KR102590772B1 - Apparatus and method for measuring length of leaf stem - Google Patents

Abstract

An apparatus for measuring inner leaf stem length and a method for measuring inner leaf stem length are presented, and the apparatus for measuring inner leaf stem length according to an embodiment disclosed herein includes an input/output unit for acquiring an image, and a memory storing a program for performing deep learning. , and a control unit that performs deep learning on the image frame by executing the program, wherein the control unit acquires the position of the stump from the image frame in which the crop is photographed using the learned first deep learning model, and learns When the position of the inner leaf is obtained from the image frame using the second deep learning model, the length of the inner leaf stem is calculated based on the position of the stump and the position of the inner leaf.

Description

Device for measuring inner leaf stem length and method for measuring inner leaf stem length {APPARATUS AND METHOD FOR MEASURING LENGTH OF LEAF STEM}

The embodiments disclosed in this specification relate to an apparatus for measuring inner leaf stem length and a method for measuring inner leaf stem length. More specifically, it relates to an apparatus and method for measuring the length of an inner leaf stem by identifying the inner leaf using deep learning.

Recently, interest in implementing smart farms has been increasing.

Smart farm is an agricultural system that improves not only production efficiency but also convenience by automatically managing crops, livestock, and marine products to maintain an appropriate growth environment.

For example, in a smart farm, crop growth can be diagnosed by automatically measuring the speed at which crop stems grow.

Meanwhile, the indicators indicating growth may be different for each plant. For example, in plants with inner leaves, such as strawberries, the inner leaf growth rate plays an important role in the indicator of plant growth. For example, as the inner leaf grows, the stem of the inner leaf becomes longer. Depending on the lengthening speed, the growth rate of the inner leaf can be confirmed, and this can be used to infer the growth rate of strawberries.

However, since the inner leaves are often located together with other leaves or stems, it is difficult to identify the inner leaves.

In relation to this, the crop growth control system proposed in Korean Patent Publication No. 10-2020-0122612 includes a sensor that senses the growth status of crops inside the greenhouse and environmental information inside the greenhouse, an environmental controller to manage the environment inside the greenhouse, It includes a crop cultivation analysis device that obtains the control set value of the environmental controller and the control set value of the nutrient solution based on the growth status information of the crop sensed by the nutrient solution and sensor that supplies nutrient solution to the crops and the environmental information inside the greenhouse. However, the system only describes technology for supplying nutrient solution to crops such as strawberries and does not suggest technology for measuring inner leaf stems.

Therefore, technology to solve the above-mentioned problems has become necessary.

Meanwhile, the above-described background technology is technical information that the inventor possessed for deriving the present invention or acquired in the process of deriving the present invention, and cannot necessarily be said to be known technology disclosed to the general public before filing the application for the present invention. .

The purpose of the embodiments disclosed herein is to present a device for measuring inner leaf stem length and a method for measuring inner leaf stem length.

Additionally, the embodiments disclosed herein are aimed at quickly and accurately measuring inner leaf stem length using deep learning.

Additionally, the embodiments disclosed herein are aimed at monitoring the growth of crops by measuring the length of inner leaf stems.

As a technical means for achieving the above-described technical problem, according to one embodiment, there is a device for measuring the inner leaf stem of a crop, including an input/output unit for acquiring an image, a memory storing a program for performing deep learning, and It includes a control unit that performs deep learning on the image frame by executing the program, wherein the control unit acquires the position of the stump from the image frame in which the crop is photographed using the learned first deep learning model, and uses the learned first deep learning model to obtain the position of the stump. 2If the position of the inner leaf is obtained from the image frame using a deep learning model, the length of the inner leaf stem can be calculated based on the position of the stump and the position of the inner leaf.

According to another embodiment, the inner leaf stem measuring device is a method for measuring the inner leaf stem of a crop, comprising the steps of acquiring the position of the stump from an image frame photographing the crop using a learned first deep learning model, and the learned first deep learning model. 2It may include obtaining the position of the inner leaf from the image frame using a deep learning model, and calculating the length of the inner leaf stem based on the position of the stump and the position of the inner leaf.

According to another embodiment, a computer-readable recording medium on which a program for performing a method of measuring inner leaf stems is recorded, comprising: acquiring the position of the stump from an image frame photographing a crop using a learned first deep learning model; It may include obtaining the position of the inner leaf from the image frame using the learned second deep learning model, and calculating the length of the inner leaf stem based on the position of the stump and the position of the inner leaf.

According to another embodiment, it is performed by an inner leaf stem measuring device and is a computer program stored in a medium for performing the inner leaf stem measuring method, wherein the position of the stump is obtained from an image frame of a crop photographed using a learned first deep learning model. Obtaining the position of the inner leaf from the image frame using a learned second deep learning model, and calculating the length of the inner leaf stem based on the position of the stump and the position of the inner leaf. can do.

According to one of the means for solving the above-mentioned problem, the purpose is to present a device for measuring inner leaf stem length and a method for measuring inner leaf stem length.

In addition, according to any one of the above-mentioned problem solving means, the inner leaf stem length can be measured quickly and accurately using deep learning.

In addition, according to any one of the above-mentioned problem solving means, the growth of crops can be monitored by measuring the length of the inner leaf stem.

The effects that can be obtained from the disclosed embodiments are not limited to the effects mentioned above, and other effects not mentioned are clear to those skilled in the art to which the disclosed embodiments belong from the description below. It will be understandable.

1 is a configuration diagram illustrating a device for measuring inner leaf stems according to an embodiment disclosed herein.
Figure 2 is a block diagram showing an inner leaf stem measuring device according to an embodiment disclosed herein.
Figures 3 to 5 are exemplary diagrams for explaining an inner leaf stem measuring device according to an embodiment disclosed herein.
Figure 6 is an exemplary diagram for explaining a method of measuring inner leaf stems according to an embodiment disclosed herein.

Below, various embodiments will be described in detail with reference to the attached drawings. The embodiments described below may be modified and implemented in various different forms. In order to more clearly explain the characteristics of the embodiments, detailed descriptions of matters widely known to those skilled in the art to which the following embodiments belong have been omitted. In addition, in the drawings, parts that are not related to the description of the embodiments are omitted, and similar parts are given similar reference numerals throughout the specification.

Throughout the specification, when a configuration is said to be “connected” to another configuration, this includes not only cases where it is “directly connected,” but also cases where it is “connected with another configuration in between.” In addition, when a configuration “includes” a configuration, this means that other configurations may be further included rather than excluding other configurations, unless specifically stated to the contrary.

Hereinafter, embodiments will be described in detail with reference to the attached drawings.

However, before explaining this, we first define the meaning of the terms used below.

'Internal leaves' are leaves that emerge from the base (or crown) and grow into the leaves of crops.

Terms that require explanation other than those defined above are explained separately below.

1 is a configuration diagram illustrating a device for measuring inner leaf stems according to an embodiment disclosed herein.

According to one embodiment, a program for performing a deep learning model may be installed and driven in the inner leaf stem measuring device 100, and the inner leaf stem measuring device 100 may be configured as described herein under the control of the driven computer program. The inner leaf stem measurement method according to the example can be performed.

For example, the inner leaf stem measurement device 100 can predict the growth of crops by identifying inner leaves using deep learning from frames that make up images of crops and calculating the length of inner leaf stems.

This internal leaf stem measuring device 100 may be implemented as a terminal with an application installed that can interact with the manager or as a server-client system. When implemented as a server-client system, it can be used as an online service for interaction with the manager. It may include a terminal where the application is installed.

For example, the inner leaf stem measurement device 100 implemented as a server-client system may be composed of one or more terminals 10 and 11 and a server 20, as shown in FIG. 1, and the terminal 10, 11) and the server 20 can communicate through the network (N).

The terminals 10 and 11 may be implemented as computers, portable terminals, televisions, wearable devices, etc. that can connect to a remote server through a network (N) or connect to other terminals and servers. Here, the computer includes, for example, a laptop, desktop, laptop, etc. equipped with a web browser, and the portable terminal is, for example, a wireless communication device that guarantees portability and mobility. , PCS (Personal Communication System), PDC (Personal Digital Cellular), PHS (Personal Handyphone System), PDA (Personal Digital Assistant), GSM (Global System for Mobile communications), IMT (International Mobile Telecommunication)-2000, CDMA (Code) All types of handhelds such as Division Multiple Access)-2000, W-CDMA (W-Code Division Multiple Access), Wibro (Wireless Broadband Internet), Smart Phone, Mobile WiMAX (Mobile Worldwide Interoperability for Microwave Access), etc. (Handheld)-based wireless communication device may be included. Additionally, television may include IPTV (Internet Protocol Television), Internet TV (Internet Television), terrestrial TV, cable TV, etc. Furthermore, a wearable device is a type of information processing device that can be worn directly on the human body, such as a watch, glasses, accessories, clothing, or shoes, and can connect to a remote server or other terminal via a network directly or through another information processing device. can be connected to

And the server 20 is implemented as a computer capable of communicating through a network (N) with terminals (10, 11) installed with an application or web browser for interaction with the manager of the inner leaf stem measurement device (100), or as a cloud computing server. It can be implemented. Additionally, the server 20 may be performed on a plurality of physically separated servers or on a single server, and may also include a storage device capable of storing data or may store data through a third server.

Therefore, the inner leaf stem measurement device 100 implemented as a server-client system, for example, photographs crops while the terminal 11 moves on a rice field, field, or orchard or moves inside a greenhouse, and captures the images obtained accordingly. As transmitted to the server 20, the server 20 measures the length of the inner leaf stem and outputs the measured result or crop growth information according to the measurement result so that the manager can check it through the terminal 10. can do.

Meanwhile, Figure 2 is a block diagram showing an inner leaf stem measuring device 100 according to an embodiment.

Referring to FIG. 2, the inner leaf stem measuring device 100 according to an embodiment may include an input/output unit 110, a control unit 120, a communication unit 130, and a memory 140. For example, if the inner leaf stem measurement device 100 is implemented as a server-client system, each component of the input/output unit 110, control unit 120, communication unit 130, and memory 140 is connected to the terminal 10. , 11) or may be implemented through the server 20.

The input/output unit 110 may include an input unit for receiving input from a user and an output unit for displaying information such as a task performance result or the status of the user terminal 100. For example, the input/output unit 110 may include an operation panel that receives user input and a display panel that displays a screen.

Specifically, the input unit may include devices that can receive various types of user input, such as a keyboard, physical button, touch screen, camera, or microphone. Additionally, the output unit may include a display panel or a speaker. However, the input/output unit 110 is not limited to this and may include a configuration that supports various inputs and outputs.

According to the embodiment, the input unit may generate an image of a crop as an object, or may receive an image generated through an external device as an input. Additionally, the output unit may output growth information, etc., which is a result of calculation on the image, or output an alarm according to growth.

The control unit 120 controls the overall operation of the inner leaf stem measurement device 100 and may include a processor such as a CPU. The control unit 120 may control other components included in the inner leaf stem measuring device 100 to perform operations corresponding to user input received through the input/output unit 110.

For example, the control unit 120 may execute a program stored in the memory 140, read a file stored in the memory 140, or store a new file in the memory 140. More details regarding the control unit 120 will be described later.

The communication unit 130 may perform wired or wireless communication with other devices or networks. To this end, the communication unit 130 may include a communication module that supports at least one of various wired and wireless communication methods. For example, a communication module may be implemented in the form of a chipset.

Wireless communication supported by the communication unit 130 may be, for example, Wi-Fi (Wireless Fidelity), Wi-Fi Direct, Bluetooth, UWB (Ultra Wide Band), or NFC (Near Field Communication). Additionally, wired communication supported by the communication unit 130 may be, for example, USB or HDMI (High Definition Multimedia Interface).

Various types of data, such as files, applications, and programs, can be installed and stored in the memory 140. The control unit 120 may access and use data stored in the memory 140, or may store new data in the memory 140. Additionally, the control unit 120 may execute a program installed in the memory 140. For example, a program for performing a method of measuring inner leaf stems may be installed in the memory 140.

According to one embodiment, when an image of a crop is acquired through the input/output unit 110, the control unit 120 executes a program stored in the memory 140 to perform a method of measuring inner leaf stems.

Meanwhile, with reference to FIGS. 3 to 5, the operation of the control unit 120 according to an embodiment disclosed herein will be described.

In relation to this, FIGS. 3 to 5 are exemplary diagrams for explaining the inner leaf stem measurement device 100 according to an embodiment disclosed herein, through the screen of the terminal 10 on which the inner leaf stem measurement device 100 is implemented. This shows an example of a video frame that appears.

The control unit 120 may analyze image frames in which crops are photographed.

At this time, preprocessing may be performed on the image frame for accurate analysis of the image frame. For example, the control unit 120 may correct the color, saturation, etc. of the image frame based on the amount of sunlight at the time the image was captured. there is.

According to one embodiment, the control unit 120 may use a learned deep learning model to obtain the position of the stump from an image frame in which a crop is photographed.

For this purpose, a deep learning model can be trained to output the result of detecting the stump within the video frame. At this time, as shown in FIG. 3, the control unit 120 can identify the stump within the image frame 300 through object detection and display a bounding box 310 surrounding the identified stump within the image frame. . This deep learning model may be implemented as a network model, for example, a convolutional neural network (CNN) or a recurrent neural network (RNN).

For example, when the control unit 120 obtains the bounding box 310 surrounding the stump, it can set one of the midpoints or vertices of the bounding box 310 as the position of the stump. For example, as shown in FIG. 3, the midpoint 320 within the bounding box 310 can be set as the position of the base.

Alternatively, for example, when the control unit 120 obtains the bounding box 310 surrounding the stump, it can analyze the color within the bounding box 310 and set a pixel where a color of a predetermined value is located as the position of the stump. That is, for example, the position of a pixel with a color value related to brown within the bounding box 310 can be set as the position of the base. Or, for example, if there are a plurality of pixels with a brown color value in the bounding box 310 and some of the pixels are located adjacent to each other, the position of any pixel among the plurality of pixels is selected as the position of the base. You can set it.

If the position of the stump is identified in this way, the control unit 120 can obtain the position of the inner leaf from the image frame using the learned deep learning model.

For this purpose, a deep learning model can be trained using video frames to output results showing the inner leaves within the video frame. For example, a deep learning model can be trained to identify inner leaves within an image frame and output a bounding box surrounding the identified inner leaves. At this time, the deep learning model may be implemented as a network model, such as a convolutional neural network (CNN) or a recurrent neural network (RNN).

To this end, the control unit 120 can obtain a cropped area by cropping a predetermined area within the image frame based on the position of the stump. Alternatively, the control unit 120 may obtain the crop area by cropping the inside based on the boundary of the bounding box surrounding the stump.

And the control unit 120 can identify the location of the inner leaf by inputting the crop area into the learned deep learning model. That is, focusing on the fact that among crop leaves, the inner leaves are located around the base, the crop area is obtained by cropping the area within a certain radius based on the position of the base, and the location of the inner leaves is identified by inputting the crop area into a deep learning model. can do.

According to one embodiment, the inner leaf location can be obtained from the crop area using a deep learning model.

For example, as shown in Figure 4, by inputting the crop area into a deep learning model, leaves within the crop area can be identified through segmentation.

The learned deep learning model can output a bounding box surrounding the inner leaf as it receives the crop area. For example, as shown in FIG. 4, the control unit 120 can output a bounding box in three steps, that is, as shown in (a) of FIG. 4, the crop area is converted to gray scale. A deep learning model learned to output the boundary of an object within the preprocessed crop area 410 can be used according to the scale, and by inputting the crop area 410 into the learned deep learning model, the As shown in (b), an object with a predetermined circular shape as a closed curve can be identified as a 'leaf', and the deep learning model learned by inputting the frame and crop area of (b) of Figure 4 together is shown in Figure 4. As shown in (c), the control unit 120 can identify the leaf in the crop area 430 by providing an output value surrounding the identified leaf 431 with a bounding box.

As described above, the control unit 120 can train the deep learning model to output a bounding box surrounding the inner leaf by inputting the crop area so that the deep learning model can output the result.

In addition, the control unit 120 can set the midpoint, vertex, or arbitrary point within the bounding box surrounding the identified leaf as the position of the inner leaf, or a closed curve estimated to be a leaf based on the fact that the leaf is usually connected to the stem. The point where the lines presumed to be the stem meet can be set as the location of the inner leaf.

According to another embodiment, the control unit 120 may identify leaf veins from the crop area using a deep learning model. For example, the boundary of an object within the crop area 410 can be identified through segmentation, and by identifying a line inside the object that has a predetermined circular shape as a closed curve, the line can be identified as a leaf vein.

To this end, the control unit 120 may train a deep learning model to segment and display at least one of the leaf veins and leaves when the crop area is input.

Additionally, the control unit 120 may identify a plurality of leaf veins within a closed curve presumed to be a 'leaf'. Therefore, when 'leaf' and 'leaf vein' are identified using the learned deep learning model, the control unit 120 identifies the position as the position of the inner leaf by, for example, identifying the position where a plurality of leaf veins overlap, or, for example, For example, the main vein (costa) where multiple leaf veins are all connected can be identified and one point at both ends of the main vein can be set as the location of the inner leaf.

Additionally, the control unit 120 can track the growth of the inner leaf stem by labeling the identified inner leaf.

According to one embodiment, if an inner leaf located in an area within a predetermined radius is identified based on the previously identified inner leaf at a time point (t-1) before the analysis immediately before the current time point (t), it is determined that the identified inner leaf has grown, and the inner leaf is determined to have grown. It can be processed to have the same labeling as the identified inner leaves. At this time, the previous time point (t) refers to the time when the inner leaf stem was measured immediately before measuring the current inner leaf stem. For example, the control unit 120 determines that the inner leaf A has grown in the upper inner leaf within a predetermined radius based on the inner leaf A labeled at the previous time point (t-1), and selects the inner leaf at the current time point (t). It can be labeled as 'A', but the inner leaf located above and below within a predetermined radius based on inner leaf A is judged to be a new inner leaf, and the inner leaf can be labeled with new identification information at the current time (t).

In this regard, the deep learning model for determining the location of the stump from the image frame may be referred to as the first deep learning model, and the deep learning model for determining the location of the inner leaf from the image frame may be referred to as the second deep learning model. Each of the first deep learning model and the second deep learning model may be a different network or the same network.

If the position of the inner leaf is identified as described above, the control unit 120 can calculate the length of the inner leaf stem by measuring the distance from the position of the base to the position of the inner leaf.

That is, as shown in FIG. 5, within the image frame 500, the control unit 120 sets a point within the bounding box 510 identified as the stump as the position of the stump and the bounding box 520 identified as the inner leaf. My work point can be determined by the position of the inner leaf.

And according to one embodiment, the control unit 120 can calculate the straight line distance from the position of the stump to the position of the inner leaf as the length of the inner leaf stem. For example, the coordinates of the stump position within the frame and the frame The inner leaf stem length can be calculated by calculating the difference between the inner leaf position coordinates within the inner leaf. Accordingly, as shown in FIG. 5, the control unit 120 can calculate the inner leaf stem length 550.

According to another embodiment, the control unit 120 can identify a stem connected from the position of the base to the position of the inner leaf, and calculate the length of the curved stem using a definite integral.

According to another embodiment, the control unit 120 overlaps the frame at the previous time point (t-1) and the frame at the current time point (t), matches the position of the stump, and accordingly matches the position of the inner leaf at the previous time point (t-1). The difference between the position of the inner leaf at the current time (t) can be calculated. The length of the inner leaf stem at the current time (t) can be calculated by adding the difference to the length of the inner leaf stem at the previous time point (t-1).

In relation to this, when the position of the stump is set at the current time point (t), an error may occur compared to the position of the stump at the previous time point (t-1). For example, the midpoint of the bounding box identified as the base at the current time point (t) can be set to the position of the base, and one vertex of the bounding box identified as the base at the previous time point (t-1) is the position of the base at the current time point (t). The midpoint is set differently depending on the vertex of the bounding box or the length of the horizontal or vertical sides, so that the position of the base at the current time (t) can be set at a different point from the position of the base at the previous time (t-1). . Since it is difficult to move the stump in the near future, a more accurate inner leaf length can be calculated by matching the position of the stump when overlapping the frame at the previous time (t-1) and the frame at the current time (t).

The control unit 120 may reflect the length of the inner leaf stem calculated in this way on the inner leaf stem growth graph.

The inner leaf stem growth graph is a graph showing the change in length of each inner leaf over time and can be updated each time the inner leaf length is calculated.

Additionally, the control unit 120 may provide growth information, which is information about the growth state of the crop, by analyzing the inner leaf stem growth graph. At this time, the growth information may be any information to indicate the growth condition of the crop, and may include, for example, whether the crop's growth condition is good, the growth stage of the crop, etc.

For example, if the slope of the inner leaf stem growth graph converges to 0 for a predetermined period, the control unit 120 may determine that there is a problem with crop growth and provide an alarm to the manager.

Or, for example, the control unit 120 may provide a result predicting how many fruits will be opened by analyzing the inner leaf stem growth graph, or may provide the manager with an alarm indicating the time of irrigation as a growth stage.

Meanwhile, Figure 6 is a flowchart for explaining a method of measuring inner leaf stems according to an embodiment.

The method of measuring inner leaf stems according to the embodiment shown in FIG. 6 includes steps processed in time series in the inner leaf stem measuring device 100 shown in FIGS. 1 to 5. Therefore, even if the content is omitted below, the content described above regarding the inner leaf stem measuring device 100 shown in FIGS. 1 to 5 can also be applied to the inner leaf stem measuring method according to the embodiment shown in FIG. 6.

As shown in FIG. 6, the inner leaf stem measuring device 100 can acquire the position of the stump from an image frame in which a crop is photographed using the learned first deep learning model (S610).

For example, the first deep learning model can be learned to output a bounding box surrounding the base from the image frame, and accordingly, the inner leaf stem measurement device 100 uses the learned first deep learning model to output a bounding box surrounding the base from the image frame. The location of the stump can be identified.

And the inner leaf stem measuring device 100 can acquire the position of the inner leaf from the image frame using the learned second deep learning model (S620).

For example, the inner leaf stem measurement device 100 can acquire the cropped area by cropping the area within the image frame based on the position of the stump, and obtain the inner leaf position by inputting the cropped area into the second deep learning model. there is. At this time, the second deep learning model may be learned to identify inner leaves by identifying leaf veins within the cropped area. In addition, the inner leaf stem measuring device 100 can identify a plurality of leaf veins within the inner leaf and identify positions where the plurality of leaf veins overlap.

Based on the position of the base and the inner leaf obtained as described above, the inner leaf stem measurement device 100 can calculate the length of the inner leaf stem (S630).

For example, the inner leaf stem measuring device 100 overlaps the frame at the previous time point (t-1) with the frame at the current time point (t), matches the position of the base, and accordingly matches the frame of the inner leaf at the previous time point (t-1). The difference between the position and the position of the inner leaf at the current time (t) can be calculated. The length of the inner leaf stem at the current time (t) can be calculated by adding the difference to the length of the inner leaf stem at the previous time point (t-1).

In addition, the inner leaf stem measuring device 100 can provide growth information by reflecting the calculated length of the inner leaf stem in the inner leaf stem growth graph (S640).

Therefore, for example, if there are elements necessary for the healthy growth of the crop, the inner leaf stem measuring device 100 can request the manager through an alarm, etc., and if the growth rate of the inner leaf to be analyzed is slower than the average growth rate, the inner leaf stem measuring device 100 can make a request to the manager through an alarm, etc. You can determine that management is necessary and provide an alarm to the manager.

The term '~unit' used in the above embodiments refers to software or hardware components such as FPGA (field programmable gate array) or ASIC, and the '~unit' performs certain roles. However, '~part' is not limited to software or hardware. The '~ part' may be configured to reside in an addressable storage medium and may be configured to reproduce on one or more processors. Therefore, as an example, '~ part' refers to components such as software components, object-oriented software components, class components, and task components, processes, functions, properties, and procedures. , subroutines, segments of program code, drivers, firmware, microcode, circuits, data, databases, data structures, tables, arrays, and variables.

The functions provided within the components and 'parts' may be combined into a smaller number of components and 'parts' or may be separated from additional components and 'parts'.

In addition, the components and 'parts' may be implemented to regenerate one or more CPUs within the device or secure multimedia card.

The method for measuring inner leaf length according to the embodiment may also be implemented in the form of a computer-readable medium that stores instructions and data executable by a computer. At this time, instructions and data can be stored in the form of program code, and when executed by a processor, they can generate a certain program module and perform a certain operation. Additionally, computer-readable media can be any available media that can be accessed by a computer and includes both volatile and non-volatile media, removable and non-removable media. Additionally, computer-readable media may be computer recording media, which are volatile and non-volatile implemented in any method or technology for storage of information such as computer-readable instructions, data structures, program modules, or other data. It can include both volatile, removable and non-removable media. For example, computer recording media may be magnetic storage media such as HDDs and SSDs, optical recording media such as CDs, DVDs, and Blu-ray discs, or memory included in servers accessible through a network.

Additionally, the method for measuring inner leaf stems according to the embodiment may be implemented as a computer program (or computer program product) including instructions executable by a computer. A computer program includes programmable machine instructions processed by a processor and may be implemented in a high-level programming language, object-oriented programming language, assembly language, or machine language. . Additionally, the computer program may be recorded on a tangible computer-readable recording medium (eg, memory, hard disk, magnetic/optical medium, or solid-state drive (SSD)).

Therefore, the method for measuring inner leaf stems according to the embodiment can be implemented by executing the computer program as described above by a computing device. The computing device may include at least some of a processor, memory, a storage device, a high-speed interface connected to the memory and a high-speed expansion port, and a low-speed interface connected to a low-speed bus and a storage device. Each of these components is connected to one another using various buses and may be mounted on a common motherboard or in some other suitable manner.

Here, the processor can process instructions within the computing device, such as displaying graphical information to provide a graphic user interface (GUI) on an external input or output device, such as a display connected to a high-speed interface. These may include instructions stored in memory or a storage device. In other embodiments, multiple processors and/or multiple buses may be utilized along with multiple memories and memory types as appropriate. Additionally, the processor may be implemented as a chipset consisting of chips including multiple independent analog and/or digital processors.

Memory also stores information within a computing device. In one example, memory may be comprised of volatile memory units or sets thereof. As another example, memory may consist of non-volatile memory units or sets thereof. The memory may also be another type of computer-readable medium, such as a magnetic or optical disk.

And the storage device can provide a large amount of storage space to the computing device. A storage device may be a computer-readable medium or a configuration that includes such media, and may include, for example, devices or other components within a storage area network (SAN), such as a floppy disk device, a hard disk device, an optical disk device, Or it may be a tape device, flash memory, or other similar semiconductor memory device or device array.

The above-described embodiments are for illustrative purposes, and those skilled in the art will recognize that the above-described embodiments can be easily modified into other specific forms without changing the technical idea or essential features of the above-described embodiments. You will understand. Therefore, the above-described embodiments should be understood in all respects as illustrative and not restrictive. For example, each component described as unitary may be implemented in a distributed manner, and similarly, components described as distributed may also be implemented in a combined form.

The scope sought to be protected through this specification is indicated by the patent claims described later rather than the detailed description above, and should be interpreted to include the meaning and scope of the claims and all changes or modified forms derived from the equivalent concept. .

100: Inner leaf stem measuring device
110: input/output unit 120: control unit
130: Communication unit 140: Memory

Claims (12)

A device for measuring the inner leaf stem of a crop,
An input/output unit for acquiring an image;
Memory in which a program for performing deep learning is stored; and
It includes a control unit that performs deep learning on image frames by executing the program,
The control unit,
When the position of the stump is obtained from an image frame in which a crop is photographed using the learned first deep learning model, and the position of the inner leaf is obtained from the image frame using the learned second deep learning model, the position of the stump and Calculate the length of the inner leaf stem based on the position of the inner leaf,
An inner leaf stem measurement device that acquires the cropped area by cropping the area within the image frame based on the position of the stump and acquires the inner leaf position by inputting the cropped area into the second deep learning model. delete According to paragraph 1,
The control unit,
An inner leaf stem measurement device that trains the second deep learning model to identify inner leaves by identifying leaf veins within the crop area. According to paragraph 3,
The control unit,
An inner leaf stem measuring device that identifies the position of the inner leaf by identifying a plurality of leaf veins and identifying a position where the plurality of leaf veins overlap. According to paragraph 1,
The control unit,
An inner leaf stem measuring device that provides growth information by reflecting the calculated inner leaf stem length on an inner leaf stem growth graph. An inner leaf stem measuring device is a method for measuring the inner leaf stem of a crop,
Obtaining the position of the stump from an image frame in which a crop is photographed using the learned first deep learning model;
Obtaining the position of the inner leaf from the image frame using a learned second deep learning model; and
Comprising the step of calculating the length of the inner leaf stem based on the position of the stump and the position of the inner leaf,
The step of acquiring the position of the inner leaf is,
Obtaining a cropped area by cropping an area within the image frame based on the position of the stump; and
A method of measuring inner leaf stems, comprising the step of acquiring the inner leaf position by inputting the cropped area into the second deep learning model. delete According to clause 6,
The second deep learning model is,
A method of measuring inner leaf stems, which is learned to identify inner leaves by identifying leaf veins within the crop area. According to clause 8,
The step of acquiring the position of the inner leaf is,
A method for measuring inner leaf stems, comprising identifying the position of the inner leaf by identifying a plurality of leaf veins and identifying a position where the plurality of leaf veins overlap. According to clause 6,
A method of measuring inner leaf stems, further comprising providing growth information by reflecting the calculated length of inner leaf stems on a growth graph of inner leaf stems. A computer-readable recording medium on which a program for performing the method according to claim 6 is recorded. A computer program stored on a medium for performing the method according to claim 6, which is performed by an inner leaf stem measurement device.