; wherein the growth cycle change table at least includes: the correspondence between the growth condition, the height and the color of the crop; use computer vision technology to calculate the height and color of the crop in the first collected picture and the second collected picture; determine the growth condition of the crops in the currently monitored farmland based on the growth cycle change table, height, and color; wherein the growth condition includes poor development, normal development, and excessive growth.

[0047] The system also includes a location determination module for obtaining the total area of ​​the monitored farmland; based on the monitoring range of the preset monitoring equipment, the total area is divided into several sub-areas, the central geographical location of each sub-area is determined as the placement point of the current preset monitoring equipment, and the placement of the preset monitoring equipment is completed by the drone; wherein the preset monitoring equipment is divided into at least: pest and disease detector, insect trap, soil pH meter, soil moisture sensor, soil nitrogen, phosphorus and potassium content tester.

[0048] The above is a method embodiment of the present application. Based on the same inventive concept, the present application embodiment also provides an automatic farmland monitoring device. Figure 3 As shown, the device includes: a processor; and a memory on which executable codes are stored. When the executable codes are executed, the processor executes an automatic farmland monitoring method as described in the above-mentioned embodiment.

[0049] Specifically, the server side obtains preset monitoring equipment placed in the monitored farmland and collects real-time farmland data; wherein the preset monitoring equipment at least includes: a pest and disease detector, an insect trap, a soil pH meter, a soil moisture sensor, a soil nitrogen, phosphorus and potassium content tester, and a monitoring camera, and the real-time farmland data at least includes: whether there are pests and diseases, insect trapping increment, soil pH value, soil moisture, soil nitrogen, phosphorus and potassium content, and a first collected picture; the second collected picture is collected by a drone; whether there is abnormal data is determined based on the relationship between the real-time farmland data and the preset threshold; when abnormal data exists, the corresponding processing terminal is determined according to the specific content corresponding to the abnormal data; the weather information corresponding to the monitored farmland is obtained, and the growth status of the crops in the current monitored farmland is determined through the first collected picture and the second collected picture; the weather information, soil moisture, soil nitrogen, phosphorus and potassium content and growth status are input into the trained neural network to obtain output recommendation information; wherein the training data of the neural network is weather information, soil moisture, soil nitrogen, phosphorus and potassium content, growth status and preset recommendation information; when it is checked that the actual picture of the farmland corresponds to the harvest of crops, the recommended crops are generated according to the current real-time data of the farmland.

[0050] In addition, an embodiment of the present application further provides a non-volatile computer storage medium on which executable instructions are stored. When the executable instructions are executed, an automatic farmland monitoring method as described above is implemented.

[0051] So far, the technical solutions of the present disclosure have been described in combination with the above multiple embodiments, but it is easy for those skilled in the art to understand that the protection scope of the present disclosure is not limited to these specific embodiments. Without departing from the technical principles of the present disclosure, those skilled in the art can split and combine the technical solutions in the above-mentioned various embodiments, and can also make equivalent changes or replacements to the relevant technical features. Any changes, equivalent replacements, improvements, etc. made within the technical concept and / or technical principle of the present disclosure will fall within the protection scope of the present disclosure.

Claims

1. A method for automatic monitoring of farmland, characterized in that: The method comprises: Based on the preset monitoring equipment placed in the monitored farmland, real-time farmland data is collected; wherein the preset monitoring equipment includes at least: a pest detector, an insect trap, a soil pH meter, a soil moisture sensor, a soil nitrogen, phosphorus and potassium content tester, and a monitoring camera; the real-time farmland data includes at least: whether there are pests and diseases, insect trap increment, soil pH value, soil moisture, soil nitrogen, phosphorus and potassium content, and a first collected picture; the second collected picture is collected by a drone; Determine whether there is abnormal data through the relationship between the real-time farmland data and the preset threshold; when there is abnormal data, determine the corresponding processing terminal according to the specific content corresponding to the abnormal data; Acquire weather information corresponding to the monitored farmland, and determine the growth conditions of crops in the currently monitored farmland through the first collected picture and the second collected picture; Weather information, soil moisture, soil nitrogen, phosphorus and potassium content, and growth conditions are input into the trained neural network to obtain output recommendation information; the training data of the neural network are weather information, soil moisture, soil nitrogen, phosphorus and potassium content, growth conditions, and preset recommendation information; when it is checked that the actual picture of the farmland corresponds to the crop harvest, the recommended crops are generated based on the current real-time data of the farmland.

2. The automatic farmland monitoring method according to claim 1, characterized in that: The method further comprises: Monitoring cameras are installed at the four corners of the monitored farmland, and multiple sets of first collected pictures are obtained according to the offset of the monitoring camera angles; and second collected pictures are obtained through the camera device carried by the drone; Computer vision technology is used to calculate and obtain three-dimensional data of crops in the first collected picture and the second collected picture; wherein the three-dimensional data at least includes: the specific position, height, color, and shape of the crops; the three-dimensional data is applied with digital twin technology to present a three-dimensional image and realize remote modeling.

3. The automatic farmland monitoring method according to claim 1 or 2, characterized in that: Determining the growth status of crops in the currently monitored farmland through the first collected picture and the second collected picture specifically includes: Identifying the variety of crops in the currently monitored farmland through the first collected picture or the second collected picture; Obtaining a growth cycle change table corresponding to the current crop variety; wherein the growth cycle change table at least includes: a corresponding relationship between growth conditions, crop height and color; Using computer vision technology, calculate and obtain the height and color of the crops in the first collected image and the second collected image; The growth status of the crops in the currently monitored farmland is determined based on the growth cycle change table, height, and color; the growth status includes poor development, normal development, and excessive growth.

4. The automatic farmland monitoring method according to claim 1, characterized in that: Through the relationship between the real-time farmland data and the preset threshold, determine whether there is abnormal data, including: When it is determined that there are pests and diseases, or the insect trapping increment exceeds the increment threshold, or the soil pH value is not within the preset pH value range, or the soil moisture is not within the preset moisture range, or the soil nitrogen, phosphorus and potassium content is not within the preset normal range, it is determined that there is abnormal data; Identify the variety and growth cycle of the crops in the first collected image; determine a corresponding trained pest and disease recognition algorithm according to the variety and growth cycle of the crops, and then use the first collected image as an input of the corresponding trained pest and disease recognition algorithm to obtain a recognition result; The training process of the pest and disease identification algorithm is as follows: Collect sample images of several crops, and divide the sample images into several sets according to the varieties and growth cycles of the crops; The sample images and preset pest and disease identifications in the set are used as sample data to train the preset image recognition algorithm to obtain the trained pest and disease identification algorithms corresponding to each set.

5. The automatic farmland monitoring method according to claim 1, characterized in that: Before using the real-time farmland data based on the preset monitoring equipment placed in the monitored farmland, the method further includes: Obtain the total area of ​​the monitored farmland; divide the total area into several sub-areas based on the monitoring range of the preset monitoring equipment, determine the central geographical location of each sub-area as the placement point of the current preset monitoring equipment, and complete the placement of the preset monitoring equipment through the drone; wherein the preset monitoring equipment is divided into at least: pest and disease detector, insect trap, soil pH meter, soil moisture sensor, soil nitrogen, phosphorus and potassium content tester.

6. A farmland automatic monitoring system, characterized in that: The system comprises: The collection module is used to collect real-time data of farmland based on preset monitoring equipment placed in the monitored farmland; wherein the preset monitoring equipment at least includes: a pest detector, an insect trap, a soil pH meter, a soil moisture sensor, a soil nitrogen, phosphorus and potassium content tester, and a monitoring camera; the real-time data of farmland at least includes: whether there are pests and diseases, insect trap increment, soil pH value, soil moisture, soil nitrogen, phosphorus and potassium content, and a first collected picture; the second collected picture is collected by a drone; The central processing module is used to determine whether there is abnormal data through the relationship between the real-time farmland data and the preset threshold; when there is abnormal data, determine the corresponding processing terminal according to the specific content corresponding to the abnormal data; obtain the weather information corresponding to the monitored farmland, and determine the growth status of the crops in the current monitored farmland through the first collected picture and the second collected picture; input the weather information, soil moisture, soil nitrogen, phosphorus and potassium content and growth status into the trained neural network to obtain output recommendation information; wherein the training data of the neural network is weather information, soil moisture, soil nitrogen, phosphorus and potassium content, growth status and preset recommendation information; when it is checked that the actual picture of the farmland corresponds to the crop harvest, generate recommended crops for planting according to the current real-time data of the farmland.

7. The automatic farmland monitoring system according to claim 6, characterized in that: The central processing module includes a growth recognition unit, Used to identify the variety of crops in the currently monitored farmland through the first collected picture or the second collected picture; Obtaining a growth cycle change table corresponding to the current crop variety; wherein the growth cycle change table at least includes: a corresponding relationship between growth conditions, crop height and color; Using computer vision technology, calculate and obtain the height and color of the crops in the first collected image and the second collected image; The growth status of the crops in the currently monitored farmland is determined based on the growth cycle change table, height, and color; the growth status includes poor development, normal development, and excessive growth.

8. The automatic farmland monitoring system according to claim 6, characterized in that: The system also includes a location determination module, Used to obtain the total area of ​​monitored farmland; Based on the monitoring range of the preset monitoring equipment, the total area is divided into several sub-areas, and the central geographical location of each sub-area is determined as the placement point of the current preset monitoring equipment, and the placement of the preset monitoring equipment is completed by drones; among them, the preset monitoring equipment is divided into at least: pest and disease detector, insect trap, soil pH meter, soil moisture sensor, soil nitrogen, phosphorus and potassium content tester.

9. An automatic farmland monitoring device, characterized in that: The device comprises: processor; and a memory having executable codes stored thereon, which, when the executable codes are executed, enable the processor to execute a method for automatic monitoring of farmland as described in any one of claims 1-5.

10. A non-volatile computer storage medium, characterized in that: Computer instructions are stored thereon, and when the computer instructions are executed, an automatic farmland monitoring method as described in any one of claims 1-5 is implemented.

Images