Vegetation evaluation system using ultraviolet fluorescence and vegetation evaluation method using same

Abstract

In a vegetation evaluation system using ultraviolet fluorescence and a vegetation evaluation method using same, the vegetation evaluation system comprises an optical device and a determination unit. The optical device obtains an RGB image or a vegetation index image for a first region, and obtains the RGB image or an ultraviolet fluorescence image for a suspicious region selected from the first region. The determination unit determines a vegetation state on the basis of the RGB image or the ultraviolet fluorescence image for the suspicious region.

Description

Vegetation evaluation system using ultraviolet fluorescence and vegetation evaluation method using the same

[0001] The present invention relates to a vegetation evaluation system using ultraviolet fluorescence and a vegetation evaluation method using the same. More specifically, it relates to a vegetation evaluation system using ultraviolet fluorescence and a vegetation evaluation method using the same that uses a combination of vegetation images by wavelength and ultraviolet fluorescence to more precisely detect plant pests and vegetation and predict functional substances based on the stress state of plants.

[0002] Acquiring information about vegetation is essential for conducting vegetation assessments, and accordingly, various technologies for acquiring such information are being developed.

[0003] In particular, the most representative method for acquiring information about vegetation is to obtain images of the vegetation using a camera or similar device; a representative example is the technology for acquiring images through shooting to identify the damage status of crops, as described in Korean Patent Publication No. 10-2024-0052610.

[0004] In addition, as disclosed in Korean Registered Patent No. 10-2579082, a technology for acquiring multispectral images of a target and calculating spectrum-based information is also disclosed, and in Japanese Registered Patent No. 6938275, a technology for acquiring various vegetation information using measurement light having different wavelengths by utilizing an optical filter is also being developed.

[0005] However, the technologies developed to date for acquiring the aforementioned images or videos have limitations in that they obtain vegetation information based on fragmentary images or videos, making it difficult to make more precise and accurate judgments regarding vegetation or pests.

[0006] In particular, when measuring conventional chlorophyll fluorescence, there were limitations in directly determining the actual stress state of plants by simply performing fluorescence for chlorophyll.

[0007] <Prior Art Literature>

[0008] <Patent Literature>

[0009] (Patent Document 1) Republic of Korea Published Patent No. 10-2024-0052610

[0010] (Patent Document 2) Republic of Korea Registered Patent No. 10-2579082

[0011] (Patent Document 3) Japanese Registered Patent No. 6938275

[0012] Accordingly, the technical problem of the present invention is conceived from this point, and the objective of the present invention is to provide a vegetation evaluation system using ultraviolet fluorescence that can more precisely detect plant pests and vegetation and predict functional substances based on the stress state of plants by utilizing a combination of wavelength-specific vegetation images and ultraviolet fluorescence.

[0013] In addition, another objective of the present invention is to provide a vegetation evaluation method using the vegetation evaluation system.

[0014] A vegetation evaluation system according to one embodiment for realizing the purpose of the present invention described above includes an optical device and a judgment unit. The optical device acquires an RGB image or a vegetation index image for a first area and acquires the RGB image or an ultraviolet fluorescence image for a selected suspected area among the first areas. The judgment unit determines the vegetation status based on the RGB image or the ultraviolet fluorescence image for the suspected area.

[0015] In one embodiment, the optical device may include a first imaging unit for acquiring the RGB image for the first area and the suspected area, a second imaging unit for acquiring the vegetation index image for the first area, and a third imaging unit for acquiring the ultraviolet fluorescence image for the suspected area.

[0016] In one embodiment, the first imaging unit provides light of a first polarization to the first region and the suspected region, and can receive light of a second polarization perpendicular to the first polarization from the first region and the suspected region.

[0017] In one embodiment, the ultraviolet fluorescence image can be acquired at night.

[0018] In one embodiment, the judgment unit may include a first judgment unit that primarily determines a disease from the RGB image of the suspected area, and a second judgment unit that determines the growth stage of the plant and vegetation health from the ultraviolet fluorescence image of the suspected area.

[0019] In one embodiment, the first judgment unit may use the learning result of a first learning unit that learns the relationship between the RGB image and the disease, and the second judgment unit may use the learning result of a second learning unit that learns the ultraviolet fluorescence image and the growth stage and vegetation health of the plant.

[0020] In one embodiment, the second judgment unit can determine the growth stage and vegetation health of the plant based on the distribution of phenolic compounds from the ultraviolet fluorescence image.

[0021] In one embodiment, the ultraviolet fluorescence image can be obtained using ultraviolet light having a wavelength in the range of 320 nm to 400 nm.

[0022] In one embodiment, based on the result of determining the vegetation state, an analysis unit may further include an analysis unit that analyzes the precise vegetation state based on a multispectral image and the RGB image or the ultraviolet fluorescence image.

[0023] In one embodiment, the optical device may further include a spectroscopic unit for acquiring the multispectral image for an area where an abnormal lesion or an unhealthy vegetation state is found as a result of judging the vegetation state.

[0024] In one embodiment, the analysis unit may include a first analysis unit that diagnoses a disease from the multispectral image and the RGB image, and a second analysis unit that cross-diagnoses a disease from the ultraviolet fluorescence image.

[0025] In one embodiment, the first analysis unit may use the learning result of a third learning unit that learns the relationship between the multispectral image and the RGB image and the disease, and the second analysis unit may use the learning result of a fourth learning unit that learns the relationship between the ultraviolet fluorescence image and the disease.

[0026] In one embodiment, a monitoring unit for monitoring the stress of the plant based on the ultraviolet fluorescence image may be further included.

[0027] In one embodiment, the ultraviolet shape image may be a fluorescent image resulting from irradiation with a first ultraviolet light of a first wavelength range and a second ultraviolet light of a second wavelength range.

[0028] In one embodiment, the first wavelength range may be 320 nm to 400 nm, and the second wavelength range may be 280 nm to 320 nm.

[0029] In one embodiment, the first ultraviolet light is absorbed by the flavonoid components of the cuticle layer of the plant to induce green fluorescence, and the second ultraviolet light is absorbed by the phenolic acid components of the cuticle layer to induce blue fluorescence.

[0030] In one embodiment, the flavonoid component may include quercetin or rutin, and the phenolic acid component may include chlorogenic acid, p-coumaric acid, or caffeic acid.

[0031] In a vegetation evaluation method according to an embodiment for realizing another objective of the present invention described above, an RGB image or a vegetation index image is obtained for a first region. The RGB image or an ultraviolet fluorescence image is obtained for a selected suspected region among the first regions. The vegetation state is determined based on the RGB image or the ultraviolet fluorescence image for the suspected region.

[0032] In one embodiment, the step of determining the vegetation status may include: a step of determining the disease from the RGB image of the suspected area using a learning result regarding the relationship between the previously learned RGB image and the disease; or a step of determining the growth stage of the plant and the vegetation health from the UV fluorescence image of the suspected area using a learning result regarding the previously learned UV fluorescence image and the growth stage of the plant and the vegetation health.

[0033] In one embodiment, the method may further include a step of analyzing the precise vegetation state based on the judgment result of the vegetation state judgment step. In this case, the step of analyzing the precise vegetation state may include a step of diagnosing a disease from the multispectral image of the analysis area using a previously learned result regarding the relationship between the multispectral image and the RGB image and the disease, or a step of cross-diagnosing a disease from the ultraviolet fluorescence image of the analysis area using a previously learned result regarding the relationship between the ultraviolet fluorescence image and the disease.

[0034] According to embodiments of the present invention, through a plurality of imaging units mounted on a single optical device, an RGB image or an ultraviolet fluorescence image can be obtained for a first area subject to vegetation evaluation and for a suspected area among the images of the first area where pests are suspected. Therefore, vegetation judgment can be performed not only through overall scanning of the first area but also through the image of the suspected area. Accordingly, vegetation evaluation can be performed by selecting areas suspected of pests even over a relatively wide imaging range, thereby enabling evaluation using relatively quick and small amounts of data.

[0035] At this time, the RGB image of the suspected area is used to primarily determine the disease of the plant, and in particular, the ultraviolet fluorescence image obtained for the suspected area can determine the growth stage of the plant's leaves and the health of the vegetation, thereby providing a basis for subsequent precise analysis.

[0036] In the case of the above ultraviolet fluorescence image, the growth stage and vegetation health of the plant are determined based on the distribution of phenolic compounds contained in the leaves of the plant through ultraviolet rays of a specific wavelength, so it is possible to determine the vegetation status of the plant more accurately.

[0037] In addition, based on the above primary judgment, a more precise determination of plant disease symptoms is made through an analysis unit, and since the analysis unit can perform cross-verification of the disease symptoms using the first and second analysis sections, the analysis of the disease symptoms can be performed more precisely and accurately.

[0038] In particular, since the first analysis unit diagnoses the disease based on multispectral imaging and the second analysis unit diagnoses the disease based on ultraviolet fluorescence imaging, it is possible to perform a more precise and accurate analysis of the disease by utilizing both the result of diagnosing the disease by synthesizing images in the multi-wavelength range and the result of diagnosing the disease in the fluorescence imaging range.

[0039] In addition, since both the judgment in the judgment unit and the analysis in the analysis unit utilize the results of previously learned training, the accuracy of the analysis can be further improved.

[0040] Furthermore, since the above analysis unit also enables the identification of pests, it is possible to analyze various vegetation conditions of plants.

[0041] In addition, plant stress can be assessed using first ultraviolet light in a first wavelength range and second ultraviolet light in a second wavelength range, and in particular, more accurate assessment and monitoring of plant stress can be performed based on fluorescence results regarding flavonoid and phenolic acid components of plants. Through this, functional substances of plants can be predicted, and control based on plant stress can be carried out.

[0042] FIG. 1 is a block diagram illustrating a vegetation evaluation system according to one embodiment of the present invention.

[0043] Figure 2 is a flowchart illustrating a vegetation evaluation method using the vegetation evaluation system of Figure 1.

[0044] FIG. 3a is an example of a grayscale image of an RGB image obtained for a first region, and FIG. 3b is an example of a grayscale image of a vegetation index image obtained for a first region.

[0045] Figure 4a is an image illustrating the actual movement of a mobile body equipped with the vegetation evaluation system of Figure 1, and Figure 4b is an image illustrating the result of mapping the suspected pest area of ​​Figure 2.

[0046] Figure 5a is a graph illustrating the wavelength-dependent absorption fluorescence state of chlorophyll in plants, and Figure 5b is a schematic diagram explaining the fluorescence-induced state by phenolic compounds.

[0047] Figures 6a and 6b are images of the results of analyzing plant disease symptoms using ultraviolet fluorescence imaging in the second analysis unit of Figure 1.

[0048] The present invention is susceptible to various modifications and may take various forms, and embodiments are to be described in detail in the text. However, this is not intended to limit the invention to the specific disclosed forms, and it should be understood that the invention includes all modifications, equivalents, and substitutions that fall within the spirit and scope of the invention. Similar reference numerals have been used for similar components in the description of each figure. Terms such as "first," "second," etc., may be used to describe various components, but said components should not be limited by said terms.

[0049] The above terms are used solely for the purpose of distinguishing one component from another. The terms used in this application are used merely to describe specific embodiments and are not intended to limit the invention. The singular expression includes the plural expression unless the context clearly indicates otherwise. In this application, terms such as "comprising" or "consisting of" are intended to specify the existence of the features, numbers, steps, actions, components, parts, or combinations thereof described in the specification, and should be understood as not precluding the existence or addition of one or more other features, numbers, steps, actions, components, parts, or combinations thereof.

[0050] Hereinafter, preferred embodiments of the present invention will be described in more detail with reference to the attached drawings.

[0051] FIG. 1 is a block diagram illustrating a vegetation evaluation system according to one embodiment of the present invention.

[0052] First, referring to FIG. 1, the vegetation evaluation system (10) according to the present embodiment includes an optical device (100), a region selection unit (200), a mapping unit (300), a judgment unit (400), an analysis unit (500), a learning unit (600), and a monitoring unit (700).

[0053] Furthermore, the above vegetation evaluation system (10) may additionally include a pest control unit (800) to perform direct pest control on crops infected with pests based on the judgment results and monitoring results, in addition to simply judging the vegetation condition.

[0054] In particular, since the above vegetation evaluation system (10) is mounted on a mobile body (20, see FIG. 4a) and can move together with the mobile body (20) in the space where crops are cultivated, the above pest control unit (800) is also mounted on the mobile body (20) at the same time, so that pest control can be performed directly on crops infected with pests.

[0055] The optical device (100) includes first to third shooting units (110, 120, 130) and a spectroscopic unit (140), the judgment unit (400) includes first and second judgment units (410, 420), the analysis unit (500) includes first and second analysis units (510, 520), and the learning unit (600) includes first to fourth learning units (610, 620, 630, 640).

[0056] In the following, the vegetation evaluation system of FIG. 1 and the vegetation evaluation method using it will be explained simultaneously for the convenience of explanation.

[0057] Figure 2 is a flowchart illustrating a vegetation evaluation method using the vegetation evaluation system of Figure 1.

[0058] Referring to FIGS. 1 and 2, in the vegetation evaluation method using the vegetation evaluation system (10), first, an RGB image or a vegetation index image is obtained for a first area (step S100).

[0059] FIG. 3a is an example of a grayscale image of an RGB image obtained for a first region, and FIG. 3b is an example of a grayscale image of a vegetation index image obtained for a first region.

[0060] That is, as shown in FIG. 3a, the first capturing unit (110) acquires an RGB image of the first area. To do this, the first capturing unit (110) may be an RGB camera.

[0061] At this time, the first area is defined as the maximum area that can be photographed through the first shooting unit (110) with respect to the area where vegetation (or crops) are distributed. Accordingly, if the area where vegetation is distributed is smaller than the maximum area that can be photographed through the first shooting unit (110), the first area may be defined as the entire area where vegetation is distributed. Conversely, if the area where vegetation is distributed is larger than the maximum area that can be photographed through the first shooting unit (110), the area where vegetation is distributed may be divided into multiple areas, and each area may be defined as the first area. Thus, the first shooting unit (110) performs photography on a single divided area defined as the first area.

[0062] At this time, the first area needs to be defined as the widest possible area, and accordingly, RGB images can be acquired relatively quickly through the first shooting unit (110) over the entire area where the vegetation is distributed.

[0063] Additionally, when the RGB image is acquired through the first imaging unit (110), the polarization of the light emitted from the first imaging unit (110) and the polarization of the light received through the first imaging unit (110) may be perpendicular to each other. That is, the first imaging unit (110) emits light of first polarization to provide light to the first region and the suspected region, and the light reflected and received from the first region and the suspected region may receive light of second polarization that is polarized perpendicular to the first polarization.

[0064] By doing so, the accuracy and reliability of the judgment or monitoring results regarding the vegetation can be further improved by minimizing the shimmering image on the surface of the vegetation (or crop) in the first area and the suspected area and receiving light.

[0065] In contrast, referring to FIG. 3b, the second imaging unit (120) acquires a vegetation index image for the first area. To this end, the second imaging unit (120) may be a multi-wavelength camera for measuring vegetation. At this time, a vegetation index image refers to an image from which a vegetation index can be derived for the vegetation, and typically includes a normalized difference vegetation index (NDVI), a reflection vegetation index image, or a fluorescence vegetation index image.

[0066] In addition, the area captured by the second shooting unit (120) is also the first area, and is identical to the area where the first shooting unit (110) performs shooting.

[0067] As described above, an RGB image of a first area is obtained through the first shooting unit (110), and a vegetation index image of the first area is obtained through the second shooting unit (120). Then, referring to FIGS. 1 and 2, the area selection unit (200) derives a suspected pest area from the obtained image and maps it through the mapping unit (300) (step S200).

[0068] That is, as shown in FIG. 3a, suspected areas (A, A') suspected of being infected with pests are derived from the RGB image of the first area, and as shown in FIG. 3b, suspected areas (B, B') suspected of being infected with pests are also derived from the vegetation index image of the first area.

[0069] In addition, the above-mentioned suspected areas derived in this way are mapped based on the location information of the moving body (20).

[0070] Figure 4a is an image illustrating the actual movement of a mobile body equipped with the vegetation evaluation system of Figure 1, and Figure 4b is an image illustrating the result of mapping the suspected pest area of ​​Figure 2.

[0071] That is, referring to FIG. 4a, the moving body (20) moves over the entire area (30) that is the subject of vegetation judgment while equipped with the vegetation evaluation system (10), and simultaneously with this movement, the first to third shooting units (110, 120, 130) capture the necessary images.

[0072] Thus, from the above-mentioned images, the area selection unit (200) selects the suspected areas, and for the selection of these suspected areas, information about each location where the moving body (20) moves is required.

[0073] Accordingly, based on the location information of the above-mentioned moving body (20), for example, GPS information, the above-mentioned captured images at each location can be stored in a database, and through this, information on the captured images at each location can be obtained. Thus, when the above-mentioned suspected area is selected, information on which location the captured result of the suspected area is from can also be obtained, as exemplified in FIG. 4b.

[0074] That is, among the entire area (30) subject to the vegetation judgment, a suspected area where pest infection is strongly suspected is selected, and the suspected area can be selected by classifying it into a suspected area selected based on RGB images (Class 1, A, A') and a suspected area selected based on vegetation index images (Class 2, B, B'). Furthermore, a normal area (Class 3) is also selected.

[0075] Accordingly, each of the regions classified as above (Class 1, Class 2, Class 3) can be mapped onto the entire region (30) as shown in FIG. 4b.

[0076] In particular, in the case of the present embodiment, RGB images, vegetation index images, and ultraviolet fluorescence images are obtained for the suspected area through the first to third imaging units (110, 120, 130), so such mapping is absolutely necessary. Thus, when imaging vegetation at the same location using different imaging units, information about vegetation at an accurate location can be obtained while minimizing errors caused by errors in location information.

[0077] Meanwhile, the above-mentioned mobile body (20) may be an unmanned mobile body capable of autonomous driving, but may also be driven by external control or driven with a person directly on board. At this time, the above-mentioned mobile body (20) is equipped with a device capable of acquiring a GPS signal at the location where it is moved, and thus provides location information at each location where the above-mentioned mobile body (20) is moved to the mapping unit (300) described above.

[0078] Then, referring to FIGS. 1 and FIGS. 2, the RGB image or the ultraviolet fluorescence image is obtained for selected suspected regions (A, A', B, B') among the first regions (step S300).

[0079] That is, the first imaging unit (110) acquires the RGB image with respect to the suspected area (step S300), and the third imaging unit (130) acquires the so-called ultraviolet fluorescence image with respect to the suspected area (step S320). At this time, the acquisition of the RGB image and the acquisition of the ultraviolet fluorescence image are performed independently, and may be performed sequentially or simultaneously.

[0080] Meanwhile, as will be described later, when acquiring an ultraviolet fluorescence image of the suspected area, in order to monitor plant stress, the third imaging unit (130) can acquire the ultraviolet fluorescence image by irradiating a first ultraviolet light of a first wavelength range and a second ultraviolet light of a second wavelength range.

[0081] At this time, the first ultraviolet light may be, for example, UV-A, and the first wavelength range may be 320 nm to 400 nm. Alternatively, the second ultraviolet light may be, for example, UV-B, and the second wavelength range may be 280 nm to 320 nm.

[0082] Thus, the ultraviolet fluorescence image of the suspected area obtained through the first ultraviolet and the second ultraviolet is separately provided to the monitoring unit (700).

[0083] Meanwhile, as previously explained, the first shooting unit (110) may be an RGB camera and may include a zoom lens. Thus, the RGB image can be obtained as an enlarged image with respect to the suspected area.

[0084] That is, through the first shooting unit (110), an enlarged image is acquired only for the suspected area, that is, only for a relatively narrow area, and based on this, more accurate and precise image information is provided for subsequent determination of vegetation conditions.

[0085] Additionally, the third imaging unit (130) acquires the ultraviolet fluorescence image with respect to the suspected area, and for this purpose, the third imaging unit (130) may be an ultraviolet camera that provides ultraviolet light of a specific wavelength.

[0086] At this time, the wavelength of the ultraviolet light provided above may be configured to include UV-A and UV-B as previously explained, but is not limited thereto. Meanwhile, if the wavelength of the ultraviolet light is UV-A, it may be, for example, 365 nm.

[0087] In addition, since the third shooting unit (130) also needs to capture an enlarged image of a specific suspected area, it may include a zoom lens like the first shooting unit (110).

[0088] As described above, when an RGB image or an ultraviolet fluorescence image is acquired for the suspected area, the acquired result is provided to the judgment unit (400). At this time, the ultraviolet fluorescence image is also provided to the monitoring unit (700). Below, the judgment of the vegetation status by the judgment unit (400) is explained first, and the monitoring of the stress of the plants by the monitoring unit (7000) will be described later.

[0089] First, referring to FIGS. 1 and 2, the judgment unit (400) performs a judgment on the vegetation status for the suspected areas (A, A', B, B') based on the RGB image or the ultraviolet fluorescence image (step S400).

[0090] As a judgment in the above judgment unit (400), the first judgment unit (410) primarily performs a judgment on the disease of the plant leaves in the suspected area as a vegetation state based on the RGB image.

[0091] That is, the first judgment unit (410) performs a primary basic judgment regarding the disease of the plant's leaves, that is, whether there is an infection by various pathogens. Thus, based on the result of the primary judgment regarding the disease in the first judgment unit (410), subsequent additional analysis is performed.

[0092] In addition, as a judgment in the above judgment unit (400), the first judgment unit (410) determines the growth stage of the plant leaves and the health of the vegetation as a vegetation state based on the ultraviolet fluorescence image.

[0093] In this regard, we first explain that it is possible to determine the growth stage of plant leaves and vegetation health through ultraviolet fluorescence imaging. As a conventional method for measuring chlorophyll fluorescence, measuring chlorophyll fluorescence using ultraviolet light with a wavelength of 400 nm to 470 nm had a problem in that it was difficult to determine whether a polyphenol layer was present in the plant leaves because it penetrated the polyphenol layer. In other words, conventional chlorophyll fluorescence measurement had the limitation of only being able to determine the state of chlorophyll.

[0094] However, phenolic compounds such as the polyphenol layer present in the above-mentioned plant leaves are very important substances for determining the degree of stress in plants, and thus, by determining the amount of the above-mentioned phenolic compounds, it is possible to determine not only the degree of stress in the plant leaves but also the growth stage of the plant leaves and vegetation health accordingly.

[0095] That is, in the case of the ultraviolet fluorescence image obtained through the third imaging unit (130) according to the present embodiment, as previously explained, by providing a wavelength of 320 nm to 400 nm, more specifically 365 nm, to obtain the ultraviolet fluorescence image, an image of the degree of phenolic compounds present in the plant leaf can be obtained.

[0096] Accordingly, when ultraviolet fluorescence images are acquired at the above wavelength, relatively young leaves or leaves with a low stress index are acquired as red fluorescence, whereas relatively mature leaves or leaves with a high stress index are acquired as blue or green fluorescence.

[0097] This is because, in the case of relatively young leaves or leaves with a low stress index, when ultraviolet rays of the above wavelength are irradiated, there is less phenolic compound formed or accumulated on the surface, so the ultraviolet rays easily pass through the surface and react with chlorophyll, so they are displayed as fluorescence such as red.

[0098] In contrast, in the case of relatively mature leaves or leaves with a high stress index, when irradiated with ultraviolet rays of the above wavelength, the surface has a large amount of phenolic compounds formed or accumulated, so the ultraviolet rays do not easily pass through the surface and react with the phenolic compounds, resulting in fluorescence such as blue or green.

[0099] Figure 5a is a graph illustrating the wavelength-dependent absorption fluorescence state of chlorophyll in plants, and Figure 5b is a schematic diagram explaining the fluorescence-induced state by phenolic compounds.

[0100] Referring to FIG. 5a, when fluorescence is obtained using ultraviolet light with a wavelength of 400 nm to 470 nm in the conventional method, it is difficult to confirm the presence of phenolic compounds formed or accumulated on the surface of leaves because the wavelength is only a wavelength that reacts with chlorophyll. However, when fluorescence is obtained using ultraviolet light with a wavelength of 365 nm, specifically in the range of 320 nm to 400 nm, fluorescence of phenolic compounds located on the surface of leaves is possible, so as previously explained, the growth stage of the plant leaves and the health of the vegetation can be determined based on the amount of said phenolic compounds.

[0101] More specifically, the fluorescence state of a plant leaf at each growth stage is explained as follows. The second judgment unit (420) can determine the growth stage of the plant leaf based on the results obtained using ultraviolet fluorescence from the third imaging unit (130).

[0102] That is, in the case of young leaves in the growth stage that produce relatively little phenolic compounds including polyphenols, they are marked in red as a result of ultraviolet fluorescence, and thus the leaves in the area marked in red can be identified as leaves in the growth stage.

[0103] In contrast, leaves in the mature stage are marked in blue or green as a result of ultraviolet fluorescence because phenolic compounds containing polyphenols are formed or accumulated, and thus the leaves in the area marked in blue or green can be identified as leaves in the mature stage.

[0104] Furthermore, in the leaves of the declining or senescent stage, phenolic compounds containing polyphenols gradually disappear and are displayed in red as a result of ultraviolet fluorescence, and thus the leaves in the area displayed in red can be identified as leaves of the declining or senescent stage.

[0105] However, in the case of the above-mentioned declining leaves, if the shape of the leaves and other factors are additionally considered, it can be determined that they correspond to the declining stage in contrast to the growth stage, so even if they are displayed in the same red color, the growth stage of the leaves can be distinguished and determined.

[0106] Meanwhile, in the first judgment unit (410), when primarily determining whether there is a disease in the suspected area, the result of the learning of the first learning unit (610) can be utilized. That is, the first learning unit (610) performs learning based on RGB images and previously stored data regarding the presence or absence of a disease, and the result of such learning can be utilized in the first judgment unit (410) to determine the disease from the RGB images provided through the first shooting unit (110).

[0107] Likewise, in determining the leaf growth stage and vegetation health status with respect to the suspected area in the second judgment unit (420), the results of the learning of the second learning unit (620) can be utilized. That is, the second learning unit (620) performs learning based on previously stored data regarding the ultraviolet fluorescence image and the leaf growth stage, and the ultraviolet fluorescence image and vegetation health, and the results of such learning can be utilized in determining the growth stage and vegetation health status in the second judgment unit (420) from the ultraviolet shape image provided through the third shooting unit (130).

[0108] Meanwhile, referring again to FIGS. 1 and FIGS. 2, based on the judgment result of the judgment unit (400) as described above, the analysis unit (500) performs an analysis on the precise vegetation condition (step S500).

[0109] That is, when the first judgment unit (410) determines the disease in the suspected area as a result of the first judgment, and determines that the disease exists or an abnormal pattern is found, the first analysis unit (510) classifies the disease or pest using the multispectral imaging of the spectroscopic unit (140) and the previously acquired RGB image (step S510).

[0110] At this time, the detection of the above abnormal pattern can be made by utilizing the result of the classification by growth stage in the second judgment unit (420) to detect whether there is an abnormal pattern for each growth stage. Accordingly, the first analysis unit (510) can simultaneously utilize the judgment result from the first judgment unit (410) as well as the judgment result from the second judgment unit (420).

[0111] Consequently, the first analysis unit (510) can classify pests and diseases by using more specific multispectral images and RGB images for the area or leaf where the disease is found, based on the judgment result regarding the disease from the first judgment unit (410). Additionally, the first analysis unit (510) can classify pests and diseases by combining the information regarding each growth stage of the plant leaf from the second judgment unit (420) with the judgment result regarding the disease from the first judgment unit (410), and if an abnormal pattern is found for each growth stage, by using more specific multispectral images for the area or leaf where the abnormal pattern is found.

[0112] For example, the first analysis unit (510) classifies and determines pests and diseases through multispectral imaging via the spectroscopic unit (140) and the RGB image for areas or leaves where such disease symptoms are found, or areas or leaves where abnormal patterns by growth stage are found.

[0113] In this case, since pests and diseases have different reflectances at specific wavelengths for each type, the reflected images for each wavelength are combined to classify the pests and diseases and to identify and determine what they are. Through this, the types of pests and diseases infected in each area or leaf can be analyzed collectively.

[0114] Meanwhile, in classifying and determining pests using the multispectral image and RGB image of the first analysis unit (510) mentioned above, the learning results of the third learning unit (630) mentioned above can be used.

[0115] That is, the third learning unit (630) performs learning based on data stored regarding multispectral images, RGB images, and types of pests, and the results of such learning can be used to analyze the types of pests in the first analysis unit (510) from the multispectral images provided through the spectroscopic unit (140) and the RGB images acquired earlier.

[0116] In addition, as an analysis of the precise vegetation status of the analysis unit (500), the second analysis unit (520) can classify and determine pests using a fluorescence imaging technique based on the ultraviolet fluorescence judgment result from the second judgment unit (420).

[0117] Generally, in the case of pests, the amount or pattern of phenolic compounds containing polyphenols within the plant can be varied depending on the type. Accordingly, based on the ultraviolet fluorescence image utilized in the second judgment unit (420), the second analysis unit (520) can analyze the type of pest. Furthermore, the analysis of the type of pest in the second analysis unit (520) can be used for cross-verification with the result of analyzing the type of pest in the first analysis unit (510).

[0118] Figures 6a and 6b are images of the results of analyzing plant disease symptoms using ultraviolet fluorescence imaging in the second analysis unit of Figure 1.

[0119] Referring to FIGS. 6a and FIGS. 6b, it can be seen that the second analysis unit (520) analyzes the types of pests based on ultraviolet fluorescence images and shows different fluorescence results for each pest.

[0120] In particular, the learning results of the fourth learning unit (640) can also be utilized in the second analysis unit (520). That is, the fourth learning unit (640) performs learning based on previously stored data regarding ultraviolet fluorescence images and types of pests, and the results of such learning can be utilized in the second analysis unit (520) to analyze the types of pests.

[0121] Accordingly, the result of analyzing the type of pest in the second analysis unit (520) can also have high reliability, and in particular, since cross-verification can be performed with the analysis result in the first analysis unit (510), the result of analyzing the type of pest can be more accurate.

[0122] As described above, based on the results of analyzing the types of pests through the analysis unit (500), the control unit (800) performs control on the leaves or areas infected with pests according to the type of pest (step S700).

[0123] Meanwhile, through the second analysis unit (520) above, in addition to analyzing disease symptoms on leaves or areas, it is also possible to analyze pests. That is, in the case of pests, since the ultraviolet fluorescence images may differ depending on the type of pest, the type of pest can be determined based on this.

[0124] Of course, the type of pest can also be determined through the first analysis unit (510) above, and this can be used for cross-verification with the analysis result of the second analysis unit (520).

[0125] Furthermore, referring again to FIGS. 1 and FIGS. 2, the monitoring unit (700) monitors plant stress based on the fluorescence images of the first ultraviolet and the second ultraviolet light mentioned above (step S600).

[0126] At this time, although Figure 2 illustrates the monitoring of plant stress in the suspected area based on the ultraviolet fluorescence image obtained in the suspected area, it is evident that monitoring of plant stress in all overall areas, in addition to the suspected area, can be performed.

[0127] Generally, plants include a cuticle layer on the outermost layer, and a wax layer is formed on the surface of the cuticle layer. The wax layer contains a mixture of fatty acids, alcohols, aldehydes, ketones, etc., and is coated on the surface of the cuticle layer. As a layer that prevents moisture loss, it serves to inhibit the attachment of pests or diseases or inhibit the germination of spores.

[0128] The above cuticle layer forms an external surface structure of the plant that connects the cell walls located below, and serves as a barrier to the penetration of pests and diseases, thereby limiting their spread. At this time, the cuticle layer has a structure in which flavonoids and phenolic acids are distributed on cutin.

[0129] The above cutin is a polymer of C16 and C18 fatty acids that is deposited on the outer wall of epidermal cells to prevent water loss and serves as a mechanical support and physical barrier against pathogens. The above phenolic acid includes, for example, chlorogenic acid, p-coumaric acid, or caffeic acid, and plays a preemptive role in antioxidant activity and defense against pathogens. In particular, the above phenolic acid fluoresces in a second wavelength range of 280 nm to 320 nm, which is mainly in the UV-B region, thereby inducing blue fluorescence.

[0130] That is, when the aforementioned second ultraviolet light is irradiated, the phenolic acid in the cuticle layer reacts fluorescently with the second ultraviolet light and induces blue fluorescence.

[0131] The above flavonoids include, for example, quercetin or rutin, and perform continuous UV protection and antioxidant functions within plant tissues. In particular, the above flavonoids fluoresce in a first wavelength range of 320 nm to 400 nm, which is mainly in the UV-A region, and induce green fluorescence.

[0132] That is, when the first ultraviolet light described above is irradiated, the flavonoid in the cuticle layer reacts fluorescently to the first ultraviolet light and induces green fluorescence.

[0133] In addition, the phenolic acid is composed of a relatively small size compared to the flavonoid, and thus reacts more quickly to external stress than the flavonoid and has the characteristic of moving rapidly. Furthermore, the phenolic acid is produced early in the development of plant leaves, whereas the flavonoid is a substance that increases as the plant leaves mature.

[0134] Ultimately, through the fluorescence reaction to the above-mentioned phenolic acid, the occurrence of plant stress (i.e., biological stress) due to pests and diseases can be determined early. Furthermore, among the stresses of the plant, if the cuticle layer is damaged due to abiotic stress, the flavonoid component contained in the cuticle layer decreases. Therefore, when the first ultraviolet light is irradiated while the flavonoid component is reduced as described above, the first ultraviolet light can be absorbed by chlorophyll to induce red fluorescence. Through this, it is also possible to determine the state of occurrence of abiotic stress among the stresses of the plant.

[0135] Here, biological stress of a plant refers to a disease caused by factors such as pests and diseases, and abiotic stress of a plant refers to a physiological disorder caused by environmental factors. Accordingly, as described above, the monitoring unit (700) can perform more accurate judgment and monitoring of plant stress based on the results of the fluorescence reaction to the first ultraviolet light and the second ultraviolet light, particularly based on the fluorescence results of the flavonoid components and phenolic acid components of the plant. Through this, it may be possible to predict so-called functional substances of the plant.

[0136] Furthermore, based on the monitoring results of the stress of the plants monitored by the monitoring unit (700), the control unit (800) can perform appropriate control (step S700).

[0137] According to the embodiments of the present invention as described above, through a plurality of imaging units mounted on a single optical device, an RGB image or an ultraviolet fluorescence image can be obtained for a first area subject to vegetation evaluation and for a suspected area among the images of the first area where pests are suspected. Therefore, vegetation judgment can be performed not only through overall scanning of the first area but also through the image of the suspected area. Accordingly, vegetation evaluation can be performed by selecting areas suspected of pests even over a relatively wide imaging range, thereby enabling evaluation using relatively quick and small amounts of data.

[0138] At this time, the RGB image of the suspected area is used to primarily determine the disease of the plant, and in particular, the ultraviolet fluorescence image obtained for the suspected area can determine the growth stage of the plant's leaves and the health of the vegetation, thereby providing a basis for subsequent precise analysis.

[0139] In the case of the above ultraviolet fluorescence image, the growth stage and vegetation health of the plant are determined based on the distribution of phenolic compounds, including polyphenols, contained in the leaves of the plant through ultraviolet rays of a specific wavelength, so it is possible to determine the vegetation status of the plant more accurately.

[0140] In addition, based on the above primary judgment, a more precise determination of plant disease symptoms is made through an analysis unit, and since the analysis unit can perform cross-verification of the disease symptoms using the first and second analysis sections, the analysis of the disease symptoms can be performed more precisely and accurately.

[0141] In particular, since the first analysis unit diagnoses the disease based on multispectral imaging and the second analysis unit diagnoses the disease based on ultraviolet fluorescence imaging, it is possible to perform a more precise and accurate analysis of the disease by utilizing both the result of diagnosing the disease by synthesizing images in the multi-wavelength range and the result of diagnosing the disease in the fluorescence imaging range.

[0142] In addition, since both the judgment in the judgment unit and the analysis in the analysis unit utilize the results of previously learned training, the accuracy of the analysis can be further improved.

[0143] Furthermore, since the above analysis unit also enables the identification of pests, it is possible to analyze various vegetation conditions of plants.

[0144] In addition, plant stress can be assessed using first ultraviolet light in a first wavelength range and second ultraviolet light in a second wavelength range, and in particular, more accurate assessment and monitoring of plant stress can be performed based on fluorescence results regarding flavonoid and phenolic acid components of plants. Through this, functional substances of plants can be predicted, and control based on plant stress can be carried out.

[0145] Although the present invention has been described above with reference to preferred embodiments, those skilled in the art will understand that various modifications and changes can be made to the invention without departing from the spirit and scope of the invention as set forth in the following claims.

[0146] <Explanation of Symbols>

[0147] 10 : Vegetation evaluation system 20 : Mobile body

[0148] 100 : Optical device 200 : Area selection unit

[0149] 300: Mapping Unit 400: Judgment Unit

[0150] 410: 1st Judgment Unit 420: 2nd Judgment Unit

[0151] 500: Analysis Unit 510: 1st Analysis Section

[0152] 520 : 2nd Analysis Unit 600 : Learning Unit

[0153] 610: 1st Learning Department 620: 2nd Learning Department

[0154] 630 : 3rd Learning Department 640 : 4th Learning Department

[0155] 700: Monitoring Department 800: Pest Control Department

Claims

1. An optical device for acquiring an RGB image or a vegetation index image for a first region, and acquiring the RGB image or an ultraviolet fluorescence image for a selected suspected region among the first regions; and A vegetation evaluation system comprising a judgment unit that determines the vegetation status based on the RGB image or the ultraviolet fluorescence image of the suspected area.

2. In paragraph 1, the optical device is, A first capturing unit for acquiring the RGB image with respect to the first area and the suspected area; A second shooting unit for acquiring the vegetation index image for the first area; and A vegetation evaluation system characterized by including a third imaging unit that acquires the ultraviolet fluorescence image for the suspected area.

3. In paragraph 2, the first imaging unit is, Providing light of first polarization to the first region and the suspected region, and A vegetation evaluation system characterized by receiving light of second polarization perpendicular to the first polarization from the first region and the suspected region.

4. In paragraph 2, the ultraviolet fluorescence image is, Vegetation evaluation system characterized by being acquired at night.

5. In paragraph 1, the judgment unit is, A first judgment unit that primarily determines the disease from the RGB image of the suspected area; and A vegetation evaluation system characterized by including a second judgment unit for determining the growth stage of plants and vegetation health from the ultraviolet fluorescence image of the suspected area.

6. In Paragraph 5, The first judgment unit above uses the learning result of the first learning unit that learns the relationship between the RGB image and the disease, and A vegetation evaluation system characterized in that the second judgment unit above uses the learning results of the second learning unit which learns the ultraviolet fluorescence image and the growth stage and vegetation health of the plant above.

7. In paragraph 6, the above-mentioned second judgment unit, A vegetation evaluation system characterized by determining the growth stage and vegetation health of the plant based on the distribution of phenolic compounds from the above ultraviolet fluorescence image.

8. In paragraph 7, the ultraviolet fluorescence image is, A vegetation evaluation system characterized by being obtained using ultraviolet rays having a wavelength in the range of 320 nm to 400 nm.

9. In Paragraph 1, A vegetation evaluation system further comprising an analysis unit that analyzes a precise vegetation state based on a multispectral image and the RGB image or the ultraviolet fluorescence image, based on the judgment result regarding the vegetation state.

10. In paragraph 9, the optical device is, A vegetation evaluation system further comprising a spectroscopic unit for acquiring the multispectral image for an area in which an abnormal lesion or unhealthy vegetation condition is found as a result of judging the vegetation condition above.

11. In paragraph 9, the above analysis unit is, A first analysis unit for diagnosing a disease from the above multispectral image and the above RGB image; and A vegetation evaluation system characterized by including a second analysis unit that cross-diagnoses disease symptoms from the above ultraviolet fluorescence image.

12. In Paragraph 11, The first analysis unit above uses the learning result of the third learning unit that learns the relationship between the multispectral image and the RGB image and the disease, and A vegetation evaluation system characterized in that the second analysis unit above utilizes the learning results of the fourth learning unit which learns the relationship between the ultraviolet fluorescence image and the disease.

13. In Paragraph 1, A vegetation evaluation system further comprising a monitoring unit that monitors the stress of the plant based on the above ultraviolet fluorescence image.

14. In Paragraph 13, A vegetation evaluation system characterized in that the above-mentioned ultraviolet shape image is a fluorescent image resulting from irradiation with a first ultraviolet light of a first wavelength range and a second ultraviolet light of a second wavelength range.

15. In Paragraph 14, A vegetation evaluation system characterized in that the first wavelength range is 320 nm to 400 nm and the second wavelength range is 280 nm to 320 nm.

16. In Paragraph 15, The first ultraviolet light is absorbed by the flavonoid components of the cuticle layer of the plant to induce green fluorescence, and A vegetation evaluation system characterized by the fact that the above-mentioned second ultraviolet light is absorbed by the phenolic acid components of the cuticle layer to induce blue fluorescence.

17. In Paragraph 16, The above flavonoid component includes quercetin or rutin, and A vegetation evaluation system characterized in that the above-mentioned phenolic acid component includes chlorogenic acid, p-coumaric acid, or coffeec acid.

18. A step of acquiring an RGB image or a vegetation index image for the first region; A step of acquiring the RGB image or ultraviolet fluorescence image for a selected suspected area among the first areas; and A vegetation evaluation method comprising the step of determining the vegetation status based on the RGB image or the ultraviolet fluorescence image of the suspected area.

19. In paragraph 18, the step of determining the vegetation status above is, A step of primarily determining a disease from the RGB image of the suspected region using the learning result regarding the relationship between the previously learned RGB image and the disease; or A vegetation evaluation method characterized by including the step of determining the growth stage of a plant and the vegetation health from the ultraviolet fluorescence image of the suspected area using the previously learned ultraviolet fluorescence image and the learning results regarding the growth stage of a plant and the vegetation health.

20. In Paragraph 19, Based on the judgment result of the above vegetation status judgment step, it further includes a step of analyzing the precise vegetation status, and The step of analyzing the above precise vegetation status is, A step of diagnosing a disease from the multispectral image of the analysis area using the learning results regarding the relationship between the previously learned multispectral image and RGB image and the disease; or A vegetation evaluation method characterized by including a step of cross-diagnosing disease from the ultraviolet fluorescence image for the analysis area using the results of learning the relationship between the previously learned ultraviolet fluorescence image and the disease.

Images