KR20220059820A - A device for measuring crop growth using deep learning segmentation recognition technology and operation method thereof - Google Patents

Abstract

The present invention relates to a device for measuring crop growth using deep learning segmentation recognition technology and a method thereof and, more particularly, to a technology for determining the growth length of a crop and the diameter of a flower using deep learning segmentation. According to one embodiment of the present invention, a method for measuring crop growth comprises the following steps: receiving a video in which a crop is filmed; extracting stem segmentation information of a measurement target of the crop recognized in the video based on a first extraction parameter; extracting flower box information of the measurement target of the crop recognized in the video based on a second extraction parameter; and measuring at least one of the growth length of the crop and the stem thickness near a flower, based on intersection information of the stem segmentation information and the flower box information.

Description

A DEVICE FOR MEASURING CROP GROWTH USING DEEP LEARNING SEGMENTATION RECOGNITION TECHNOLOGY AND OPERATION METHOD THEREOF

The present invention relates to an apparatus and method for measuring crop growth using deep learning segmentation recognition technology, and more particularly, to a technology for obtaining a growth length of a crop and a stem thickness near a flower room using deep learning segmentation.

Recently, the Internet, short-range wireless communication technology, image processing, and various sensing ICT (Information & Communication Technology) complex technologies are being applied to crop cultivation and growth monitoring systems and are being developed. In addition, greenhouse monitoring and management systems that can manage the growing environment of crops are also developing together.

However, it takes a lot of labor and time for the operator to visually check and record the growth state of each stage for a large amount of crops at regular intervals. In addition, when the crop grows higher than the worker's height or when the worker passes by without checking the growth status of a specific stage in a specific crop among a large number of crops, it is difficult to accurately grasp the growth status of each stage of the crop. happened.

In addition, when measuring the height of crops, if the total height of some crops exceeds 2m, the angle of view is not captured, so there is a problem in that the height of the crops cannot be determined after a certain cultivation period. In addition, in image recognition after photographing, if crops that are dense by colony are photographed, when crops overlap, it is difficult to accurately recognize crop organs. In addition, there is no system capable of predicting the change in crop growth or diagnosing crop conditions by collecting currently measured crop data.

An object of the present invention is to provide a crop growth apparatus and method for obtaining a growth length of a crop and a stem thickness in the vicinity of a flower room using deep learning segmentation.

In the method for measuring crop growth according to an embodiment of the present invention, the step of receiving an image of a crop, based on a first extraction parameter, extracting stem segmentation information of a measurement target of the crop recognized in the image step, extracting the flower box information of the measurement target of the crop recognized from the image based on the second extraction parameter, and the growth length of the crop based on the intersection information of the stem segmentation information and the flower room box information And measuring at least one of the thickness of the stem in the vicinity of the flower room.

In one embodiment of the present invention, the step of measuring at least one of the growth length of the crop and the thickness of the stem in the vicinity of the flower room comprises measuring the difference between the highest point of the stem segmentation information and the lowest point of the flower room box information. It may include calculating the growth length.

In one embodiment of the present invention, measuring at least one of the growth length of the crop and the thickness of the stem near the flower room includes measuring the length of the intersection line generated at the intersection of the stem segmentation information and the flower room box information Thus, the method may further include calculating the thickness of the stem near the flower room of the crop.

In one embodiment of the present invention, after receiving the image, based on a threshold value preset for each crop, the method may further include recognizing the target crop for growth measurement in the image.

In one embodiment of the present invention, the method may further include a labeling step of labeling separately a stem and a flower room among the measurement objects of the crop, and in the labeling step, at least a part of the lowest point of the flower room box information is the stem segmentation It is possible to cross the stem of the crop according to the result of the information.

In one embodiment of the present invention, in the step of measuring at least one of the growth length of the crop and the thickness of the stem near the flower room, the change in the growth of the crop is measured in real time through a fixed camera, and compared with a reference value. , it is possible to measure the growth length of the crop.

In one embodiment of the present invention, in the step of measuring at least one of the growth length of the crop and the thickness of the stem near the flower room, the image and the absolute size of the crop are received by using a depth camera among the cameras, At least one of a growth length of a crop and a thickness of a stem near the flower room may be extracted.

In an embodiment of the present invention, the first extraction parameter includes object information used for stem segmentation of the measurement target, and the second extraction parameter is an object used to extract information about a video room of the measurement target Including information, the method may include updating the first and second parameters by using at least one of the information of the crossing line and the crop.

A crop growth measuring apparatus according to an embodiment of the present invention is an image input unit receiving an image of a crop, based on a first extraction parameter, a stem for extracting stem segmentation information of a measurement target of the crop recognized in the image A segmentation unit, a flower room box work unit for extracting flower box information of the crop recognized from the image based on a second extraction parameter, and growth of the crop based on intersection information of the stem segmentation information and the flower room box information It may include a growth calculator that measures at least one of the length and the thickness of the stem near the flower room.

In one embodiment of the present invention, the growth calculator may calculate the growth length of the crop by measuring a difference between the highest point of the stem segmentation information and the lowest point of the flower box information.

In one embodiment of the present invention, the growth calculation unit may calculate the thickness of the stem near the flower room by measuring the length of the intersection line generated at the intersection of the stem segmentation information and the flower room box information.

In an embodiment of the present invention, a crop labeling unit for classifying and labeling a stem and a flower room among the measurement objects of the crop may be further included using the stem segmentation information and the flower room box information, and in the crop labeling unit , at least a part of the lowest point of the flower box information may cross the stem of the crop according to the result of the stem segmentation information.

In one embodiment of the present invention, the growth calculation unit may measure the growth length of the crop by measuring a change in the growth of the crop in real time through a fixed camera and comparing it with a reference value.

A crop growth measurement system according to an embodiment of the present invention includes a crop growth measurement device, a camera for photographing a crop, and a user terminal displaying information of the crop growth measurement device, wherein the crop growth measurement device includes: An image input unit that receives an image of a photograph of A flower room box work unit for extracting the recognized flower room box information of the crop, and the growth length measuring at least one of the growth length of the crop and the thickness of the stem near the flower room based on the intersection information of the stem segmentation information and the flower room box information Includes calculator.

The computer program according to an embodiment of the present invention includes the steps of receiving an image of a crop, extracting stem segmentation information of a measurement target of the crop recognized from the image, based on a first extraction parameter, a second extracting the flower box information of the crop recognized from the image based on the extraction parameter, based on the intersection information of the stem segmentation information and the flower room box information, at least one of the growth length of the crop and the thickness of the stem near the flower room One is stored in the medium for executing the measuring step.

The method for measuring crop growth according to an embodiment of the present application can accurately and quickly obtain the growth length of a crop and the thickness of a stem near the flower room.

1 is a diagram of a method for measuring a conventional crop growth state.
2 is a flowchart illustrating a method for measuring crop growth according to an embodiment of the present invention.
3 is a flowchart illustrating a step of recognizing a growth measurement target crop according to an embodiment of the present invention.
4 is a flowchart illustrating the steps of measuring at least one of a growth length of a crop and a stem thickness near a flower room according to an embodiment of the present invention.
5 is a block diagram of an apparatus for measuring crop growth according to an embodiment of the present invention.
6 is a view showing a thickness measuring method according to an embodiment of the present invention.
7 is a view showing a crop growth measurement system according to an embodiment of the present invention.

Hereinafter, embodiments disclosed in the present specification will be described in detail with reference to the drawings. Unless otherwise specified in the text, like reference numbers in the drawings indicate like elements. Exemplary embodiments described above in the detailed description, drawings, and claims are not intended to be limiting, and other embodiments may be used, and other changes may be made without departing from the spirit or scope of the technology disclosed herein. Those skilled in the art may arrange, construct, combine, and design in variously different configurations the elements of the present disclosure, i.e., the elements generally described herein and illustrated in the drawings, all of which are clearly devised, and the present disclosure It can be easily understood that it forms a part of In order to clearly express various layers (or films), regions, and shapes in the drawings, the width, length, thickness, or shape of the components may be exaggerated.

Since the description of the disclosed technology is merely an embodiment for structural or functional description, the scope of the disclosed technology should not be construed as being limited by the embodiment described in the text. That is, since the embodiment may have various changes and may have various forms, it should be understood that the scope of the disclosed technology includes equivalents capable of realizing the technical idea.

The singular expression is to be understood as including the plural expression unless the context clearly dictates otherwise, and terms such as “comprises” or “have” refer to the embodied feature, number, step, action, component, part or these It is to be understood that this is intended to indicate that a combination exists, and does not preclude the possibility of addition or existence of one or more other features or numbers, steps, operations, components, parts, or combinations thereof.

All terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the disclosed technology belongs, unless otherwise defined. Terms defined in commonly used dictionaries should be interpreted as being consistent with the meanings in the context of the related art, and cannot be interpreted as having ideal or excessively formal meanings unless explicitly defined in the present application.

1 is a view of a method for measuring the state of the conventional crop growth.

Referring to FIG. 1 , conventionally, the growth point and flower room of a crop were recognized using a deep learning box recognition method. The green box at the top is when the growth point is recognized, and the green box at the bottom is when the flower room is recognized. However, when using this method to recognize the growth length of crops and the thickness of stems near the flower room, an additional algorithm is required because of a large error. . In addition, in the case of the segmentation algorithm, there may be a problem of measuring the harvest each time. In addition, since it is difficult to measure only the size of the fruit and the location of the fruit to be measured each time may be different, there may be a problem that it is difficult to obtain a uniform measurement method.

Accordingly, the present invention will be described later on a crop growth measuring apparatus and method for measuring a growth degree by combining stem segmentation information and flower box information.

2 is a flowchart illustrating a method for measuring crop growth according to an embodiment of the present invention.

Referring to FIG. 2 , the crop growth measurement method includes a step of receiving an image of a crop (S11), a step of recognizing a target crop for growth measurement in the image (S13), and a step of extracting stem segmentation information of the crop (S15) , extracting flower box information of the crop (S17), and measuring at least one of the growth length of the crop and the thickness of the stem near the flower room (S19).

In an embodiment, the thickness of the stem near the flower room may be the thickness of the stem near the first flower room that has bloomed. That is, it does not mean the thickness of the stem of the flower room itself, but the thickness of the stem portion in contact with the flower room or the thickness of the stem portion at which the flower room blooms.

In one embodiment, the growth length of the crop may be the length from the growth point to the part of the stem where the flower room is opened, where the flower room may mean the flower room that first opened, that is, the flower room closest to the growth point of the stem.

In the step S11 of receiving an image of a crop photographed, an image of a crop photographed through a camera may be input. At this time, the cameras are IP cameras, HD-SDI cameras, analog cameras, fire detection color cameras, thermal imaging cameras, HD (1920x5080, HD5080p) cameras at the resolution of SD (720x486, NTSC), IP zoom speed cameras, CCTV cameras, It may be at least one of depth cameras. The camera can be fixed and installed, and changes in the growth of crops can be checked in real time.

In the step of recognizing the target crop for growth measurement in the image ( S13 ), the target crop for growth measurement in the image may be recognized based on a preset threshold value for each crop. Since crops in the image do not grow uniformly, it is necessary to recognize the crop to be measured in the image. In one embodiment, deep learning technology may be used to recognize crops. In order to use deep learning technology, it is necessary to collect pictures of target crops from various angles and positions. In addition, in order to increase accuracy, by using an imaging technique, the number of photos of the target crop may be increased to x n for each size and angle. Through the above method, it is possible to recognize the target crop for growth measurement in the image. In this regard, specific details will be described later with reference to FIG. 3 .

In the step of extracting the stem segmentation information of the crop ( S15 ), stem segmentation information of the measurement target of the crop recognized from the image may be extracted based on the first extraction parameter. In an embodiment, the stem segmentation information may be a stem of a segmented crop. That is, the stem segmentation information may be information obtained by extracting only the stem of a crop. The first extraction parameter may include object information used for stem segmentation of the measurement target.

In the step of extracting the flower room box information of the crop (S17), based on the second extraction parameter, the flower room box information of the measurement target of the crop recognized in the image may be extracted. According to an embodiment, the flower room box information may be information obtained by box-processing flowers and flower stems. The second extraction parameter may include object information used to extract information about a room to be measured.

In one embodiment, the order of extracting the stem segmentation information of the crop (S15) and extracting the flower room box information of the crop (S17) may be changed. For example, after performing the step (S17) of extracting the flower box information of the crop, the step of extracting the stem segmentation information of the crop (S15) may be executed.

The step of measuring at least one of the growth length of the crop and the thickness of the stem near the flower room (S19) is to measure at least one of the growth length of the crop and the thickness of the stem near the flower room based on the intersection information of the stem segmentation information and the flower box information. can

In an embodiment, by measuring the difference between the highest point of the stem segmentation information and the lowest point of the flower box information, the growth length of the crop may be measured.

In another embodiment, the growth length of the crop may be measured by measuring a change in the growth of a crop in real time through a fixed camera and comparing it with an existing reference value. For example, the existing reference value may be the growth value of the previous day's crop.

In one embodiment, the thickness of the crop may be measured by measuring the length of the intersection line generated at the intersection of the stem segmentation information and the flower box information. In one embodiment, the crossing line is created in the main stem of the crop.

In an embodiment, the depth camera may be used to receive an image and an absolute size of a crop, and measure at least one of a growth length of a crop and a thickness of a stem near a flower room to increase the accuracy of growth measurement.

In an embodiment, the crossing line and the crossing information may be determined based on preset threshold information. In addition, measurement results such as crossing lines and information on crops may go through a learning process using a deep learning algorithm to update the first and second extraction parameters. By updating the first and second extraction parameters, it is possible to more accurately extract the intersection line.

In this regard, specific details will be described later with reference to FIG. 4 .

The method for measuring crop growth may further include a labeling step of classifying and labeling a stem and a flower room among measurement targets of a crop. In the labeling step, at least a part of the flower box information may be crossed with the stem of the crop according to the result of the stem segmentation information. In this case, the part intersecting the stem of the crop may be the lowest point of the flower box information.

The step of extracting the stem segmentation information of the crop ( S15 ) and the step of extracting the flower room box information of the crop ( S17 ) may extract stem segmentation information and the flower room box information, respectively, based on the labeling step.

3 is a flowchart illustrating a step of recognizing a growth measurement target crop according to an embodiment of the present invention.

Referring to FIG. 3 , the step of recognizing the growth measurement target crop (S13) is the step of setting the recognition threshold and the classification criterion of the crop (S131), the loading of the learned file (S133), and comparing the learned file with the current image It may include a step of preparing for (S135), and a step of determining whether the object accuracy is greater than a threshold (S137).

In the recognition threshold and crop classification criteria setting step S131, threshold values may be set for recognition for each crop, and criteria for classifying each crop type, composition, etc. may be set. In this case, the composition of the crop may be at least one of a growth point, a flower room, and a stem.

The learned file loading step ( S133 ) may be a step of loading an image of a crop learned with the first extraction parameter and the second extraction parameter. The first extraction parameter and the second extraction parameter are based on the criteria set in step S131 . It may be a learned algorithm. For example, the first extraction parameter and the second extraction parameter may be algorithms based on segmentation tools such as R-CNN and YOLACT. The first extraction parameter and the second extraction parameter may be trained and updated with a deep learning algorithm such as a Restricted Boltzmann Machine (M), an Autoencoder, a Convolutional Neural Network, a Recurrent Neural Network, and a Deep Q-Network.

Preparing to compare the learned file with the current image (S135) prepares to compare the step (S11) of receiving the image of the current crop and the step (S133) of receiving the file learned based on a preset standard (S133) It may be a step to

The step of determining whether the object accuracy is greater than the threshold (S137) is to determine whether the accuracy of the object is greater than a preset recognition threshold, and when the accuracy of the object is greater than the preset recognition threshold, it is determined that the growth measurement target crop has been accurately recognized can do. In this case, it may proceed to the step of extracting the stem segmentation information of the crop (S15). However, when the accuracy of the target is less than a preset recognition threshold, it may be determined that the recognition of the target crop for growth measurement is not accurate. In this case, it is possible to return to the preparation step (S135) for comparing the learned file and the current image.

4 is a flowchart specifically illustrating the step of measuring at least one of a growth length of a crop and a stem thickness near a flower room according to an embodiment of the present invention.

Referring to FIG. 4 , the method may include extracting the lowest point of flower box information ( S191 ), extracting the highest point of stem segmentation information ( S193 ), and calculating the growth length ( S195 ).

The step of extracting the lowest point of the flower box information (S191) may be a step of extracting the portion where the flower room box information and the stem segmentation information intersect since the stem of the crop and at least a part of the flower box overlap in the labeling step. According to an embodiment, the lowest point part at the bottom among the intersecting parts may be extracted.

The step of extracting the highest point of the stem segmentation information ( S193 ) may be a step of extracting the growth point of the crop. The highest point of the stem segmentation information is the top of the stem, and may be the growth point of the crop.

In an embodiment, the order of extracting the lowest point of the video box information ( S191 ) and the step of extracting the highest point of the stem segmentation information ( S193 ) may be changed from each other. For example, the step of extracting the highest point of the stem segmentation information ( S193 ) may be performed first, and the step of extracting the lowest point of the picture room box information ( S191 ) may be performed.

The step of calculating the growth length ( S195 ) may be calculating [the highest point of the stem segmentation information - the lowest point of the flower box information]. That is, it may be a step of calculating the difference from the location of the growth point to the cross line.

In an embodiment, the lowest point and the highest point may be numerically calculated, respectively, and a numerical value may be extracted from the image result by comparing the images.

In the case of the conventional segmentation algorithm, there may be a problem of measuring the harvest each time. In addition, since it is difficult to measure only the size of the fruit and the location of the fruit to be measured each time may be different, there may be a problem that it is difficult to obtain a uniform measurement method. However, in the case of the present invention, by learning the first and second extraction parameters with a deep learning algorithm, it is possible to extract the growth length of a crop simply and accurately automatically.

The step of measuring the thickness of the stem near the room may include extracting the lowest point of the box information of the room ( S193 ) and calculating the thickness of the stem near the room ( S199 ).

The step of extracting the lowest point of the flower box information (S191) may be the same as the step of extracting the lowest point of the flower box information required to measure the growth length of a crop (S191).

The step of calculating the thickness of the stem near the flower room ( S199 ) may be calculating the length of the intersection line between the lowest point of the flower room box information and the stem segmentation information. That is, the length of the intersection of the lowest point of the picture box information and the stem according to the stem segmentation information may be measured. The intersecting line may be a line representing the intersection of the stem part of the crop recognized through segmentation and the lowest point of the flower box information.

In the case of the conventional box recognition, an additional algorithm is required because of a large error, and since the growth of crops is different for each situation, a standard algorithm is required, so there is a problem in that it is difficult to commercialize. However, in the case of the present invention, it is possible to simply and accurately obtain the thickness of the stem near the flower room.

As described above, it is possible to more accurately obtain the growth length and thickness of a crop through the crop growth apparatus and method of the present invention. In addition, it is possible to automate the measurement part that requires a lot of human work, so that it is possible to easily acquire data required for crop growth analysis.

5 is a block diagram of an apparatus for measuring crop growth according to an embodiment of the present invention.

Referring to FIG. 5 , the crop growth measuring device 10 includes an image input unit 100 , a measurement target recognition unit 200 , a stem segmentation unit 400 , a flower box working unit 500 , and a growth calculation unit 600 . may include

The image input unit 100 may receive an image of crops. The image may be captured through a camera. At this time, the cameras are IP cameras, HD-SDI cameras, analog cameras, fire detection color cameras, thermal imaging cameras, HD (1920x5080, HD5080p) cameras at the resolution of SD (720x486, NTSC), IP zoom speed cameras, CCTV cameras, It may be at least one of depth cameras. The camera can be fixed and installed, and changes in the growth of crops can be checked in real time. In one embodiment, the Depth camera may provide an absolute size of the crop.

The measurement target recognition unit 200 may recognize a growth measurement target crop from the image based on a threshold value preset for each crop. Since the crops in the image do not grow uniformly, it is necessary to recognize the crop to be measured in the image.

In one embodiment, deep learning technology may be used to recognize crops. In order to use deep learning technology, it is possible to collect pictures of target crops from different angles and positions. In addition, in order to increase accuracy, by using an imaging technique, the number of photos of the target crop may be increased to x n for each size and angle.

The apparatus for measuring crop growth may further include a crop labeling unit 300 for labeling a stem and a flower room among measurement targets of crops. In the crop labeling unit 300 , at least a part of the flower box information may cross the stem of the crop according to the result of the stem segmentation information. In this case, the part that intersects the stem part of the crop may be the lowest point of the flower box information.

The stem segmentation unit 400 may extract stem segmentation information of the crop to be measured for growth recognized in the image, based on the first extraction parameter. In an embodiment, the stem segmentation information may be a stem of a segmented crop. That is, the stem segmentation information may be information obtained by extracting only the stem of a crop.

The flower room box work unit 500 of the crop may extract information on the flower room box of the measurement target of the crop recognized in the image, based on the second extraction parameter. According to an embodiment, the flower room box information may be information obtained by box-processing flowers and flower stems.

The stem segmentation unit 400 and the crop box work unit 500 may extract stem segmentation information and flower room box information, respectively, based on the crop labeling unit 300 .

The growth calculator 600 may measure at least one of a growth length of a crop and a thickness of a stem in the vicinity of the flower room based on the intersection information of the stem segmentation information and the flower room box information.

In an embodiment, by measuring the difference between the highest point of the stem segmentation information and the lowest point of the flower box information, the growth length of the crop may be measured.

In one embodiment, by measuring the change in the growth of a crop in real time through a fixedly installed camera, and comparing it with an existing reference value, the growth length of the crop may be measured. For example, the existing reference value may be the growth value of the previous day's crop.

In one embodiment, the thickness of the crop may be measured by measuring the length of the intersection line generated at the intersection of the stem segmentation information and the flower box information. The crossing line may be a distance from a stem portion close to the flower room to a stem portion farther from the flower room. For example, the intersection line may be determined as the shortest distance from the intersection point to the stem.

In an embodiment, the depth camera may be used to receive an image and an absolute size of a crop, and measure at least one of a growth length of a crop and a thickness of a stem near a flower room to increase the accuracy of growth measurement.

In an embodiment, the crossing line and the crossing information may be determined based on preset threshold information. In addition, information on crops and measured information such as cross lines may be updated in the first and second extraction parameters.

As used herein, the term “unit” or “module” refers to a hardware component such as software, FPGA, or ASIC, and “unit” or “module” performs certain roles. However, “part” or “module” is not meant to be limited to software or hardware. A “unit” or “module” may be configured to reside on an addressable storage medium or configured to reproduce one or more processors. Thus, as an example, “part” or “module” refers to components such as software components, object-oriented software components, class components and task components, processes, functions, properties, Includes procedures, subroutines, segments of program code, drivers, firmware, microcode, circuitry, data, databases, data structures, tables, arrays, and variables. Components and functionality provided within “parts” or “modules” may be combined into a smaller number of components and “parts” or “modules” or as additional components and “parts” or “modules”. can be further separated.

6 is a view showing a method for measuring the thickness of a stem near a flower room according to an embodiment of the present invention.

Referring to FIG. 6 , the thickness of the stem near the flower room of the crop may be determined based on the stem 410 and the flower room 510 . For convenience, the leaves except for the stems and flower beds of crops have been omitted.

As can be seen in FIG. 6 , the thickness of the crop can be measured by measuring the length of the crossing line 415 generated at the intersection of the stem 410 and the flower room 510 . In this case, the intersection line 415 may be an intersection of the stem of the crop extracted from the stem segmentation and the lowest point of the flower box information. That is, W is the length of the crossing line, and may be the thickness of the stem near the flower room.

In an embodiment, the stem may include a first line 411 and a second line 412 , and the second line 412 may be a line intersecting the picture room box information 510 . The crossing line 415 may be determined as the distance from the second line 412 to the lowest point of the video box information 510 and the first line 411 .

In an embodiment, the thickness of the stem near the flower room may be the thickness of the stem near the first flower room that has bloomed. That is, it does not mean the thickness of the stem of the flower room itself, but the thickness of the stem portion in contact with the flower room or the thickness of the stem portion at which the flower room blooms. The stem thickness W in the vicinity of the flower room may indicate a portion where the stem segmentation information 410 and the flower room box information 510 intersect as a line on the stem of the crop. In this case, the line may be the shortest length among the lengths that can be displayed on the stem.

In one embodiment, the growth length of the crop may be the length from the growth point to the part of the stem where the flower room is opened, where the flower room may mean the flower room that first opened, that is, the flower room closest to the growth point of the stem. The growing length of the crop is H, which may be the shortest length from the crossing line to the growing point.

7 is a view showing a crop growth measurement system according to an embodiment of the present invention.

Referring to FIG. 7 , the crop growth measurement system may include a crop growth measurement device 10 , a camera 30 , and a user terminal 50 .

The crop growth measuring device 10 may be the same as the crop growth measuring device 10 illustrated in FIG. 5 .

The camera 30 may be for photographing crops. The camera 30 is an IP camera, HD-SDI camera, analog camera, fire detection color camera, thermal camera, HD (1920x5080, HD5080p) camera, IP zoom speed camera, CCTV camera at the resolution of SD (720x486, NTSC) level. , may be at least one of a depth camera. The camera can be fixed and installed, and changes in the growth of crops can be checked in real time. At this time, the depth camera can obtain the absolute size of the crop.

The user terminal 50 may display information of the crop growth measuring device. The grower can easily check the growth state of the growth measurement target of the currently growing crop through his or her user terminal 50 . Accordingly, the grower can determine whether the crop is growing normally, and if it is determined that the growth is not performed smoothly, the cultivation process can be changed, such as supplying a larger amount of fertilizer or nutrients.

The user terminal 50 is a broad concept including a wired terminal or a wireless terminal, a personal computer (PC), an Internet protocol television (IP), a notebook-sized personal computer, a personal digital assistant (PDA), Smartphone, IMT-2000 (International Mobile Telecommunication 2000) phone, GSM (Global System for Mobile Communication) phone, GPRS (General Packet Radio Service) phone, WCDMA (Wideband Code Division Multiple Access) phone, UMTS (Universal Mobile Telecommunication Service) phone It may include a phone, a Mobile Broadband System (MBS) phone, and the like.

In an embodiment of the present invention, the computer program includes the steps of receiving an image of a crop photographed into a computer, and extracting stem segmentation information of a measurement target of the crop recognized from the image, based on a first extraction parameter , extracting the flower room box information of the crop recognized from the image based on a second extraction parameter, the growth length of the crop and the stem near the flower room based on the intersection information of the stem segmentation information and the flower room box information It may be stored in the medium to execute the step of measuring at least one of the thickness.

A computer program is not a medium that stores data for a short moment, such as a register, a cache, a memory, etc., but a medium that stores data semi-permanently and can be read by a device. Specifically, examples of the storage medium include, but are not limited to, ROM, RAM, CD-ROM, magnetic tape, floppy disk, and optical data storage device. That is, the program may be stored in various recording media on various servers accessible by the computer or in various recording media on the computer of the user. In addition, the medium may be distributed in a computer system connected to a network, and a computer-readable code may be stored in a distributed manner.

10: crop growth measuring device
30 : camera
50: user terminal
100: video input unit
200: measurement target recognition unit
300: crop labeling unit
400: stem segmentation unit
410: stem
411: first line
412: second line
415: intersecting line
500: flower box work department
510 : Hwabang
600: growth calculation unit

Claims (15)

In the method for measuring crop growth,
receiving an input image of crops;
extracting stem segmentation information of a measurement target of the crop recognized from the image based on a first extraction parameter;
extracting, based on a second extraction parameter, information on a flower room of a measurement target of the crop recognized in the image; and
and measuring at least one of a growth length of the crop and a thickness of a stem in the vicinity of a flower room based on cross information of the stem segmentation information and the flower room box information. According to claim 1,
The step of measuring at least one of the growth length of the crop and the thickness of the stem near the flower room,
and calculating the growth length of the crop by measuring a difference between the highest point of the stem segmentation information and the lowest point of the flower box information. 3. The method of claim 2,
The step of measuring at least one of the growth length of the crop and the thickness of the stem near the flower room,
The method further comprising the step of calculating a stem thickness in the vicinity of the flower room of the crop by measuring the length of the intersection line generated at the intersection of the stem segmentation information and the flower room box information. According to claim 1,
After receiving the image,
Crop growth measurement method further comprising the step of recognizing a growth measurement target crop in the image based on a threshold value preset for each crop. According to claim 1,
It further comprises a labeling step of labeling by distinguishing the stem and the flower room among the measurement target of the crop,
In the labeling step, at least a part of the lowest point of the flower box information crosses the stem of the crop according to the result of the stem segmentation information. According to claim 1,
In the step of measuring at least one of the growth length of the crop and the thickness of the stem near the flower room, the growth change of the crop is measured in real time through a fixed camera and compared with a reference value to measure the growth length of the crop A method of measuring crop growth. 7. The method of claim 6,
In the step of measuring at least one of the growth length of the crop and the thickness of the stem near the flower room, the image and the absolute size of the crop are received using a depth camera among the cameras, A method of measuring crop growth by extracting at least one of stem thickness. According to claim 1,
The first extraction parameter includes object information used for stem segmentation of the measurement target,
The second extraction parameter includes object information used to extract the picture room box information of the measurement target,
and updating the first and second parameters using at least one of the crossed line and the crop information. an image input unit for receiving an image of crops;
a stem segmentation unit for extracting stem segmentation information of a measurement target of the crop recognized from the image based on a first extraction parameter;
a flower room box work unit for extracting flower room box information of the crop recognized from the image based on a second extraction parameter; and
and a growth calculator configured to measure at least one of a growth length of the crop and a thickness of a stem in the vicinity of the flower room based on the intersection information of the stem segmentation information and the flower room box information. 10. The method of claim 9,
The growth calculation unit measures the difference between the highest point of the stem segmentation information and the lowest point of the flower box information to calculate the growth length of the crop. 10. The method of claim 9,
The growth calculation unit measures the length of the intersection line generated at the intersection of the stem segmentation information and the flower box information to calculate the thickness of the stem in the vicinity of the flower room. 10. The method of claim 9,
and a crop labeling unit for classifying and labeling a stem and a flower room among the measurement targets of the crop by using the stem segmentation information and the flower room box information,
In the crop labeling unit, at least a part of the lowest point of the flower box information crosses the stem of the crop according to the result of the stem segmentation information. 10. The method of claim 9,
The growth calculation unit measures the growth change of the crop in real time through a fixed camera, and compares it with a reference value, thereby measuring the growth length of the crop. crop growth measuring device;
a camera for photographing crops; and
It includes a user terminal for displaying information of the crop growth measurement device,
The crop growth measuring device,
an image input unit for receiving an image of crops;
a stem segmentation unit for extracting stem segmentation information of a measurement target of the crop recognized from the image based on a first extraction parameter;
a flower room box work unit for extracting flower room box information of the crop recognized from the image based on a second extraction parameter; and
and a growth calculator configured to measure at least one of a growth length of the crop and a thickness of a stem in the vicinity of the flower room based on the intersection information of the stem segmentation information and the flower room box information. on the computer,
receiving an input image of crops;
extracting stem segmentation information of a measurement target of the crop recognized from the image based on a first extraction parameter;
extracting the flower room box information of the crop recognized from the image based on a second extraction parameter;
A computer program stored in a medium for executing the step of measuring at least one of a growth length of the crop and a thickness of a stem in the vicinity of a flower room based on the intersection information of the stem segmentation information and the flower room box information.

Images