KR102516100B1 - Disease diagnosis monitering device that diagnoses diseases of crops through image analysis and operation method thereof - Google Patents

Abstract

The present invention relates to a disease diagnosis monitoring device, and more specifically, to a disease diagnosis monitoring device to diagnose diseases by analyzing the image of crops. To this end, the disease diagnosis monitoring device according to one embodiment of the present invention may comprise: a first image filming unit which films a first image of crops by means of a first camera; a second image filming unit which films a second image of the crops by means of a second camera; a disease detection unit which analyzes at least one of the first and second images of the crops which are filmed based on at least one of an edge computing and a central artificial intelligence server; and a control unit which controls by one of a disease control mode and a camera changing mode depending on whether diseases are detected based on analyzed information.

Description

Disease diagnosis monitoring device and operation method for diagnosing crop diseases through image analysis

The present invention relates to a disease diagnosis monitoring device, and more particularly, to a disease diagnosis monitoring device for diagnosing a disease by analyzing an image of a crop.

Currently, there is a lack of a disease management system that allows farmers and non-agricultural experts to timely determine disease and provide appropriate treatment. Farmers must request an expert for accurate confirmation when a disease is suspected, and since this takes a considerable amount of time, a problem in which farmers are not informed of the exact type and method of disease at a critical time frequently occurs.

In addition, smart farm technology used in existing open-field cultivation monitors crop conditions through monitoring based on external environmental information (temperature, humidity, solar radiation, wind direction, wind speed, weather information, soil sensor, RGB image). In the case of open-field cultivation, crop cultivation is being carried out based on environmental information by period and farmers' experiences, and if farmers do not manually record their farming work, problems with poor management arise.

The reality is that most farmers are neglecting their own agricultural record management.

In smart farms (facility horticulture), greenhouse integrated management and monitoring (storage of sensor data) is automatically managed by the system, but in the case of outdoor cultivation, there is currently no system that automatically records it.

An object of the present invention is to provide a disease diagnosis monitoring device for diagnosing diseases by analyzing images of crops.

An apparatus for disease diagnosis and monitoring according to an embodiment of the present invention includes a first image capture unit that captures a first image of crops through a first camera and a second image capture that captures a second image of crops through a second camera. a disease detection unit that analyzes at least one of the first and second images of the crop taken based on at least one of edge computing and a central artificial intelligence server; and a control mode according to whether or not the disease is found based on the analyzed information. and a controller controlling one of camera change modes.

In one embodiment of the present invention, the control unit, when the disease is found from the analyzed first image, controls the control mode, and when the disease is not found from the analyzed first image, the camera change mode can be controlled.

In one embodiment of the present invention, in the case of the camera change mode, the control unit may control to change the first camera to the second camera and capture the second image.

In one embodiment of the present invention, the control unit, in the case of the control mode, a precision monitoring control unit for controlling the first and second image capturing units to take an enlarged image corresponding to the location where the disease is detected, a spray pesticide It may include a spraying pesticide type selection unit for selecting the type, a location determination unit for determining the location where the disease has occurred, and a control information transmission unit for transmitting the type of pesticide spraying and the location where the disease has occurred for control.

In one embodiment of the present invention, the location determination unit can represent the disease outbreak location information as 3-dimensional location information representing x, y, and z axes from the ground and 4-dimensional location information including an expected time of onset there is.

In one embodiment of the present invention, a color marking unit for storing a colorant for displaying a colorant at the location where the disease is detected may be further included, and the colorant may be used to mark the location where the disease is detected.

In one embodiment of the present invention, the first camera may be an RGB camera, and the second camera may be a stereo spectroscopy camera.

In one embodiment of the present invention, the disease detection unit analyzes the image of the crop in real time through the edge computing, and transmits the image of the crop to the central artificial intelligence server when the analysis is impossible through the edge computing. Thus, the image of the crop can be analyzed by the central artificial intelligence server.

In one embodiment of the present invention, the edge computing analyzes at least one of the first and second images of the crop based on a first algorithm, and the central artificial intelligence server analyzes the crop based on the second algorithm. At least one of the first and second images of may be analyzed.

An operation method of a disease diagnosis monitoring device according to an embodiment of the present invention includes the steps of photographing a first image of a crop through a first camera, photographing a second image of the crop through a second camera, edge computing and Analyzing at least one of the first and second images of the crop taken based on at least one of the central artificial intelligence servers, and selecting one of a control mode and a camera change mode according to whether a disease is found based on the analyzed information. It may include a control step.

In one embodiment of the present invention, the step of controlling to one of the control mode and the camera change mode according to whether or not the disease is found based on the analyzed information, when the disease is found from the analyzed first image, the control mode , and when the disease is not found in the analyzed first image, control may be performed in the camera change mode.

In one embodiment of the present invention, in the case of the camera change mode, the first camera may be changed to a second camera, and control may be performed to capture the second image.

In one embodiment of the present invention, the step of controlling to one of the control mode and the camera change mode according to whether the disease is found based on the analyzed information is the expansion corresponding to the location where the disease is detected in the case of the control mode. It may include taking an image, selecting the type of pesticide to be sprayed, determining the location where the disease has occurred, and transmitting the type of pesticide to be sprayed and the location where the disease has occurred for control.

In one embodiment of the present invention, the step of analyzing the image of the crop taken based on at least one of the edge computing and the central artificial intelligence server includes analyzing the image of the crop in real time through the edge computing, When the analysis is impossible by the edge computing, the image of the crop may be transmitted to the central artificial intelligence server, and the image of the crop may be analyzed by the central artificial intelligence server.

A computer program according to an embodiment of the present invention may be combined with a processor and stored in a medium to perform any one of the methods described above.

A disease diagnosis monitoring system according to an embodiment of the present invention includes a disease diagnosis monitoring device and a server that manages pesticide information according to whether or not the disease is discovered by the disease diagnosis monitoring device and the type of disease, the disease diagnosis monitoring device comprising: At least one of a first image capturing unit capturing a first image of crops through a first camera, a second image capturing unit capturing a second image of crops through a second camera, edge computing, and a central artificial intelligence server It may include a disease detection unit for analyzing the image of the crop taken based on the image, and a control unit for controlling one of a control mode and a camera change mode according to whether or not a disease is found based on the analyzed information.

The disease diagnosis monitoring device of the present invention diagnoses diseases by analyzing images of crops using a plurality of cameras, thereby diagnosing diseases more precisely and accurately.

1 is a block diagram illustrating a disease diagnosis monitoring apparatus according to an embodiment of the present invention.
2 is a block diagram illustrating a disease detection unit according to an embodiment of the present invention.
3 is a block diagram showing a control unit according to an embodiment of the present invention.
4 is an example of a disease diagnosis monitoring device according to another embodiment of the present invention.
5 is an example of a disease diagnosis monitoring device according to another embodiment of the present invention.
6 is an example of a disease diagnosis monitoring device according to an embodiment of the present invention.
7 is a flowchart illustrating an operating method of a disease diagnosis monitoring apparatus according to an embodiment of the present invention.
8 is a flowchart illustrating an operating method of a disease detection unit according to an embodiment of the present invention.
9 is a flowchart illustrating an operating method of a control unit according to an embodiment of the present invention.

Hereinafter, the present disclosure will be described with reference to the accompanying drawings. Since the present disclosure can make various changes and have various embodiments, specific embodiments are illustrated in the drawings and related detailed descriptions are described. However, this is not intended to limit the present disclosure to specific embodiments, and should be understood to include all changes and/or equivalents or substitutes included in the spirit and scope of the present disclosure. In connection with the description of the drawings, like reference numerals have been used for like elements.

Terms used in the present disclosure are only used to describe specific embodiments, and are not intended to limit the present disclosure. Singular expressions include plural expressions unless the context clearly dictates otherwise.

Unless defined otherwise, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by one of ordinary skill in the art to which this disclosure belongs. Terms such as those defined in commonly used dictionaries should be interpreted as having a meaning consistent with the meaning in the context of the related art, and unless explicitly defined in the present disclosure, they should not be interpreted in an ideal or excessively formal meaning. don't

1 is a block diagram illustrating a disease diagnosis monitoring apparatus according to an embodiment of the present invention.

Referring to FIG. 1 , the disease diagnosis monitoring device 10 may include a first image capture unit 100 , a second image capture unit 200 , a disease detection unit 300 , and a control unit 400 .

The first image capturing unit 100 may capture a first image of crops through a first camera. In this case, the first camera may be an RGB camera. The RGB camera can quickly analyze the first image due to its fast analysis time.

The second image capturing unit 200 may capture a second image of crops through a second camera. In this case, the second camera may be a stereo spectroscopy camera. The analysis time of the stereo spectroscopy camera is longer than that of the RGB camera, but more accurate analysis can be achieved by longer analysis. In addition, the stereoscopic camera can increase the accuracy of the position and posture of the object.

When the controller 400 is set to the camera change mode, the second image capturing unit 200 may capture a second image of the crop through the second camera. For example, the second image capture unit 200 may capture a second image of crops through a second camera when no disease is detected in the first image and the control unit 400 controls the camera change mode. can

As another example, after capturing the first and second images, the first and second image capture units 200 may determine whether to analyze the second image according to a result of analyzing the first image. If the disease is found as a result of analyzing the first image, the second image may not be analyzed, and if the disease is not found as a result of analyzing the first image, the second image may be analyzed.

The disease detection unit 300 may analyze at least one of the first and second images based on one of the edge computing and the central artificial intelligence server to determine whether the disease is detected. For example, the disease detection unit 300 may first analyze the first image. After that, the first image analysis result may be transmitted to the controller 400 .

As another example, the disease detection unit 300 may simultaneously analyze the first and second images. After that, analysis results of the first and second images may be transmitted to the controller 400 .

The control unit 400 may control one of a control mode and a camera change mode depending on whether or not a disease is detected. For example, the control unit 400 may perform control in a control mode when a disease is found in the first image, and control in a camera change mode when no disease is found in the first image. In the camera change mode, the controller 400 may change the first camera to a second camera and control to capture a second image.

2 is a block diagram illustrating a disease detection unit according to an embodiment of the present invention.

Referring to FIG. 2 , the disease detection unit 300 may include an edge computing-based detection unit 310 and a central artificial intelligence server-based detection unit 320 .

The edge computing-based detection unit 310 may analyze one of the first and second images in real time through edge computing. For example, the edge computing-based detection unit 310 may analyze the first image to determine whether a disease is detected. As another example, when the second image capturing unit captures the second image, the edge computing-based detection unit 310 may analyze the second image to determine whether the disease is detected.

The edge computing-based detection unit 310 may analyze at least one of the first and second images of crops based on the first algorithm. In this case, the first algorithm may be a tensor flow.

The central artificial intelligence server-based detection unit 320 may transmit at least one of the first and second images to the central artificial intelligence server when it is impossible to analyze the first and second images by edge computing. The central artificial intelligence server-based detection unit 320 may analyze the first and second images based on the central artificial intelligence server to determine whether a disease is detected.

The central artificial intelligence server-based detection unit 320 may analyze at least one of the first and second images of crops based on the second algorithm. In this case, the second algorithm may be Faster rcnn series deep learning.

3 is a block diagram showing a control unit according to an embodiment of the present invention.

Referring to FIG. 3, the control unit 400 includes a precise monitoring control unit 410, a spray pesticide type selection unit 420, a location determination unit 430, a control information transmission unit 440, and a camera mode change unit 450. can include

The precision monitoring control unit 410, the spray pesticide type selection unit 420, the location determination unit 430, and the control information transmission unit 440 operate in the control mode, and when a disease is detected by the disease detection unit, they are performed. It can be.

The precision monitoring controller 410 may control the first and second image capturing units to enlarge the corresponding location and capture an enlarged image so that the corresponding location can be seen in more detail when the disease detection unit detects the disease.

For example, as a result of analyzing the first image taken through the first camera, if a disease is detected, the precision monitoring controller 410 moves the first camera closer to the corresponding crop or zooms in using the first camera, An enlarged image of the corresponding location may be captured.

As another example, if the disease is not found as a result of analyzing the first image, a second image is captured through the second camera, and the disease is detected as a result of analyzing the second image, the precise monitoring controller 410 may select the second camera. It is possible to capture an enlarged image of the corresponding position by moving it close to the corresponding crop or by enlarging it using the second camera.

The spray pesticide type selection unit 420 may select the spray pesticide type according to the type of disease detected by the disease detection unit.

The location determining unit 430 may determine at least one of a disease occurrence location and an onset location. The location determining unit 430 may represent the location information of the onset of the disease as 4-dimensional location information including location information expressed in x, y, and z axes from the ground and an expected time of onset. For example, the location determination unit 430 receives the result analyzed by the artificial intelligence server based on at least one of dew condensation due to repetitive rapid temperature change, moisture stress, and tip burn, and includes information on the expected onset time for each location. so it can be transmitted. The location determining unit 430 may determine location information where the disease has occurred using GPS information. For example, if it is determined that a disease is found in the disease detection unit, the location determination unit 430 may collect the GPS location of the disease diagnosis monitoring device at that time.

The control information transmission unit 440 may transmit the type of pesticide to be sprayed and the location where the disease has occurred to at least one of the server and the control device.

The camera mode change unit 450 may change the camera mode based on whether or not the disease detection unit detects a disease. For example, if the first image capturing unit captures the first image using the first camera and the disease detection unit analyzes the first image and no disease is detected, the camera mode changing unit 450 may change the first camera mode. may be changed to the second camera. When the first camera is changed to the second camera in the camera mode changing unit 450, the second image capture unit takes a second image of the crop, and it is possible to check whether the disease is detected through the disease detection unit.

The camera mode changing unit 450 according to an embodiment of the present invention may include first to fourth camera modes. For example, the first camera mode is a mode using the first camera, the second camera mode is a mode using the second camera, the third camera is a mode using the third camera, and the fourth camera is a mode using the fourth camera. It may be a mode to use. In this case, the first camera may be an RGB camera, the second camera may be a stereo spectroscopy camera, the third camera may be an infrared camera, and the fourth camera may be a fluorescence camera.

When the camera mode changer 450 needs to operate in the night mode, it can change to a fourth camera mode using a fourth camera. On the other hand, when there is no need to operate in the night mode, the camera mode changer 450 may change the camera mode to at least one of the second camera mode and the third camera mode according to whether or not a disease is detected through the first camera mode.

After changing the camera mode, the camera mode changing unit 450 may reconfirm whether or not a disease has been detected through the disease detection unit.

As described above, the disease diagnosis monitoring device according to an embodiment of the present invention may perform one of a control mode and a camera change mode according to whether a disease is found by taking an image of a crop in a greenhouse.

In more detail, the disease diagnosis monitoring device may capture a first image of crops through a first camera. The disease diagnosis monitoring device may capture a second image of the crop through a second camera. The disease diagnosis monitoring device may analyze at least one of the first and second images of the crop based on at least one of edge computing and a central artificial intelligence server. The disease diagnosis monitoring device may control one of a control mode and a camera change mode depending on whether a disease is found based on the analyzed information.

Accordingly, the disease diagnosis monitoring device of the present invention can diagnose diseases more precisely and accurately by analyzing images of crops using a plurality of cameras and diagnosing diseases.

4 is an example of a disease diagnosis monitoring device according to another embodiment of the present invention.

Referring to FIG. 4 , the disease diagnosis monitoring device 10_1 includes a first image capture unit 100, a second image capture unit 200, a disease detection unit 300, a control unit 400, and a precise monitoring unit 500. can include

The first image capture unit 100, the second image capture unit 200, the disease detection unit 300, and the control unit 400 are the first image capture unit 100 and the second image capture unit 200 shown in FIG. ), the disease detection unit 300, and the control unit 400 are the same or similar, so detailed descriptions will be omitted.

The precise monitoring unit 500 may capture an enlarged image by operating the magnifying arm of the disease diagnosis monitoring apparatus 10_1 based on a control signal of the precise monitoring controller of the controller 400 . For example, when a disease is detected in at least one of the first and second images, the precise monitoring unit 500 may capture an enlarged image obtained by enlarging a portion where the disease is detected.

For example, the precise monitoring unit 500 may direct the enlarged arm of the disease diagnosis monitoring device 10_1 toward crops when the precise monitoring controller transmits a control signal to capture an enlarged image.

In more detail, based on the result of analyzing the first image, if the control signal for capturing the enlarged image is transmitted, the precision monitoring unit 500 moves the magnification arm equipped with the camera corresponding to the first image capturing unit 100. can make it work. On the other hand, based on the result of analyzing the second image, if the enlarged image capture control signal is received, the precise monitoring unit 500 operates the magnification arm equipped with the camera corresponding to the second image capture unit 200. can

The precise monitoring unit 500 may capture an enlarged image through the first and second image capture units based on a control signal from the controller. The precise monitoring unit 500 may transmit the photographed enlarged image to the controller 400 and analyze the enlarged image.

5 is an example of a disease diagnosis monitoring device according to another embodiment of the present invention.

Referring to FIG. 5 , the disease diagnosis monitoring device 10_2 includes a first image capture unit 100, a second image capture unit 200, a disease detection unit 300, a control unit 400, and a dye marking unit 600. can include

The first image capture unit 100, the second image capture unit 200, the disease detection unit 300, and the control unit 400 are the first image capture unit 100 and the second image capture unit 200 shown in FIG. ), the disease detection unit 300, and the control unit 400 are the same or similar, so detailed descriptions will be omitted.

The dye marking unit 600 may store a dye for displaying the disease at a location where the disease is detected. The dye marking unit 600 may mark the location where the disease is detected using the stored dye. For example, the dye marking unit 600 may mark a corresponding location with a dye when the camera moves close to the corresponding location so that the control unit 400 enlarges and captures the portion where the disease is detected.

The dye marking unit 600 can make it easy for a manager to visually check by marking the onset location of the disease with a dye. In more detail, compared to the case where the control robot controls only with location information, it is possible to more accurately locate the location using the physically marked color marking.

6 is an example of a disease diagnosis monitoring device according to an embodiment of the present invention.

Referring to FIG. 6 , the disease diagnosis monitoring device 10 may be implemented as a robot. For example, the disease diagnosis monitoring device 10 may capture images of crops by mounting a camera at the end of a robot arm.

The disease diagnosis monitoring device 10 implemented as a robot can autonomously drive in a greenhouse and take images of crops through an installed camera.

The disease diagnosis monitoring device 10 may transmit an image of a crop photographed through a camera to a disease detection unit to determine whether the disease is detected.

If a disease is detected, the disease diagnosis monitoring device 10 may extend the arm of the robot toward the crop, take an enlarged image, and mark the location of the disease by using a colorant.

On the other hand, if no disease is detected, the disease diagnosis monitoring device 10 may change the current camera mode to another camera mode and retake an image of the crop.

7 is a flowchart illustrating an operating method of a disease diagnosis monitoring apparatus according to an embodiment of the present invention.

In step S100, the first image capturing unit may capture a first image of crops using a first camera. For example, the first image capture unit may capture a first image of crops using an RGB camera.

In step S200, the second image capture unit may capture a second image of the crop using a second camera. For example, the second image capturing unit may analyze the second image of the crop using a stereo spectroscopic camera.

In step S300, the disease detection unit may analyze at least one of the captured first and second images of crops based on at least one of edge computing and a central artificial intelligence server. For example, the disease detection unit may analyze at least one of the first and second images based on edge computing. However, when the analysis result is not derived or the analysis is impossible based on edge computing, the first and second images are transmitted to the central artificial intelligence server, and at least one of the first and second images is transmitted based on the central artificial intelligence server. can be analyzed.

In step S400, it may be controlled in one of a control mode and a camera change mode according to whether or not a disease is found. For example, when a disease is found, it operates in a control mode to collect information about the outbreak location of the disease and pesticides. On the other hand, when the disease is not found, the RGB camera may be changed to a stereo spectroscopy camera or the stereo spectroscopy camera may be changed to a third camera by operating in a camera change mode.

As described above, the disease diagnosis monitoring device according to an embodiment of the present invention acquires first and second images of crops in advance using first and second cameras, and analyzes the first images to discover diseases. If not, the second image may be analyzed.

However, the disease diagnosis monitoring device according to another embodiment acquires and analyzes a first image of a crop using a first camera, and acquires and analyzes a second image of a crop using a second camera when no disease is found. You may.

8 is a flowchart illustrating an operating method of a disease detection unit according to an embodiment of the present invention.

In step S210, the disease detection unit may determine whether edge computing is available. For example, if edge computing is turned off or a signal is not received, the disease detection unit may determine that edge computing cannot be used. If edge computing is available, the disease detection unit may perform step S220, and if edge computing is not available, the disease detection unit may perform step S230.

In step S220, the disease detection unit may analyze the crop image in real time through edge computing. For example, the disease detection unit may analyze whether crops have diseases based on edge computing. In this case, edge computing may use a first algorithm, and the first algorithm may be Tensor Flow.

In step S230, the disease detection unit may transmit the crop image to the central artificial intelligence server. For example, when edge computing is unavailable, the disease detection unit may transmit images of crops to a central artificial intelligence server for analysis based on the central artificial intelligence server.

In step S240, the disease detection unit may analyze crop images based on the central artificial intelligence server. For example, the disease detection unit can analyze crops for disease based on a central AI server. At this time, the central artificial intelligence server may use a second algorithm, and the second algorithm may be Faster rcnn series deep learning.

9 is an example of a flowchart illustrating a method of operating a controller using four cameras.

Referring to FIG. 9 , the disease diagnosis monitoring device may use an RGB camera, a stereo spectroscopy camera, an infrared camera, and a fluorescence camera.

In step S310, the controller may determine whether to operate in the night mode. For example, it may be determined whether to operate in the night mode by determining whether the current time is at night or in a dark state because all lights are turned off. If it is necessary to operate in the night mode, the controller may perform step S320, and if it is not necessary to operate in the night mode, it may perform step S330. However, if a fluorescent camera is not used, this step can be omitted.

In step S320, the controller may change the current camera to a fluorescent camera. For example, if the current camera is a stereo spectroscopy camera, the controller may change the stereo spectroscopy camera to a fluorescent camera capable of detecting crop diseases even at night. However, step S320 may be omitted, and may be performed before step S200, so that the fourth image corresponding to the fluorescent camera may be captured in advance.

In step S330, the control unit may determine whether disease is found in the image of the crop. For example, the control unit may determine whether a disease is found in the analyzed crop image based on the analysis result of the disease detection unit. If the disease is found, step S340 may be performed to implement the control mode, and if no disease is found, step S380 may be performed to implement the camera change mode.

In step S340, the controller may take an enlarged image corresponding to the location where the disease is detected. For example, the controller may capture an enlarged image of the location where the disease is detected by bringing the camera closer to the location or enlarging the screen on the camera.

In step S350, the controller may select the type of pesticide to be sprayed based on the enlarged image. For example, the control unit may acquire the type of disease through an enlarged image and select the type of pesticide to be sprayed according to the disease. At this time, the types of pesticides to be sprayed according to the disease may be pre-stored.

In step S360, the control unit can determine the location where the disease has occurred. For example, the control unit expresses the location where the disease is found in 3D location information based on the enlarged image, and further includes the expected time of onset in the 3D location information to represent the location information of the disease outbreak as 4D information. can

In step S370, the control unit may transmit the type of pesticide to be sprayed and the location where the disease has occurred.

In step S380, the controller may change the RGB camera to a stereo spectroscopy camera. However, this step may be an example for a case where the basic camera is an RGB camera. In one embodiment, step S380 may be performed before step S200, and the second image corresponding to the stereo spectroscopic camera may have been captured in advance.

In step S381, the control unit may determine whether disease is found in the crop image. Since step S381 is the same as or similar to step S330, a detailed description thereof may be omitted. If the disease is found in the crop image, step S340 may be performed to perform the control mode. On the other hand, if no disease is found in the crop image, step S390 may be performed to perform the camera change mode.

In step S390, the controller may change the stereo spectroscopic camera to an infrared camera. However, step S390 may be omitted, and may be performed before step S200 so that the third image corresponding to the infrared camera may have been captured in advance.

In step S391, the control unit may determine whether disease is found in the image of the crop. Since step S391 is the same as or similar to step S330, a detailed description thereof may be omitted. If the disease is found in the crop image, step S340 may be performed to perform the control mode.

In one embodiment of the present invention, the disease diagnosis monitoring system may include a disease diagnosis monitoring device and a server.

Since the disease diagnosis monitoring device is the same as or similar to the disease diagnosis monitoring device disclosed in FIGS. 1 to 6 , a detailed description thereof will be omitted.

The server may manage pesticide information according to whether or not the disease diagnosis monitoring device has discovered the disease and the type of disease. For example, the server may store pesticide information according to disease types. If a disease called spotted mite is found in the disease diagnosis monitoring device, the server may receive information about the spotted mite detection. The server may transmit pesticide information corresponding to the spotted mite to the disease diagnosis monitoring device based on the transmitted spotted mite discovery information.

In one embodiment, the disease diagnosis monitoring system may further include a user terminal. If the disease diagnosis monitoring system further includes a user terminal, the server may allow the user terminal and the disease diagnosis monitoring device to communicate with each other. At this time, the user terminal can check the crop image captured by the disease diagnosis monitoring device in real time.

For example, the user terminal includes a smartphone, a smart pad, a tablet PC, a wearable device, a personal communication system (PCS), a global system for mobile communication (GSM), a personal digital cellular (PDC), and a PHS ( Personal Handyphone System), PDA (Personal Digital Assistant), IMT (International Mobile Telecommunication)-2000, CDMA (Code Division Multiple Access)-2000, W-CDMA (W-Code Division Multiple Access), Wibro (WirelessBroadband Internet) terminals It may be a kind of wireless communication device, a desktop computer, and a stationary terminal such as a smart TV.

The embodiments disclosed in this specification and drawings are only presented as specific examples to easily explain the content of the present disclosure and aid understanding, and are not intended to limit the scope of the present disclosure. Therefore, the scope of the present disclosure should be construed as including all changes or modified forms derived based on the technical spirit of the present disclosure in addition to the embodiments disclosed herein.

10, 10_1, 10_2: disease diagnosis monitoring device
100: first image capture unit
200: second image capturing unit
300: disease detection unit
310: edge computing based detection unit
320: central artificial intelligence server-based detection unit
400: control unit
410: precision monitoring control unit
420: spray pesticide type selection unit
430: location determination unit
440: control information transmission unit
450: camera mode change unit
500: precision monitoring unit
600: color marking unit

Claims (16)

A first image capture unit for capturing a first image of crops through a first camera;
a second image capture unit for capturing a second image of the crop through a second camera;
a disease detection unit for detecting a disease by analyzing at least one of the first and second images of the crop taken based on at least one of edge computing and a central artificial intelligence server;
A controller for controlling one of a control mode and a camera change mode according to whether a disease is found based on the analyzed information; and
A dye marking unit for storing a dye for displaying the disease at a location where the disease is detected;
The disease detection unit determines whether the edge computing is usable, but determines that it is not usable when the edge computing is turned off or cannot receive a signal;
When the edge computing is available, analyzing the first and second images based on the edge computing;
When it is impossible to use the edge computing or an analysis result is not derived by the edge computing, the image of the crop is transmitted to the central artificial intelligence server, and the first and second images are based on the central artificial intelligence server. to analyze,
The edge computing analyzes at least one of the first and second images of the crop based on a first algorithm,
The central artificial intelligence server analyzes at least one of the first and second images of the crop based on a second algorithm based on deep learning,
The dye marking unit marks the location where the disease is detected using the dye,
The control unit,
When a disease is found from the analyzed first image, it is controlled in the control mode,
When the disease is not found in the analyzed first image, control to the camera change mode,
The control unit,
Precision monitoring control unit for controlling the first and second image capturing unit to take an enlarged image corresponding to the location where the disease is detected in case of the control mode;
a spray pesticide type selection unit that selects a spray pesticide type;
a location determination unit for determining a location where the disease has occurred; and
Including a control information transmission unit for transmitting the type of spray pesticide and the location where the disease occurred for control,
The location determination unit represents the location information of the outbreak of the disease as 3-dimensional location information representing the x, y, and z-axes from the ground and 4-dimensional location information including the expected time of onset, condensation due to repeated rapid temperature change An apparatus for diagnosing and monitoring diseases that includes the expected time of onset in the 4-dimensional location information for each location based on at least one of occurrence, moisture stress, and tip burn. delete According to claim 1,
In case of the above camera change mode,
Wherein the control unit changes the first camera to the second camera and controls to capture the second image. delete delete delete According to claim 1,
Wherein the first camera is an RGB camera, and the second camera is a stereo spectroscopy camera. According to claim 1,
The disease detection unit,
A disease diagnosis monitoring device that analyzes the image of the crop in real time through the edge computing. delete As an operating method performed by the disease diagnosis monitoring device,
Taking a first image of crops through a first camera;
photographing a second image of the crop through a second camera;
detecting a disease by analyzing at least one of the first and second images of the crop taken based on at least one of edge computing and a central artificial intelligence server; and
Based on the analyzed information, controlling to one of the control mode and the camera change mode depending on whether the disease is found,
The step of detecting the disease is,
Determining whether the edge computing is usable, but determining that it is not usable when the edge computing is turned off or does not receive a signal;
When the edge computing is available, analyzing the first and second images based on the edge computing;
When it is impossible to use the edge computing or an analysis result is not derived by the edge computing, the image of the crop is transmitted to the central artificial intelligence server, and the first and second images are based on the central artificial intelligence server. to analyze,
The edge computing analyzes at least one of the first and second images of the crop based on a first algorithm,
The central artificial intelligence server analyzes at least one of the first and second images of the crop based on a second algorithm different from the first algorithm,
The dye marking unit included in the disease diagnosis monitoring device marks the location where the disease is detected using the dye,
The control unit included in the disease diagnosis monitoring device,
When a disease is found from the analyzed first image, it is controlled in the control mode,
When the disease is not found in the analyzed first image, control to the camera change mode,
The control unit,
In the case of the control mode, a precise monitoring control unit for controlling the first and second image capturing units to take an enlarged image corresponding to the location where the disease is detected, a spraying pesticide type selection unit for selecting the type of pesticide to be sprayed, Including a control information transmission unit for transmitting the type of spray pesticide and the location where the disease has occurred for control and a location determination unit for determining the location where the disease occurs,
The location determination unit represents the location information of the outbreak of the disease as 3-dimensional location information representing the x, y, and z-axes from the ground and 4-dimensional location information including the expected time of onset, condensation due to repeated rapid temperature change A method of operating a disease diagnosis monitoring device, wherein the expected time of onset is included in the four-dimensional location information for each location based on at least one of occurrence, moisture stress, and tip burn. delete According to claim 10,
In the case of the camera change mode, the first camera is changed to a second camera and controlled to capture the second image. delete According to claim 10,
Analyzing the image of the crop taken based on at least one of the edge computing and the central artificial intelligence server,
A method of operating a disease diagnosis monitoring device that analyzes the image of the crop in real time through the edge computing. A computer program stored in a medium combined with a processor to perform the method of any one of claims 10, 12 and 14. disease diagnosis monitoring device; and
Including a server that manages pesticide information according to whether or not the disease diagnosis monitoring device has been found and the type of disease,
The disease diagnosis monitoring device,
A first image capture unit for capturing a first image of crops through a first camera;
a second image capture unit for capturing a second image of the crop through a second camera;
a disease detecting unit for detecting a disease by analyzing an image of the crop taken based on at least one of edge computing and a central artificial intelligence server;
A controller for controlling one of a control mode and a camera change mode according to whether a disease is found based on the analyzed information; and
A dye marking unit for storing a dye for displaying the disease at a location where the disease is detected;
The disease detection unit determines whether the edge computing is usable, but determines that it is not usable when the edge computing is turned off or cannot receive a signal;
When the edge computing is available, analyzing the first and second images based on the edge computing;
When it is impossible to use the edge computing or an analysis result is not derived by the edge computing, the image of the crop is transmitted to the central artificial intelligence server, and the first and second images are based on the central artificial intelligence server. to analyze,
The edge computing analyzes at least one of the first and second images of the crop based on a first algorithm,
The central artificial intelligence server analyzes at least one of the first and second images of the crop based on a second algorithm different from the first algorithm,
The dye marking unit marks the location where the disease is detected using the dye,
The control unit,
When a disease is found from the analyzed first image, it is controlled in the control mode,
When the disease is not found in the analyzed first image, control to the camera change mode,
The control unit,
Precision monitoring control unit for controlling the first and second image capturing unit to take an enlarged image corresponding to the location where the disease is detected in case of the control mode;
a spray pesticide type selection unit that selects a spray pesticide type;
a location determination unit for determining a location where the disease has occurred; and
Including a control information transmission unit for transmitting the type of spray pesticide and the location where the disease occurred for control,
The location determination unit represents the location information of the outbreak of the disease as 3-dimensional location information representing the x, y, and z-axes from the ground and 4-dimensional location information including the expected time of onset, condensation due to repeated rapid temperature change The system for diagnosing and monitoring disease, wherein the expected time of onset is included in the 4-dimensional location information for each location based on at least one of occurrence, moisture stress, and tip burn.

Images