KR20210054416A - The smart apparatus and method of determining quality and screening for plant - Google Patents

Abstract

A device and method for distinguishing and selecting a quality of a smart plant according to one embodiment of the present invention may comprise: a shape analysis part that derives an external quality of a plant through scanning of the plant; a spectral analysis part that derives an internal quality of the plant by using a spectral image of the plant; and a plant body identification part wherein a grade information for the plant is discriminated based on the external and internal qualities of the plant. Therefore, the present invention is capable of having an effect of classifying plants.

Description

Device and method of smart plant quality judgment and selection {THE SMART APPARATUS AND METHOD OF DETERMINING QUALITY AND SCREENING FOR PLANT}

The present invention relates to a smart plant quality discrimination and selection device, and more particularly, to a smart plant quality discrimination and selection device in which the grade information of the plant is determined by deriving the internal and external properties of the plant grown through the seedling.

Seedling is an outpost of crop production and is important enough to determine the success or failure of production, and the seedling industry is a future growth engine industry with very high potential for development, as well as a source business that can induce division of labor and specialization of seedlings and cultivation, as well as germination, grafting and It is a precision industry in which cultivation-related technologies such as survival, growth control, and pest management are accumulated.

As for the seedling, the area of the domestic process nursery has expanded from about 20 ha in 1997 to 159 ha in 2010, about 8 times, and the area is expected to increase significantly to 187 ha in 2015 and 224 ha in 2020. very important.

However, the increase in production costs due to climate change, rising labor costs and agricultural material prices, difficulties in the supply and demand of manpower due to the aging of rural areas, and the lack of automation technology for seedling production compared to advanced countries such as the Netherlands and lagging seedling facilities are acting as a major obstacle to the development of the seedling industry. It is necessary to develop intelligent smart seedling technology that can improve seedling productivity and automation through the application of core technologies of the 4th industrial revolution in order to solve the problems of the seedling industry at present and to continue development.

ICBM is a comprehensive precision industry in which all technologies related to cultivation, such as sowing, germination, grafting and sedimentation, growth control, pest management, storage, and transport within a period of up to two months, are integrated through a series of factory-installed processes. The availability of application of AI and robots is very high, and development of an integrated seedling management system for individual smart seedling element technology development and comprehensive verification is continuously required. In addition, in order to apply phenomics, which investigates the phenotype of plants by analyzing big data formed by accumulating data acquired during the seedling process, individual analysis of target plants is required.

In other words, it is an important factor to select good seedlings by determining the quality of plants by analyzing the internal and external characteristics of the plant during the seedling process. In the past, the quality of plants was judged by directly observing plants. However, when the user selects plants with the above subjective factors added, there is a problem that the classified plants may not actually be good hair because of the difference according to the user's determination criteria.

Therefore, it is required to develop a technology for classification by classifying the quality of plants, such as external quality such as leaf area of plants, and internal quality based on spectral data.

Republic of Korea Patent Publication No. 10-2001-0098368 (Publication date: 2001.11.08)

An object of the present invention is to provide a smart plant quality discrimination and selection device capable of classifying plants by accurately determining the external and internal properties of plants that have undergone seedling, as a countermeasure to the above-described problem.

As an embodiment of the present invention, a smart plant quality determination and selection device may be provided.

The smart plant quality determination and selection device according to an embodiment of the present invention includes a shape analysis unit in which the external quality of the plant is derived through scanning of the plant, and a spectral analysis in which the internal quality of the plant is derived using a spectral image of the plant. A plant discrimination unit may be included in which grade information for a plant is determined based on the external and internal properties of the wealth and the plant.

In the smart plant quality determination and selection device according to an embodiment of the present invention, an input unit for supplying plants to the shape analysis unit, a discharge unit for classifying and discharging the plants whose grade information is determined in the plant body determination unit to the packaging unit, and an input unit. A transfer unit provided with a moving module for moving the plant supplied through to the discharge unit may be further included.

In the smart plant quality determination and selection device according to an embodiment of the present invention, the shape analysis unit includes a light source unit for irradiating light toward the plant, a detection unit for identifying light reflected from the plant, and a point group obtained as a result of the identification. A shape analysis module from which shape information of a plant is derived based on (point cloud) may be included.

In the smart plant quality discrimination and selection device according to an embodiment of the present invention, the spectroscopic analyzer includes a spectral sensor for obtaining a spectral image of the plant and a spectral reflectance for each wavelength band extracted from the spectral image to normalize the plant. A spectroscopic analysis module for calculating the vegetation index may be included.

The smart plant quality determination and selection method according to an embodiment of the present invention may include preparing a spectral library from the average of the spectrum obtained from the hyperspectral image.

The smart plant quality determination and selection method according to an embodiment of the present invention may further include the step of matching the plant with the seedling process and transmitting it to a database after producing the spectral library.

According to an embodiment of the present invention, there is an effect of classifying the plant by accurately determining the external quality and the internal quality of a plant that has undergone seedling.

1 is an exemplary view showing a smart plant quality determination and selection device according to an embodiment of the present invention.
Figure 2 is a block diagram showing a smart plant quality determination and selection device according to an embodiment of the present invention.
3 is a flowchart of a method for determining and selecting the quality of a smart plant according to an embodiment of the present invention.
4 is an exemplary view showing a process in which an external quality of a plant is derived from a shape analysis unit in a smart plant quality determination and selection device according to an embodiment of the present invention.
5 is an exemplary view showing a process in which the internal quality of a plant is derived by a spectroscopic analysis unit in the smart plant quality determination and selection device according to an embodiment of the present invention.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings so that those of ordinary skill in the art can easily implement the present invention. However, the present invention may be implemented in various different forms and is not limited to the embodiments described herein. In the drawings, parts irrelevant to the description are omitted in order to clearly describe the present invention, and similar reference numerals are attached to similar parts throughout the specification.

The terms used in the present specification will be briefly described, and the present invention will be described in detail.

Terms used in the present invention have selected general terms that are currently widely used as possible while taking functions of the present invention into consideration, but this may vary according to the intention or precedent of a technician working in the field, the emergence of new technologies, and the like. In addition, in certain cases, there are terms arbitrarily selected by the applicant, and in this case, the meaning of the terms will be described in detail in the description of the corresponding invention. Therefore, the terms used in the present invention should be defined based on the meaning of the term and the overall contents of the present invention, not a simple name of the term.

When a part of the specification is said to "include" a certain component, it means that other components may be further included rather than excluding other components unless specifically stated to the contrary. In addition, terms such as "... unit" and "module" described in the specification mean a unit that processes at least one function or operation, which may be implemented as hardware or software or a combination of hardware and software. . In addition, when a part is said to be "connected" with another part throughout the specification, this includes not only the case of being "directly connected", but also the case of being connected "with another element in the middle."

Hereinafter, the present invention will be described in detail with reference to the accompanying drawings.

As an embodiment of the present invention, a smart plant quality determination and selection device 10 may be provided.

In the present specification, the plant 50 includes not only a part of the plant itself, a plant tissue, or a plant cell, and all plants and trees for cultivation such as seedlings in a seedling process may be included. In addition, the type of cultivation is also irrelevant. For example, differentiated plants according to differentiation cultivation may be included in the plant.

In addition, it is natural that it can be included in the plant regardless of the purpose of cultivation. In addition, the plant is a plant produced by artificial manipulation such as transformation, and may include plants that do not exist originally in the natural world.

1 is an exemplary view showing a smart plant quality determination and selection device 10 according to an embodiment of the present invention, Figure 2 is a smart plant quality determination and selection device 10 according to an embodiment of the present invention. It is a block diagram.

1 and 2, the smart plant quality determination and selection device 10 according to an embodiment of the present invention is a shape analysis unit in which the external quality of the plant 50 is derived through scanning of the plant 50 (100), using the spectral image 212 of the plant 50, based on the spectral analysis unit 200 and the internal quality of the plant 50, the internal quality of the plant 50 is derived, and the plant ( 50) may be included in the plant determination unit 300 for determining the grade information.

That is, the smart plant quality determination and selection device 10 according to an embodiment of the present invention is suitable for seedling or simulated external product for transplanting through a seedling process for a predetermined period of time, such as a sowing process, a growth process, a grafting process, and a cultivation process. It refers to a device that selects good seedlings in consideration of quality and inner quality.

The plant 50 refers to a seedling or a seedling as a bud of a seed cultivated for transplantation, and there is no limitation on the type of the seed.

In addition, the external quality is a feature related to the shape of the plant body 50 as described later, and the external quality includes the leaf area, plant height, number of nodes, side width, side length, etc. of the plant body 50, and the inner quality is the plant body 50 ), such as the degree of vitality, and the internal quality may include the spectral reflectance of the plant 50, the normalized vegetation index, and the degree of pests.

In addition, in the smart plant quality determination and selection device 10 according to an embodiment of the present invention, the input unit for supplying the plant body 50 to the shape analysis unit 100, and the grade information is determined by the plant body determination unit 300 A discharging unit for classifying and discharging the plant 50 to the packaging unit, and a transfer unit having a moving module for moving the plant 50 supplied through the input unit to the discharging unit may be further included.

The plant 50 may be supplied to the transfer module in a predetermined direction from the input unit. That is, in the input unit, the plant body 50 or the plant body port containing the plant body 50 may be aligned and moved to be introduced into the transfer module at predetermined time intervals.

In the discharge unit, the plant body 50 whose grade information is determined by the plant body determination unit 300 may be classified and discharged to the packaging unit. That is, in the discharge unit, the plants 50 classified in consideration of the external and internal characteristics of the plant 50 form a plurality of movement paths and may be transferred to each collection point.

For example, as shown in Figure 1, plants 50 classified as A grade are classified as A grade collection stations 510, plants 50 classified as B grade are classified as B grade collection stations 520, C The plants 50 classified into grades may be moved to the C grade collection points 530 through individual routes. However, the grades A, B, and C are merely exemplary and may be set to various grades.

In addition to the moving module, the transfer unit may further include a drive module providing power to operate the moving module, and a transfer structure including the moving module and the drive module. The moving module may correspond to a conveying device driven by a conveyor method. That is, the moving module may further include a plurality of pulleys and belts for the operation of the conveyor.

Hereinafter, a method of determining and selecting plant quality using the smart plant quality determination and selection device 10 described above will be described.

3 shows a flow chart of a smart plant quality determination and selection method according to an embodiment of the present invention.

Referring to Figure 3, the smart plant quality determination and selection method, the step of recording the start time of the movement of the transfer conveyor for transporting the plant (50) (S10); Obtaining dual lidar data in which external traits for the plant 50 are stored or hyperspectral images in which inner traits are stored (S20); And calculating a 3D mesh from the dual lidar data or analyzing a spectrum obtained from the hyperspectral image to analyze a seedling process for the plant 50 (S30).

The step (S10) of recording the time at which the movement of the transport conveyor for transporting the plant 50 starts (S10) is a process of recording the time from the point at which the transport conveyor operates in a standby state. The transfer conveyor moves at a constant speed, and the scanned plant 50 can be checked based on time and speed.

The step (S20) of acquiring dual lidar data in which the external trait is stored for the plant 50 or a hyperspectral image in which the inner trait is stored (S20) includes an external trait such as a leaf area of a plant, and an inner trait based on spectral data. This is the process of classifying the quality of ).

The step of calculating a 3D mesh from the dual lidar data or analyzing the spectrum obtained from the hyperspectral image to analyze the seedling process for the plant 50 may include analyzing the measured hyperspectral reflected light and shape information to provide a superior plant body. This is the process of determining (50) and outputting or transmitting the diagnosis result data.

In this process, it may include the step (S30) of producing a spectral library from the average of the spectrum obtained from the hyperspectral image. After the production of the spectral library, the step of matching the plant 50 and the seedling process may further include transmitting to a database.

In this regard, according to an embodiment of the present invention, a shape analysis unit 100 and a spectroscopic analysis unit may be included, and each will be described in detail with reference to FIGS. 4 and 5 below.

4 is an exemplary view showing a process in which the external quality of the plant 50 is derived by the shape analysis unit 100 in the smart plant quality determination and selection device 10 according to an embodiment of the present invention.

4, in the smart plant quality determination and selection device 10 according to an embodiment of the present invention, the shape analysis unit 100 includes a light source unit 110 for irradiating light toward the plant 50, A detection unit 120 for identifying the light reflected from the plant 50 and a shape analysis module 130 for deriving shape information of the plant 50 based on the point cloud obtained as a result of the identification may be included. have.

The light source unit 110 and the detection unit 120 may correspond to scan sensors for scanning an object. For example, the light source unit 110 and the detection unit 120 may correspond to a lidar sensor.

The lidar sensor refers to a sensor capable of detecting distance, direction, speed, temperature, material distribution, and concentration characteristics to the object by shining light on the object. That is, the light source unit 110 and the detection unit 120 may correspond to a two-dimensional lidar sensor or a three-dimensional lidar sensor.

In the smart seedling sorting apparatus of the present invention, the light source unit 110 and the detection unit 120 are preferably 3D lidar sensors for acquiring 3D shape information for the plant 50.

That is, the light irradiated from the light source unit 110 through the lidar sensor is reflected from the plant 50 and detected, so that scan information corresponding to a point cloud may be obtained. In addition, the scan information may correspond to data in which data obtained through individual scanning by one lidar sensor or data acquired through a plurality of lidar sensors are combined.

That is, the scan information may be obtained through scanning by at least one lidar sensor.

Various sensors may be used in place of the lidar sensor as the scan sensor. For example, a fluorescence sensor, a thermal image sensor, an infrared sensor, a near infrared sensor, a CCD image sensor, an ultrasonic sensor, and the like can be used.

That is, the shape of the plant 50 may be extracted using at least one of the fluorescence sensor, thermal image sensor, infrared sensor, near infrared sensor, CCD image sensor, and ultrasonic sensor.

After the point group is acquired through 3D scanning, at least one convex hull may be extracted, and a shape model may be generated from the convex hull. The convex hull refers to the smallest convex set including points or regions in Euclidean space. In addition, the shape model 135 may be generated in a three-dimensional mesh structure. The mesh refers to a unit for expressing one object (Ex. plant 50) in three dimensions based on at least one of a point, a line, and a plane.

The shape information of the plant 50 may be derived (S130) using the shape model 135 generated as described above. As described above, the shape information may include at least one of the plant height, leaf length, leaf width and leaf area, and number of nodes of the plant body 50.

The user can grasp the external quality of the plant 50 by using shape data 100 such as the plant height, leaf length, leaf width, leaf area, and number of nodes of the plant body 50 derived according to the above process.

The shape analysis unit 100 may further include a first housing 170. As shown in FIG. 1, a light source unit 110 and a detection unit 120 may be provided at a predetermined position of the first housing 170, and the first housing 170 surrounds a specific portion of the moving module. It can be formed in a shape.

That is, the first housing 170 may be formed to be positioned at the upper end of the moving module to provide a passage for allowing the plant 50 to be moved through the moving module. In addition, as shown in FIG. 1, the light source unit 110 and the detection unit 120 may be attached to the inner upper surface of the housing so as to face the lower end of the housing.

In addition, a first collection point 190 may be additionally included in the shape analysis unit 100. Plants 50 that do not meet predetermined reference shape information about the external quality of the plant 50 derived from the shape analysis unit 100 may be separated and collected by a first collection point 190. That is, when the plant body 50 is deformed, it may be separated into the first collection point 190.

For example, with respect to the plant 50 from which the plant height, leaf area, and number of nodes are derived through the shape analysis module 130, whether or not it satisfies the reference shape information (Ex. It can be judged additionally.

If the plant 50, a leaf area, and the number of nodes are out of the reference plant length range, the reference leaf area range, and the reference node range, the plant body 50 a may be separated into a deformity and moved to the first collection point 190. have.

On the contrary, if the plant 50 and leaf area b meets the standard plant length range and the reference leaf area range, but the number of nodes does not meet only the reference node range, it is not classified as a deformity and is not separated into the first collection point 190. It may be moved to the spectroscopic analysis unit 200 for determining the inner quality to be described later.

However, the above-described method for determining whether the plant body 50 is deformed is merely exemplary and is not limited thereto, and whether or not the plant body 50 is deformed may be determined by various methods.

For example, the shape analysis unit 100 may additionally include a first image sensor (not shown) for determining whether or not it is deformed inside the first housing. The first image sensor may be a photographing device using an image sensor such as a charge coupled device (CCD) or a CMOS image sensor (CIS).

The shape analysis unit 100 divides the first image of the plant 50 obtained through the first image sensor into a plurality of regions, derives the number of pixels of the plant 50 from the divided regions, and Whether or not it is deformed may be determined according to a ratio of the number of pixels for each of a plurality of divided regions.

That is, it is determined whether or not the plant is deformed based on the number of pixels of the plant 50 of the first image acquired through the first image sensor, so that the plant 50 to be separated into the first collection point may be determined.

5 is an exemplary view showing a process of deriving the inner quality of the plant 50 by the spectroscopic analysis unit in the smart plant quality determination and selection device according to an embodiment of the present invention.

5, in the smart plant quality determination and selection device 10 according to an embodiment of the present invention, the spectroscopic analysis unit 200 includes a spectroscopic image for obtaining a spectroscopic image 212 of the plant 50. A spectroscopic analysis module 220 for calculating the normalized vegetation index of the plant 50 based on the spectral reflectance for each wavelength band extracted from the sensor 210 and the spectral image 212 may be included.

The spectral sensor 210 may include a multispectral sensor, a hyperspectral sensor, and the like. However, in the smart plant quality determination and selection device 10 according to an embodiment of the present invention, the spectral sensor 210 is preferably configured as a hyperspectral image sensor.

The hyperspectral image sensor refers to a sensor that records an electromagnetic wave reflected by an object as an electromagnetic wave or radiated from an object as a continuous spectral wavelength of several hundred or more. In addition, data of a continuous spectral wavelength obtained by the hyperspectral image sensor can be used for detection of various objects of land and vegetation.

Various image data may be used instead of the spectral image. The image data may include a fluorescent image, a thermal image, an infrared image, a near infrared image, a CCD image, an ultrasonic image, and the like. That is, at least one of a fluorescence image, a thermal image, an infrared image, a near infrared image, a CCD image, and an ultrasonic image may be used to determine the inner quality of the plant 50.

The spectroscopic image 212 obtained through the spectroscopic sensor 210 as described above may be stored in a database (not shown). That is, the continuous spectral reflectance of the plant 50 may be stored in the database. In addition, the shape information obtained in the above-described process may also be stored in the database.

That is, the spectroscopic image 212 refers to an image acquired using a spectroscopic image sensor such as a hyperspectral sensor. In other words, a spectral image 212 of the plant 50 may be obtained using the hyperspectral image sensor, and the spectral image 212 is based on the spectral reflectance of the plant 50 for each of a plurality of wavelength bands. It may correspond to the generated image.

The wavelength band may include a red wavelength band (600nm ~ 700nm), a blue wavelength band (400nm ~ 500nm), a green wavelength band (500nm ~ 600nm), a near-infrared (NIR) wavelength range (750nm ~ 900nm), etc. have.

However, since the above-described wavelength band is merely exemplary, the present invention is not limited thereto, and the spectroscopic image 212 may be generated based on various wavelength bands according to a user's purpose and use.

In addition, the spectral image 212 may additionally perform pre/post processing such as color space conversion, binarization processing, masking processing, and the like. That is, a color space conversion process may be performed on the spectral image 212.

Information related to an RGB color space expressing colors using light characteristics may be obtained from the obtained spectral image 212, and the RGB data may be converted into an HSL color space for the purpose of image processing.

That is, information on hue, saturation, and lightness may be obtained from RGB pixel data through the color space conversion process.

The binarization of the spectral image 212 may be performed on the basis of information obtained according to the color space conversion. Various known methods can be applied in relation to image binarization. For example, the spectral image 212 may be binarized based on a threshold value obtained according to the Otsu method. The Otsu method refers to a method for finding a threshold value for optimally dividing an image into two classes for an image histogram.

That is, when the image information data (ex. pixel value, brightness, etc.) is classified into two classes based on the threshold value T, the threshold value T that maximizes the value of the between-class variance. Refers to how to find.

As the spectral image 212 is binarized based on a threshold for classification using the aforementioned Otsu method, a binarized image may be generated.

Thereafter, as image masking is performed on the generated binarized image, a masked image may be generated. A known method may be applied in relation to image masking, and a masking image may be obtained as masking is performed so that only the plant 50 is represented in the binarized image.

That is, a masking process may be performed so that only vegetation portions are extracted from the spectroscopic image 212 and the remaining portions (Ex. background, etc.) are hidden.

Next, the normalized vegetation index (NDVI) may be derived through spectrum analysis of the spectral image 212. The normalized vegetation index corresponds to an index for grasping the vegetation characteristics of the plant 50 based on the spectral difference generated between the near-infrared (NIR) region and the visible ray region.

The normalized vegetation index may be used as an index to indicate the degree of vegetation vitality for the plant 50. That is, the user may acquire information related to the degree of vitality of the plant 50 by using the obtained normalized vegetation index. In addition, it is possible to derive whether the spectral image 212 is infected with a pest or not through spectrum analysis.

In addition, a second housing 270 may be further provided in the spectroscopic analysis unit 200. As shown in FIG. 1, a spectroscopic sensor 210 may be provided at a predetermined position of the second housing 270, and the second housing 270 may be formed to surround a specific portion of the moving module. I can.

The second housing 270 may be formed at an upper end of the moving module to provide a passage for allowing the plant 50 to be moved through the moving module. In addition, as shown in FIG. 1, a spectroscopic sensor 210 as well as an additional lighting unit may be attached to the inner upper surface of the housing so as to face the lower end of the housing.

The lighting unit 205 serves as a light source in photographing the spectral image 212 using the spectral sensor 210, and the lighting unit 205 may include lighting devices such as an LED lamp and a krypton lamp.

In addition, a second collection station 290 may be additionally included in the spectroscopic analysis unit 200. To this end, a second image sensor (not shown) may be additionally provided in the spectroscopic analyzer 200 inside the second housing.

The second image sensor may be a photographing device using an image sensor such as a charge coupled device (CCD) or a CMOS image sensor (CIS). The second image acquired through the second image sensor is an above-ground image of the plant body 50, and the spectroscopic analysis unit 200 uses a pre-generated underground portion prediction model to determine the underground portion of the plant body 50. The shape of can be predicted.

The basement part refers to a part of the plant tissue, such as a root in the basement. That is, the spectroscopic analysis unit 200 may predict the shape of the underground part of the plant 50 based on the above-ground image of the plant 50 acquired through the second image sensor.

The underground part prediction model may be generated based on a result of learning based on a deep learning algorithm for the correlation between the above-ground image and the underground image for each variety of the plant 50 obtained in advance.

The deep learning algorithm may include a generative adversarial network (GAN), a variational autoencoder (VAE), and the like. However, not only the deep learning algorithms described above can be used, but various artificial intelligence technologies such as machine learning algorithms such as Bayesian network and support vector machine (SVM) as well as regression analysis methods can be used. have.

That is, the spectral analysis unit 200 may generate an underground image of the plant 50 from the above-described image using various methods described above, and may secondaryly separate the plant 50 from the generated underground image.

For example, the plant 50 to be separated may be selected based on the distribution of roots, the number of roots, and the length of the roots in the generated underground image. Accordingly, the secondly separated plant 50 may be separated into the second collection point.

For the plants 50 that have not been separated and collected at the first collection point and the second collection point through the above-described process, the grade information can be determined based on the internal and external properties of the plant 50 in the above (300). . The grade information may be classified based on the quantified results for the inner trait and the outer trait, respectively.

Specifically, in consideration of differences in development characteristics for each cultivar of the plant 50, information for determining the internal and external characteristics may be determined differently according to the cultivar. For example, if the proportion of the leaf area of the plant 50 is larger than the plant height according to the variety of the plant 50, the inner quality can be quantified only with the leaf area without considering the plant height and the number of nodes of the plant 50.

Alternatively, the weight may be determined differently according to the importance of each variety of the inner trait.For example, a weight of 0.7 may be assigned to the leaf area, and a weight of 0.15 may be assigned to each plant length and number of nodes to quantify the inner trait. Can be.

As described above, the external quality of the plant 50 may be quantified as different weights are given according to information such as a normalized vegetation index or a disease and pest index according to the variety as described above. That is, in the plant determining unit 300, the grade of the plant 50 may be classified based on the inner and outer traits of the plant 50 quantified by the above method.

The above description of the present invention is for illustrative purposes only, and those of ordinary skill in the art to which the present invention pertains will be able to understand that other specific forms can be easily modified without changing the technical spirit or essential features of the present invention. will be. Therefore, it should be understood that the embodiments described above are illustrative and non-limiting in all respects. For example, each component described as a single type may be implemented in a distributed manner, and similarly, components described as being distributed may also be implemented in a combined form.

The scope of the present invention is indicated by the claims to be described later rather than the detailed description, and all changes or modified forms derived from the meaning and scope of the claims and their equivalent concepts should be construed as being included in the scope of the present invention.

10: Smart plant quality determination and sorting device
50: plant
100: shape analysis unit
110: light source unit 110, 120: detection unit
130: shape analysis module 135: shape model
138: shape information
200: spectroscopic analysis unit 205: lighting unit
210: spectral sensor 212: spectral image
220: spectral analysis module 222: external quality
300: Plant identification unit

Claims (7)

In the smart plant quality determination and selection device,
A shape analysis unit for deriving an external quality of the plant through the scanning of the plant body;
A spectroscopic analysis unit for deriving the inner quality of the plant by using the spectroscopic image of the plant; And
Smart plant quality determination and selection device comprising a plant body determining unit for determining the grade information on the plant based on the external and internal properties of the plant.
The method of claim 1,
An input unit for supplying the plant to the shape analysis unit;
A discharge unit for classifying the plants whose grade information is determined by the plant body determining unit and discharging them to the packaging unit; And
Smart plant quality determination and selection device further comprising a transfer unit provided with a moving module for moving the plant supplied through the input unit to the discharge unit.
The method of claim 1,
The shape analysis unit,
A light source unit for irradiating light toward the plant;
A detection unit for identifying the light reflected from the plant; And
Smart plant quality determination and selection device further comprising a shape analysis module for deriving the shape information of the plant based on the point cloud obtained as a result of the identification.
The method of claim 1,
The spectroscopic analysis unit,
A spectroscopic sensor for acquiring a spectroscopic image of the plant; And
Smart plant quality determination and selection device further comprising a spectroscopic analysis module for calculating the normalized vegetation index of the plant based on the spectral reflectance of each wavelength band extracted from the spectral image.
In the smart plant quality determination and selection method for scanning plants using a transfer conveyor,
Recording a time at which the movement of the transfer conveyor for transferring the plants starts;
Acquiring dual lidar data in which an external trait for the plant is stored or a hyperspectral image in which the inner trait is stored; And
Comprising the step of calculating a 3D mesh from the dual lidar data, or analyzing the spectrum obtained from the hyperspectral image to analyze the seedling process for the plant.
The method of claim 5,
Smart plant quality determination and selection method comprising the step of producing a spectral library from the average of the spectrum obtained from the hyperspectral image.
The method of claim 6,
After preparing the spectral library,
Smart plant quality determination and selection method further comprising the step of matching the plant and the seedling process and transmitting to a database.

Images