Crop disease recognition robot and crop disease recognition system

Abstract

The invention provides a crop disease identification robot and a crop disease identification system. The crop disease identification robot comprises a camera device, an electric signal acquisition device and a control device, wherein the camera device is used for shooting crop images of crops in a target area; the electric signal acquisition device is used for acquiring crop electric signal information of each crop in a target area; the control device is used for identifying the crop type of each crop in the target area according to the crop image; and determining a disease identification result of each crop in the target area according to the crop electric signal information of each crop, the crop image and the identification model corresponding to each crop type. The electric signals of the crops can be influenced by the growth states of the crops, so that the electric signals can correspondingly change when diseases and insect pests occur, and the disease recognition precision can be effectively improved by adding the electric signals of the crops in the image recognition process.

Description

Technical Field

[0001] The invention belongs to the technical field of crop growth monitoring, and in particular relates to a crop disease identification robot and a crop disease identification system. Background Art

[0002] During the growth process of crops, they are often infected by various diseases and pests due to the influence of factors such as weather, microorganisms, and growth environment, which affect the normal growth of crops and thus lead to reduced crop yields.

[0003] Current crop pest and disease detection usually involves inputting crop images into corresponding machine learning models for image recognition to achieve crop pest and disease detection. However, due to limitations in the clarity of captured images and recognition errors in machine learning, it is unable to accurately identify various pests and diseases, and the recognition accuracy is low. Summary of the invention

[0004] In view of this, the present invention provides a crop disease identification robot and a crop disease identification system, aiming to solve the problem of low identification accuracy of crop disease identification equipment in the prior art.

[0005] A first aspect of an embodiment of the present invention provides a crop disease identification robot, comprising:

[0006] A camera device, an electric signal acquisition device, and a control device; the camera device and the electric signal acquisition device are both connected to the control device;

[0007] The camera device is used to capture crop images of each crop in the target area;

[0008] The electric signal collection device is used to collect the crop electric signal information of each crop in the target area;

[0009] The control device is used to identify the crop type of each crop in the target area according to the crop image; and determine the disease identification result of each crop in the target area according to the crop electrical signal information of each crop, the crop image and the identification model corresponding to each crop type.

[0010] In some possible implementations, a closed collection chamber is provided in the crop disease identification robot, and the electrical signal collection device is provided in the closed collection chamber;

[0011] The control device is also used to determine the position of the leaves to be detected of each crop based on the crop images of each crop in the target area; place the leaves to be detected in a closed collection chamber, and instruct the electrical signal collection device to collect the electrical signals of the leaves to be detected as the crop electrical signal information of the crop to which the leaves to be detected belong.

[0012] In some possible implementations, an environment adjustment device is provided in the closed collection chamber; the crop electrical signal information includes a first electrical signal and a second electrical signal; the first electrical signal is an electrical signal measured when the leaf to be detected is in a first preset environment; the second electrical signal is an electrical signal measured when the leaf to be detected is in a second preset environment;

[0013] The control device is specifically used to: calculate the growth activity of each crop based on the first electrical signal and the second electrical signal of each crop; input the growth activity and crop image of each crop into the recognition model corresponding to each crop type to obtain the disease recognition results of each crop in the target area.

[0014] In some possible implementations, the control device is specifically used to:

[0015] According to the growth activity of each crop, the average growth activity of crops in the target area is calculated;

[0016] The difference between the growth activity and the mean growth activity was taken as the degree of crop weakness;

[0017] The crop weakness degree of each crop and the crop image of each crop are input into the recognition model corresponding to each crop type to obtain the disease recognition results of each crop in the target area.

[0018] In some possible implementations, the crop disease identification robot further includes a fruit gas detection device;

[0019] The fruit gas detection device is connected to the control device;

[0020] The control device is also used to determine whether the fruits of each crop in the target area are bagged according to the crop image, and when the fruits of each crop in the target area are bagged, instruct the fruit gas detection device to collect gas information of multiple fruits of each crop;

[0021] The control device is also used to input the growth activity of each crop, the crop image of each crop, and the gas information of multiple fruits of each crop into the recognition model corresponding to each crop type to obtain the disease recognition results of each crop in the target area.

[0022] In some possible implementations, the crop disease identification robot also includes a flying device.

[0023] In some possible implementations, the crop disease identification robot further includes a GPS module;

[0024] The control device is also used to determine the position to be detected of each crop based on the crop image of each crop; based on the position information and the position to be detected reported by the GPS module, a detection point closest to the robot is used as the starting point to obtain multiple paths, and then all paths are traversed to determine the path with the shortest total length as the optimal path.

[0025] In some possible implementations, the electrical signal acquisition device includes: a patch electrode and an acupuncture electrode.

[0026] In some possible implementations, the crop disease identification robot further includes a solar cell.

[0027] A second aspect of the embodiments of the present invention provides a crop disease identification system, comprising: a terminal and the crop disease identification robot according to the first aspect.

[0028] The crop disease identification robot and crop disease identification system provided by the embodiment of the present invention include a camera device, an electric signal acquisition device and a control device; wherein the camera device is used to capture crop images of each crop in the target area; the electric signal acquisition device is used to collect crop electric signal information of each crop in the target area; the control device is used to identify the crop type of each crop in the target area according to the crop image; and the disease identification results of each crop in the target area are determined according to the crop electric signal information of each crop, the crop image and the identification model corresponding to each crop type. The electric signal of the crop will be affected by its growth state, so its electric signal will change accordingly when pests and diseases occur. Therefore, by adding the electric signal of the crop in the process of image recognition, the accuracy of disease identification can be effectively improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0030] Figure 1 is a schematic diagram of the structure of a crop disease identification robot provided by an embodiment of the present invention;

[0031] Figure 2 Schematic diagram of the structure of a crop disease identification system provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0032] In the following description, specific details such as specific system structures, technologies, etc. are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present invention. However, it should be clear to those skilled in the art that the present invention may be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to prevent unnecessary details from obstructing the description of the present invention.

[0033] Figure 1 Schematic diagram of the structure of the crop disease identification robot provided by the embodiment of the present invention. Figure 1 As shown, in some embodiments, a crop disease identification robot includes:

[0034] Camera device 11, electric signal acquisition device 12, control device 13;

[0035] Among them, the camera device 11 and the electrical signal acquisition device 12 are both connected to the control device 13; the camera device 11 is used to capture crop images of each crop in the target area; the electrical signal acquisition device 12 is used to collect crop electrical signal information of each crop in the target area; the control device 13 is used to identify the crop type of each crop in the target area based on the crop image; based on the crop electrical signal information of each crop, the crop image and the recognition model corresponding to each crop type, the disease recognition result of each crop in the target area is determined.

[0036] In the embodiment of the present invention, the crop image captured by the camera device 11 may be a partial image of the crop or an overall image of the crop, which is not limited herein.

[0037] In the embodiment of the present invention, after the crop image is acquired, it needs to be preprocessed to remove the corresponding background and leave the image of the features such as stems and leaves. The recognition model can be a neural network model, a deep learning model, etc., which are not limited here. The control device 13 can be an MCU, a single chip microcomputer, etc., which are not limited here.

[0038] Due to the complexity of crop types, each crop has multiple diseases and pests, and direct identification is difficult to achieve. Therefore, an identification model can be set for each crop, that is, each identification model only identifies several diseases and pests of one crop.

[0039] The recognition model for each crop can be trained using transfer learning methods.

[0040] For example, when training a corn recognition model, multiple corn leaf images are required, including healthy leaves and leaves affected by pests and diseases (such as corn gray spot, corn stripe rust, and corn leaf spot). At the same time, the electrical signals of these healthy leaves and the electrical signals of the leaves affected by pests and diseases are collected. The type and degree of the disease need to be labeled before model training. The size of the leaf image can be uniformly set to 256×256 pixels.

[0041] The images and electrical signals of these leaves are used as input, and the types and degrees of labeled pests and diseases are used as output to form samples. The samples are set as training sets and test sets in a ratio of 4:1 to achieve the training of the recognition model.

[0042] The transfer learning ResNet model uses the weight file of the ResNet50 model pre-trained on ImageNet, retains the convolution part of ResNet50, and changes the multiple layers to improve the recognition accuracy; the transfer learning VGG16 model uses the model pre-trained on ImageNet to retain the weight file, removes the top three fully connected layers, and adds a custom fully connected layer to improve the recognition accuracy;

[0043] The specific parameters in the process of transfer learning and fine-tuning are: using the SGD+0.9 momentum optimization method, the batch size is 128, the learning rate is 0.0001, the number of ResNet iterations in the experiment is 50 times, and the number of VGG16 model iterations is 10 times.

[0044] In the embodiment of the present invention, after the initial electrical signal of the crop is collected, it is necessary to sample the one-dimensional initial electrical signal at equal intervals to obtain a discrete electrical signal sequence, and then decompose the discrete electrical signal sequence into a two-dimensional matrix composed of eigenstate functions according to the empirical mode decomposition algorithm, and evaluate the two-dimensional matrix according to the corresponding evaluation model to obtain crop electrical signal information, which is used to represent the health status of the crop. The evaluation model can be a convolutional neural network model.

[0045] In the embodiment of the present invention, the electrical signal of the crop will be affected by its growth state, so when pests and diseases occur, its electrical signal will change accordingly. Therefore, by adding the electrical signal of the crop to the image recognition process, the accuracy of disease recognition can be effectively improved.

[0046] In addition, the electrical signals of crops will also reflect the trend of crop diseases. When the crop electrical signals reflect that the health of the crop is poor, if the image recognition does not identify the occurrence of diseases, the crop may still have or will soon have diseases and pests. Therefore, after adding electrical signals to conventional image recognition, it is not only possible to evaluate the occurrence of crop diseases from multiple angles, but also to determine the trend of crop diseases and achieve advanced prediction of diseases.

[0047] In some embodiments, a closed collection chamber is provided in the crop disease identification robot, and the electrical signal collection device 12 is provided in the closed collection chamber.

[0048] Correspondingly, the control device 13 is also used to determine the position of the leaves to be detected of each crop based on the crop images of each crop in the target area; place the leaves to be detected into a closed collection chamber, and instruct the electrical signal collection device 12 to collect the electrical signals of the leaves to be detected as the crop electrical signal information of the crop to which the leaves to be detected belong.

[0049] In an embodiment of the present invention, after obtaining the crop images of each crop in the target area, it is necessary to remove the weeds and background interference in the image, leaving only the leaves of the crop, and randomly select one or more leaves from the leaves of the crop as the leaves to be detected. The weeds and background interference in the image can be removed by using a neural network, a leaf contour extraction algorithm based on threshold segmentation, etc., which are not limited here. After the leaves are extracted, a three-dimensional coordinate system is established in the crop image, and the actual position of the leaves can be determined according to the position of the leaves in the image, so that the leaves to be detected are placed in a closed collection chamber. In addition to being affected by their own growth state, the electrical signals of crops are also affected by external environments such as temperature and light. In order to eliminate the influence of the external environment and enable the electrical signals to accurately reflect the growth state of the plants, a closed collection chamber is set on the robot, and the leaves are placed in a closed collection chamber with a constant environment for electrical signal prediction.

[0050] In some embodiments, an environmental adjustment device is provided in the closed collection chamber; the crop electrical signal information includes a first electrical signal and a second electrical signal; the first electrical signal is an electrical signal measured when the leaves to be detected are in a first preset environment; the second electrical signal is an electrical signal measured when the leaves to be detected are in a second preset environment.

[0051] Correspondingly, the control device 13 is specifically used to: calculate the growth activity of each crop based on the first electrical signal and the second electrical signal of each crop; input the growth activity of each crop and the crop image into the recognition model corresponding to each crop type to obtain the disease recognition results of each crop in the target area.

[0052] In the embodiment of the present invention, the electrical signal of the crop is a tiny signal, and there is a noise signal in the collected electrical signal. Changes in external environment such as temperature and light will only affect the change of the crop electrical signal, and will not affect the noise signal. Therefore, by testing the crop leaves in two different environments, the noise in the electrical signal can be effectively eliminated.

[0053] In an embodiment of the present invention, the photosynthetic efficiency of healthy crops and diseased crops is different, so two different environments can be simulated in a closed collection room to measure the electrical signals in the two environments, that is, to reflect the photosynthetic efficiency of the leaves in the two environments, and then the electrical signals in the two environments are subtracted and dimensionless processed to obtain the growth activity of the crop, that is, to reflect the difference in photosynthetic efficiency of the crop in the two environments, and then combined with traditional disease image recognition technology, accurate identification of the disease is achieved through the recognition model. Among them, if the difference of the electrical signals in the two environments is greater than the preset difference, that is, the growth activity is greater than the preset difference, then the crop is considered healthy, and if the difference of the electrical signals in the two environments is not greater than the preset difference, that is, the growth activity is not greater than the preset difference, then the crop is considered diseased.

[0054] Correspondingly, based on the growth activity, the image of the leaves and the growth activity are used as input, and the type and degree of labeled pests and diseases are used as output to form samples. The samples are set as training sets and test sets in a ratio of 4:1, thereby realizing the training of the recognition model.

[0055] In the embodiment of the present invention, the environment adjustment device specifically includes a lamp with uniform and adjustable illumination, a heater, a fan, a temperature sensor, a light sensor, and the like.

[0056] The control device 13 controls the operation of environment adjustment devices such as lamps, heaters and fans according to the temperature and light collected by the temperature sensor and the light sensor, thereby controlling the environmental changes in the closed collection room.

[0057] In some embodiments, the control device 13 is specifically used to: calculate the average growth activity of crops in the target area based on the growth activity of each crop; use the difference between the growth activity and the average growth activity as the degree of crop weakness; input the crop weakness degree of each crop and the crop image of each crop into the recognition model corresponding to each crop type to obtain the disease identification results of each crop in the target area.

[0058] In the embodiment of the present invention, the electrical signal of the crop is not only affected by the environment such as temperature and light, but also by factors such as soil salinity, humidity, fertility, etc. However, the soil conditions in the same area are usually not very different, so the mean growth activity of the target area can be calculated as a criterion for judging the growth activity. If the difference between the mean growth activity and the growth activity of a crop is less than the preset difference, it indicates that the crop is in a weak state.

[0059] Correspondingly, after obtaining the degree of weakness of the crop based on the growth activity and the average value of growth activity, the image of the leaves and the degree of weakness of the crop are used as input, and the type and degree of pests and diseases marked with labels are used as output to form samples. The samples are set as training sets and test sets in a ratio of 4:1, thereby realizing the training of the recognition model.

[0060] Therefore, by adding the growth activity and the average growth activity of each crop to the traditional disease image recognition technology, the accuracy of disease identification can be effectively improved.

[0061] In some embodiments, the crop disease identification robot also includes a fruit gas detection device; the fruit gas detection device is connected to the control device 13.

[0062] The corresponding control device 13 is also used to determine whether the fruits of each crop in the target area are bagged based on the crop image, and when the fruits of each crop in the target area are bagged, instruct the fruit gas detection device to collect gas information of multiple fruits of each crop; the control device 13 is also used to input the growth activity of each crop, the crop image of each crop, and the gas information of multiple fruits of each crop into the recognition model corresponding to each crop type to obtain the disease recognition results of each crop in the target area.

[0063] Many crops need to be bagged during their growth, and the main parts for plant disease and insect pest detection are leaves and fruits. For bagged fruits, it is impossible to detect diseases using traditional image recognition technology. Based on the crop image, it is determined whether the fruits of each crop in the target area are bagged. The technical means used is conventional image recognition technology. The bagging process is the same as the leaf position identification process mentioned above, which will not be described here.

[0064] Therefore, in an embodiment of the present invention, a fruit gas detection device is added, wherein the fruit gas detection device may include an electronic nose and a collecting needle. During detection, the collecting needle is inserted into the gap of the bag to collect a certain amount of gas in the bag. The electronic nose detects whether there are nitrogen oxides, methane, sulfides, terpenes, alcohols and some aromatic compounds in the gas in the bag, which are gases produced by fruit decay. Then, the image of the crop leaves, the growth activity measured on the leaves, and the gas information of the fruit are input into the recognition model to obtain the crop disease recognition result.

[0065] The electronic nose is equipped with a W5S sensor (sensitive to nitrogen oxides, with a detection limit of 1ppm), a W5C sensor (sensitive to alkanes and aromatic compounds, with a detection limit of 1ppm), a W2W sensor (sensitive to aromatic components, terpenes, and organic sulfides, with a detection limit of 1ppm), and a W2S sensor (sensitive to alcohols and some aromatic compounds, with a detection limit of 100ppm).

[0066] During the measurement, the sensor coating adsorbs volatile substances in the sample to produce conductivity changes. The ratio G / G0 of the resistivity G after the sensor adsorbs the sample volatiles to the conductivity G0 of the air adsorbed by the sensor after being filtered by activated carbon is recorded. The greater the response gas concentration, the more the G / G0 value deviates from 1 (greater than or less than 1). If the concentration is lower than the detection limit or there is no induced gas, the ratio is close to or even equal to 1.

[0067] Correspondingly, when the recognition model is trained, the detected gases must be added to the input of its samples, and the disease labels corresponding to each gas combination and the degree of lesions corresponding to the gas content must be added to the output.

[0068] In some embodiments, the crop disease identification robot also includes a flying device.

[0069] In the embodiment of the present invention, after the flying device is provided on the crop disease identification robot, it can be convenient for the crop disease identification robot to detect diseases in various complex terrains and vegetation environments.

[0070] In some embodiments, the crop disease identification robot also includes a GPS module; the control device 13 is also used to determine the position to be detected of each crop based on the crop image of each crop; based on the position information and the position to be detected reported by the GPS module, a detection point closest to the robot is used as the starting point to obtain multiple paths, and then all paths are traversed, and the path with the shortest total length is taken as the optimal path.

[0071] In the embodiment of the present invention, when inspecting crops, the robot needs to take multiple images and inspect multiple leaves and fruits. The robot is limited by its own size and the power of the battery it carries is limited, so it is necessary to save energy consumption as much as possible.

[0072] Specifically, the control device needs to establish a three-dimensional coordinate system for the entire target area, and then obtain the positions of leaves, fruits, etc. that need to be detected through crop image recognition, mark them in the three-dimensional coordinate system, obtain the positions to be detected, connect the various positions to be detected, and use the detection point closest to the robot as the starting point to obtain multiple paths, and then determine to traverse all paths, and take the path with the shortest total length as the optimal path to complete the process of planning the detection path. The robot detects along the optimal path, thereby saving energy consumption.

[0073] In some embodiments, the electrical signal acquisition device 12 includes: patch electrodes and acupuncture electrodes.

[0074] In the embodiment of the present invention, the detection accuracy of the acupuncture electrode is higher, but it will damage the leaves and is not suitable for use on crops such as seedlings. The patch electrode should be replaced.

[0075] In an embodiment of the present invention, patch electrodes and acupuncture electrodes can also be used at the same time. The patch electrodes are first used to measure the electrical signals of the leaves in a normal state, and then the acupuncture electrodes are used to measure the electrical signals of the plants in a stimulated state. The growth activity of the crops is calculated jointly by the electrical signals in the normal state and the electrical signals in the stimulated state.

[0076] The battery of the crop disease identification robot can be a lead-acid battery, a lithium battery, etc., which is not limited here.

[0077] In some embodiments, the crop disease identification robot further includes a solar cell.

[0078] Figure 2 Schematic diagram of the structure of the crop disease identification system provided by the embodiment of the present invention. Figure 2 As shown, in some embodiments, a crop disease identification system includes: a crop disease identification robot 21 and a terminal 22 as shown in any of the above embodiments.

[0079] In the embodiment of the present invention, the terminal 22 can be connected to multiple crop disease identification robots 21. The user can send a detection instruction to each crop disease identification robot 21 connected to the terminal 22, so that the robot arrives at the target area corresponding to the detection instruction for detection.

[0080] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0081] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0082] In the embodiments provided by the present invention, it should be understood that the disclosed devices / terminals and methods can be implemented in other ways. For example, the device / terminal embodiments described above are only schematic, for example, the division of modules or units is only a logical function division, and there may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0083] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0084] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0085] If the integrated module / unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. Computer-readable media may include: any entity or device capable of carrying computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electrical carrier signal, telecommunication signal and software distribution medium, etc.

[0086] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention, and should be included in the protection scope of the present invention.

Claims

1. A crop disease identification robot, characterized in that: include: A camera device, an electric signal acquisition device, and a control device; the camera device and the electric signal acquisition device are both connected to the control device; The camera device is used to capture crop images of each crop in the target area; The electrical signal collection device is used to collect crop electrical signal information of each crop in the target area; The control device is used to identify the crop type of each crop in the target area based on the crop image; and determine the disease identification results of each crop in the target area based on the crop electrical signal information of each crop, the crop image and the identification model corresponding to each crop type.

2. The crop disease identification robot according to claim 1, characterized in that: The crop disease identification robot is provided with a closed collection chamber, and the electrical signal collection device is arranged in the closed collection chamber; The control device is also used to determine the position of the leaves to be detected of each crop based on the crop images of each crop in the target area; place the leaves to be detected into the closed collection chamber, and instruct the electrical signal collection device to collect the electrical signals of the leaves to be detected as the crop electrical signal information of the crop to which the leaves to be detected belong.

3. The crop disease identification robot according to claim 2, characterized in that: The closed collection chamber is provided with an environment adjustment device; the crop electrical signal information includes a first electrical signal and a second electrical signal; the first electrical signal is an electrical signal measured when the leaf to be detected is in a first preset environment; the second electrical signal is an electrical signal measured when the leaf to be detected is in a second preset environment; The control device is specifically used to: calculate the growth activity of each crop according to the first electrical signal and the second electrical signal of each crop; input the growth activity of each crop and the crop image into the recognition model corresponding to each crop type to obtain the disease recognition result of each crop in the target area.

4. The crop disease identification robot according to claim 3, characterized in that: The control device is specifically used for: Calculating the average growth activity of the crops in the target area according to the growth activity of each crop; The difference between the growth activity and the mean growth activity was taken as the degree of crop weakness; The crop weakness degree of each crop and the crop image of each crop are input into the recognition model corresponding to each crop type to obtain the disease recognition results of each crop in the target area.

5. The crop disease identification robot according to claim 3, characterized in that: The crop disease identification robot also includes a fruit gas detection device; The fruit gas detection device is connected to the control device; The control device is also used to determine whether the fruits of each crop in the target area are bagged according to the crop image, and when the fruits of each crop in the target area are bagged, instruct the fruit gas detection device to collect gas information of multiple fruits of each crop; The control device is also used to input the growth activity of each crop, the crop image of each crop, and the gas information of multiple fruits of each crop into the recognition model corresponding to each crop type to obtain the disease recognition results of each crop in the target area.

6. The crop disease identification robot according to claim 1, characterized in that: The crop disease identification robot also includes a flying device.

7. The crop disease identification robot according to claim 6, characterized in that: The crop disease identification robot also includes a GPS module; The control device is also used to determine the position to be detected of each crop based on the crop image of each crop; based on the position information reported by the GPS module and the position to be detected, a detection point closest to the robot is used as the starting point to obtain multiple paths, and then determine to traverse all the paths and take the path with the smallest total length as the optimal path.

8. The crop disease identification robot according to claim 1, characterized in that: The electrical signal acquisition device comprises: a patch electrode and an acupuncture electrode.

9. The crop disease identification robot according to claim 1, characterized in that: The crop disease identification robot also includes a solar cell.

10. A crop disease identification system, characterized in that: include: A terminal and a crop disease identification robot as described in claim 1 above.

Images