Method and system for diagnosing object

Abstract

To provide a diagnostic method and a diagnostic system that can diagnose, at an early stage, an object that develops a sound portion over time due to natural factors, with a simple configuration.SOLUTION: The present invention provides a method and system for diagnosing an object including: an imaging step of diagnosing an object that develops a sound portion and an unsound portion over time and is discolored due to natural factors including disease, and acquiring photographed images of the object; a wavelength information acquisition step of acquiring wavelength information for each pixel from the photographed image; a two-dimensional scatter diagram acquisition step of acquiring a two-dimensional scatter diagram viewed from a direction where a difference between the sound portion and the unsound portion of a three-dimensional scatter diagram where each pixel is plotted in a color space based on the acquired wavelength information, is clearly visible; and a diagnosis step of diagnosing a factor of a phenomenon in which the object is discolored and a state of the object from the acquired two-dimensional scatter diagram.SELECTED DRAWING: Figure 7

Description

[Technical field]

[0001] The present invention relates to a method and system for diagnosing an object that changes color over time due to natural factors, including disease, and develops healthy and unhealthy parts. [Background technology]

[0002] A deterioration of the growth environment, pathogen infection, and the like can cause poor plant growth and progress in plant yellowing. In plant cultivation, it is important to accurately grasp and deal with the progress of plant yellowing. For this reason, for example, a pathogen infection diagnostic device has been proposed that irradiates a plant body with excitation light of a wavelength that excites pathogen tissue and grasps the pathogen infection of the plant (see, for example, Patent Document 1). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Patent Publication No. 2013-36889 Summary of the Invention [Problem to be solved by the invention]

[0004] However, the pathogen infection diagnostic device described in Patent Document 1 is a device specialized for diagnosing plant diseases, and is an expensive dedicated device equipped with a light source that irradiates excitation light of a specific wavelength at a predetermined intensity. In addition to plant diseases, there are various objects that change color over time due to natural factors that cause healthy and unhealthy parts, and there is a wide demand for detecting the discoloration of such objects and appropriately diagnosing them. In particular, there is a strong demand for diagnosis from an early stage without using an expensive dedicated device.

[0005] The present invention has been made in consideration of the above-mentioned problems, and has an object to provide a diagnostic method and diagnostic system that are simply configured and can diagnose, at an early stage, an object in which unhealthy parts develop over time due to natural factors. [Means for solving the problem]

[0006] One embodiment of the method for diagnosing the subject of the present invention comprises the steps of: A method for diagnosing an object that changes color over time due to natural factors, including disease, and develops healthy and unhealthy areas, comprising: An imaging step of acquiring a photographed image of the object; a wavelength information acquiring step of acquiring wavelength information for each pixel from the captured image; a two-dimensional scatter diagram acquisition step of acquiring a two-dimensional scatter diagram obtained by viewing a three-dimensional scatter diagram in which each pixel is plotted in a color space based on the acquired wavelength information from a direction in which a difference between a healthy portion and an unhealthy portion is significantly shown; a diagnosis step of diagnosing a cause of the discoloration phenomenon of the object and a state of the object from the acquired two-dimensional scatter diagram; A method for diagnosing an object, comprising:

[0007] One embodiment of the diagnostic system of the subject of the present invention comprises: A system for diagnosing an object that changes color over time due to natural factors, including disease, and develops healthy and unhealthy parts, An imaging unit for acquiring a photographed image of the object; a wavelength information acquisition unit that acquires wavelength information for each pixel from the captured image; a two-dimensional scatter diagram acquisition unit that acquires a two-dimensional scatter diagram obtained by viewing a three-dimensional scatter diagram in which each pixel is plotted in a color space based on the acquired wavelength information from a direction in which a difference between a healthy portion and an unhealthy portion is significantly shown; a diagnosis unit that diagnoses a cause of the discoloration phenomenon of the object and a state of the object from the acquired two-dimensional scatter diagram; The present invention relates to a diagnostic system for an object, comprising: Effect of the Invention

[0008] As described above, the above aspects can provide a diagnostic method and diagnostic system that can diagnose, at an early stage, an object in which unhealthy parts develop over time due to natural factors, with a simple configuration. [Brief description of the drawings]

[0009] [Figure 1A] 1 is a block diagram showing a configuration of an yellowing information acquisition system according to one embodiment of the present invention. [Figure 1B] FIG. 11 is a block diagram showing a configuration of a yellowing information acquisition system according to another embodiment of the present invention. [Diagram 2] FIG. 2 is a diagram showing an image acquisition section of a plant leaf. [Diagram 3] This is a three-dimensional scatter plot in which the RGB light intensity of each pixel of an acquired image is plotted in three-dimensional space, and is a view of a sheet-like point cloud viewed from a direction opposite to the sheet surface. [Figure 4] This is a three-dimensional scatter plot in which the RGB light intensity of each pixel of an acquired image is plotted in three-dimensional space, and is a view of a sheet-like point cloud viewed in a direction along the sheet surface. [Diagram 5] FIG. 13 is a diagram showing an example of a group of points obtained by plotting the RG light intensity of each pixel of an acquired image on an RG plane, and an example of an approximation straight line thereof. [Figure 6A] FIG. 13 is a diagram showing an example of an approximation straight line of a group of points obtained by plotting the RG light intensity of each pixel of an acquired image on a plane. [Figure 6B] FIG. 13 is a diagram showing an example of an approximation curve of a group of points obtained by plotting the RG light intensity of each pixel of an acquired image on a plane. [Figure 7] FIG. 13 is a diagram showing an example of a two-dimensional scatter diagram acquired based on the RGB light intensity of each pixel of an acquired image. [Figure 8A] FIG. 1 shows an example of a two-dimensional scatter plot of yellowed areas on leaves of a plant infected with a pathogen. [Figure 8B] FIG. 1 shows an example of a two-dimensional scatter plot of yellowed areas on leaves of a plant infected with a pathogen. [Figure 8C] FIG. 1 shows an example of a two-dimensional scatter plot of yellowed areas on leaves of a plant infected with a pathogen. [Figure 8D] FIG. 1 shows an example of a two-dimensional scatter plot of yellowed areas on leaves of a plant infected with a pathogen. [Figure 9] FIG. 1 is a diagram showing an example of a three-dimensional scatter plot in which the HSV intensity of each pixel of an image of a plant leaf is plotted in a three-dimensional space. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0010] Hereinafter, an embodiment for carrying out the present invention will be described with reference to the drawings. In each drawing, corresponding components having the same function are given the same reference numerals. In consideration of the explanation or ease of understanding of the main points, the embodiments may be shown separately for convenience, but partial replacement or combination of the configurations shown in different embodiments is possible. In the embodiments described below, the description of matters common to the above-mentioned embodiments will be omitted, and only the differences will be described. In particular, similar actions and effects due to similar configurations will not be mentioned in each embodiment.

[0011] (Method and system for diagnosing an object) Due to natural factors, healthy and unhealthy parts occur in an object over time, and discoloration occurs in various situations. For example, the object may be the surface of a leaf, shoot, flower, bud, stem, branch, trunk, or fruit of a plant, where discoloration occurs due to natural factors including disease. Furthermore, the object may not only be a plant, but may also be the surface of the skin, mucous membrane, or teeth of an animal or human, where discoloration occurs due to natural factors including disease. Furthermore, the object is not limited to a living thing, and may be a member having a metal surface that rusts.

[0012] A method for diagnosing an object according to one embodiment of the present invention is a method for diagnosing an object that changes color over time due to natural factors including disease, which cause healthy and unhealthy parts to appear. Specifically, the method for diagnosing according to this embodiment includes an imaging step of acquiring a photographed image of the object, a wavelength information acquisition step of acquiring wavelength information for each pixel from the photographed image, a two-dimensional scatter diagram acquisition step of acquiring a three-dimensional scatter diagram in which each pixel is plotted in a color space based on the acquired wavelength information, and a two-dimensional scatter diagram viewed from a direction in which the difference between healthy and unhealthy parts is clearly shown, and a diagnosis step of diagnosing the cause of the phenomenon in which the object changes color and the state of the object from the acquired two-dimensional scatter diagram.

[0013] A diagnostic system that implements such a diagnostic method is a system that diagnoses an object that becomes discolored over time due to natural factors including disease, as healthy and unhealthy parts appear, and the system includes an imaging unit that acquires a photographed image of the object, a wavelength information acquisition unit that acquires wavelength information for each pixel from the photographed image, a two-dimensional scatter plot acquisition unit that acquires a three-dimensional scatter plot in which each pixel is plotted in a color space based on the acquired wavelength information, and a two-dimensional scatter plot viewed from a direction in which the difference between healthy and unhealthy parts is clearly evident, and a diagnostic unit that diagnoses the cause of the discoloration of the object and the condition of the object from the acquired two-dimensional scatter plot.

[0014] For example, when the leaves of a plant turn yellow, this may be due to natural causes or a specific disease or injury, but the cause of the yellowing cannot be identified by visual inspection. Similarly, when rust appears on a metal surface, it cannot be identified by visual inspection whether it is red rust or black rust. However, by using the above-mentioned diagnostic method and diagnostic system, the cause of the yellowing can be identified.

[0015] Such discoloration due to natural phenomena has at least one or more stages and one or more characteristic parts, and when a particular characteristic part is at a particular discoloration stage, the color before and after the discoloration is constant, and the direction of the color change is the same.

[0016] For example, in the case of yellowing of plant leaves, there are two stages: from green to yellow and from yellow to death. In addition, there are two types of specific parts: the mesophyll part containing yellowing and the vein part. For leaves containing withering and yellowing, four sheets can be obtained: the mesophyll part containing yellowing, the vein part, the dead part, and the dead vein part. These can be evaluated separately, or some characteristic relationship between them can be found and evaluated.

[0017] The above-mentioned method for diagnosing an object and a diagnosing system for implementing this diagnosing method will be described in detail below, taking the yellowing of plant leaves as an example.

[0018] (Yellowing Information Acquisition System) First, a yellowing information acquisition system according to an embodiment of the present invention will be described with reference to Fig. 1A and Fig. 1B. Fig. 1A is a block diagram showing the configuration of a yellowing information acquisition system according to one embodiment of the present invention. Fig. 1B is a block diagram showing the configuration of a yellowing information acquisition system according to another embodiment of the present invention.

[0019] 1A and 1B each include an image acquisition unit 10 that acquires images of plant leaves, a control unit 80, and a display device 50 that displays the information acquired by the control unit 80. In order to control the implementation of a etiolation information acquisition method described below, the control unit 80 includes a light information acquisition unit 20, a scatter diagram acquisition unit 30, and an etiolation information acquisition unit 40. Image data of plant leaves acquired by the image acquisition unit 10 can be transmitted to the control unit 80, and the information acquired under the control of the control unit 80 can be transmitted to the display device 50 for display.

[0020] In the etiolation information acquisition system shown in FIG. 1A, the image acquisition unit 10 and the display device 50 are provided in the mobile terminal 70. The mobile terminal 70 and a server having a control unit 80 are connected by wireless or wired two-way communication. The user photographs the leaves of the plant with the image acquisition unit (camera) 10 of the mobile terminal 70 and transmits the photographed image data to a server having a remote control unit 80 by wireless or wired communication. The control unit 80 performs control processing based on the image data received from the mobile terminal 70, and transmits to the mobile terminal 70 a two-dimensional scatter diagram and etiolation information based on the two-dimensional scatter diagram acquired in the control processing. The two-dimensional scatter diagram and the etiolation information based on the two-dimensional scatter diagram will be described in detail later. The user can accurately grasp the state of etiolation of the plant by displaying the received two-dimensional scatter diagram and etiolation information on the display device 50 of the mobile terminal 70.

[0021] In the etiolation information acquisition system 100 shown in FIG. 1B, an image acquisition section 10, a control section 80, and a display device 50 are provided in one etiolation information acquisition device 90. A user photographs a plant leaf with the image acquisition section (camera) 10 of the etiolation information acquisition device 90. The photographed image data is transmitted to the control section 80. Based on the image data transmitted from the image acquisition section 10, the control section 80 acquires a two-dimensional scatter diagram and etiolation information based on the two-dimensional scatter diagram, and transmits the acquired data to the display device 50. The transmitted two-dimensional scatter diagram and etiolation information are displayed on the display device 50. The information displayed on the display device 50 allows the user to accurately grasp the etiolation state of the plant.

[0022] The yellowing information acquisition system of the present invention is not limited to the above configuration, and the image acquisition unit 10, the control unit 80, and the display device 50 may all be separate devices, or some of the elements may be provided in the same device.

[0023] (Method of acquiring yellowing information according to the first embodiment) Next, with reference to FIG. 2 to FIG. 7, the etiolation information acquisition method according to the first embodiment of the present invention, which is implemented in the above-mentioned etiolation information acquisition system 100, will be described. FIG. 2 is a diagram showing an image acquisition portion of a plant leaf. FIG. 3 is a three-dimensional scatter diagram in which the RGB light intensity of each pixel of an acquired image is plotted in a three-dimensional space, and is a diagram showing a sheet-like point cloud viewed from a direction opposite to the sheet surface. FIG. 4 is a three-dimensional scatter diagram in which the RGB light intensity of each pixel of an acquired image is plotted in a three-dimensional space, and is a diagram showing a sheet-like point cloud viewed from a direction along the sheet surface. FIG. 5 is a diagram showing an example of a point cloud in which the RG light intensity of each pixel of an acquired image is plotted on an RG plane, and an example of an approximate straight line thereof. FIG. 6A is a diagram showing an example of an approximate straight line of a point cloud in which the RG light intensity of each pixel of an acquired image is plotted on a plane. FIG. 6B is a diagram showing an example of an approximate curve of a point cloud in which the RG light intensity of each pixel of an acquired image is plotted on a plane. FIG. 7 is a diagram showing an example of a two-dimensional scatter diagram acquired based on the RGB light intensity of each pixel of an acquired image.

[0024] In the first embodiment, the light intensity of red light (hereinafter referred to as "R light intensity"), the light intensity of green light (hereinafter referred to as "G light intensity"), and the light intensity of blue light (hereinafter referred to as "B light intensity") for each pixel are obtained from the acquired image of a plant leaf, and control processing is performed.

[0025] <Image acquisition process> An image acquisition step is performed by the image acquisition unit 10 to acquire images of plant leaves. In FIG. 2, the leaf image acquisition portion is shown surrounded by a rectangle. Any image capture device, including a commercially available camera or a camera provided in a mobile terminal, can be used as the image acquisition unit 10. Leaves illuminated by natural light may be photographed, or leaves illuminated by white light from a light source may be photographed. When a light source is used, it is preferable to use a surface light source.

[0026] Next, the light information acquiring unit 20 of the control unit 80 performs a light information acquiring step of acquiring the R light intensity, the G light intensity, and the B light intensity of each pixel based on the image of the leaf region (see FIG. 2) acquired by the image acquiring unit 10. This allows data on the R light intensity, the G light intensity, and the B light intensity of each pixel of the leaf image to be obtained.

[0027] <Scatter plot acquisition process> The inventors have found, through repeated testing and analysis, that when the R light intensity, G light intensity, and B light intensity of each pixel are plotted in a three-dimensional space with the R axis, the G axis, and the B axis as the three axes, at least a part of the point cloud corresponding to each pixel is distributed in a sheet-like shape. Figure 3 is a diagram of the sheet-like point cloud as viewed from a direction opposite to the sheet surface, and Figure 4 is a diagram of the sheet-like point cloud as viewed from a direction along the sheet surface. That is, Figures 3 and 4 are diagrams as viewed from directions that are approximately perpendicular to each other.

[0028] As is clear from Figures 3 and 4, the point cloud is distributed to show a sheet shape standing on the RG plane. In the illustrated example, the sheet-like point cloud is distributed extending in a direction approximately perpendicular to the RG plane. However, this is not limited to this, and the sheet-like point cloud may be slightly inclined compared to the illustrated example. The sheet-like point cloud may have a certain thickness and may be curved, wavy, etc.

[0029] The inventors have found, through repeated testing and analysis, that when the sheet-like point cloud is viewed from a direction substantially perpendicular to the sheet surface, the difference between healthy and unhealthy parts of the leaf is clearly visible. FIG. 3 shows a view from a direction nearly perpendicular to the sheet surface. On the other hand, in a view from a direction along the sheet surface as shown in FIG. 4, the difference between healthy and unhealthy parts of the leaf cannot be grasped. As will be described later, the parts where the point cloud spreads vertically and horizontally and gathers in a planar shape, as indicated by arrow P in FIG. 3, are healthy parts, and the parts where the point cloud stretches out in an elongated shape, as indicated by arrow Q in FIG. 3, are unhealthy yellowed parts.

[0030] Regardless of the initial color difference of the plant leaves, at least a part of the point cloud corresponding to each pixel of the image is distributed in a sheet shape standing on the RG plane, as shown in Figures 3 and 4. For example, green leaves in the initial stage are distributed in a position away from the R axis where the G light intensity value is large on the RG plane. Red leaves in the initial stage are distributed in a position away from the G axis where the R light intensity value is large on the RG plane.

[0031] Therefore, the inventors have found that a two-dimensional scatter diagram that clearly shows the difference between healthy and unhealthy parts of a leaf can be obtained by the following method, without handling a three-dimensional scatter diagram that requires a large amount of data processing. Specifically, a two-dimensional scatter diagram that clearly shows the difference between healthy and unhealthy parts of a leaf can be obtained by a plan view with two axes, the value obtained from the projection of the light intensity of each pixel onto the RG plane and the corresponding B light intensity. Since the point cloud is distributed in a sheet-like state upright on the RG plane, the two-dimensional scatter diagram obtained in this way is a plan view similar to that when viewed from a direction approximately perpendicular to the sheet-like distribution of the point cloud shown in the three-dimensional scatter diagram.

[0032] In the scatter diagram acquisition step, the R light intensity and the G light intensity of each pixel are plotted on the RG plane to obtain an RG plane diagram as shown in FIG. 5. This RG plane diagram is equivalent to an RG projection diagram of the point cloud shown in the three-dimensional scatter diagram. Then, an approximation line or an approximation curve of the point cloud plotted on this RG plane is formed. To obtain the approximation line or the approximation curve, any known linear regression method or nonlinear regression method such as the least squares method can be adopted.

[0033] Then, the scatter diagram acquisition unit 30 plots the values ​​of each point using the obtained approximate line or curve as a gauge and the B light intensity corresponding to the point on a plane to acquire a two-dimensional scatter diagram as shown in Fig. 7. "Using an approximate line or curve as a gauge" will be explained below with reference to the approximate line shown in Fig. 6A and the approximate curve shown in Fig. 6B.

[0034] In Fig. 6A and Fig. 6B, the approximation line and the approximation curve are shown extending from the origin. If the approximation line and the approximation curve do not pass through the origin, the offset can be adjusted to correct them so that they pass through the origin. An example of a value determined using the approximation line or the approximation curve as a gauge is the distance from the origin to each point along the approximation line or the approximation curve. If the coordinates of each point in Fig. 6A and Fig. 6B are X and Y, then in the case of the approximation line shown in Fig. 6A, the value determined using the approximation line as a gauge is X. 2 +Y 2 It can be expressed as the square root of

[0035] However, this is not limited to this. In the case of an approximate straight line, the sum of the R intensity and the G intensity of each point (i.e., X+Y) can be used, or the sum of squares (X 2 +Y 2 In the case of an approximation curve, it can be, for example, the value of a line integral expressed as the distance from the origin. Furthermore, any method relating to variance and distribution, such as the Mahalanobis distance, can be used to determine a value using an approximation line or curve as a gauge.

[0036] As described above, in the scatter diagram acquisition process, the R light intensity and G light intensity of each pixel are plotted on the RG plane, and the value of each point is plotted on the plane using the approximate straight line or curve of the point group as a gauge, and the corresponding B light intensity of each pixel is plotted. In this way, a two-dimensional scatter diagram as shown in FIG. 7 is acquired.

[0037] The two-dimensional scatter diagram shown in Fig. 7 clearly shows the difference between healthy and unhealthy parts of the leaves, similar to the three-dimensional scatter diagram shown in Fig. 3, which shows the sheet-like point cloud from a direction substantially perpendicular to the sheet surface. When dealing with a three-dimensional scatter diagram, for example, a large amount of data processing is required when tilting or rotating the point cloud in three-dimensional space. In this embodiment, a two-dimensional scatter diagram can be obtained with simple control processing that requires less data processing than when dealing with a cubic scatter diagram.

[0038] Furthermore, in the two-dimensional scatter plot obtained in this way, the area on the surface spreading left and right as indicated by the arrow P in Figure 7 indicates the healthy part, and the elongated angular area extending to the right as indicated by the arrow Q indicates the unhealthy part where yellowing has occurred. Even in the early stage when yellowing begins, the angular area extending from the planar area spreading left and right of the point cloud can be confirmed. Therefore, the state of yellowing of the plant can be grasped from the early stage by using the two-dimensional scatter plot.

[0039] As will be described later, the elongated, angular region indicated by arrow Q tends to have a specific shape depending on the pathogen that has infected it, and the shape can also be used to identify the pathogen that caused the yellowing.

[0040] As described above, the inventors have discovered that at least a portion of the points shown in the three-dimensional scatter plot are distributed in a sheet-like shape standing up from the RG plane, and the following is thought to be the cause of this.

[0041] In the initial state, the distribution of each dot in the image of the plant leaf is not scattered, and as time passes, yellowing areas appear, but each dot in the captured image does not reverse in either direction of the RGB axis, but changes in a certain direction. Therefore, each dot generally draws a straight or curved trajectory, and the direction of change of each dot tends to be the same. For this reason, at least a part of the dot group shown in the three-dimensional scatter diagram is considered to be distributed in a sheet shape.

[0042] For the same leaf, even over time, a sheet-like distribution of points on the same RG plane will be obtained. Therefore, without creating a three-dimensional scatter diagram using three-dimensional data, by creating a two-dimensional scatter diagram with the values ​​obtained from the approximate straight line or approximate curve of the points plotted on the RG plane and the B light intensity as the two axes, as described above, it is possible to grasp the changes over time between healthy and unhealthy parts of the leaf.

[0043] In this way, since there is no need to process huge amounts of data and display and handle three-dimensional data, the processing time can be significantly shortened and the memory capacity required for processing can be reduced.In addition, by using two-dimensional dispersion images, which clearly show the difference between healthy and unhealthy parts, users can distinguish healthy and unhealthy parts visually without the need for black box analysis using AI, or they can distinguish healthy and unhealthy parts using a simpler algorithm.

[0044] Discoloration caused by natural phenomena has at least one or more stages and one or more characteristic parts, and when a particular characteristic part is at a particular discoloration stage, the color before and after discoloration is fixed, and the direction of the color change is considered to be the same. Taking a leaf that is green in the early stages as an example, there are two stages mentioned above: the stage where it turns from green to yellow, and the stage where it turns from yellow to withering and dying. In addition, there are two types of specific parts: the leaf pulp part, which includes yellowing, and the leaf vein part. Furthermore, perforations may occur as specific parts due to withering or insect damage.

[0045] For leaves that are dead and yellowed, a three-dimensional scatter plot may give a sheet-like distribution of points for each of the four clusters: yellowed mesophyll, vein, dead mesophyll, and dead vein. In this case, two-dimensional scatter plots for each cluster can be obtained. These two-dimensional scatter plots can be evaluated individually, or some characteristic relationship can be found between the two-dimensional scatter plots for evaluation.

[0046] Depending on the condition of the leaf, the mesophyll and vein parts may not show distinct sheet-like distributions, and the point cloud distributed in a single sheet may tend to have a slightly larger variation in the thickness direction of the sheet. To grasp the condition of the plant leaves more accurately, it is preferable that the point cloud distributed in a sheet-like manner shown in the three-dimensional scatter plot has less variation in the thickness direction of the sheet. Therefore, the vein part is identified from the captured image using image software, and the light intensity data of the vein part is removed as necessary, thereby obtaining a sheet-like distribution of the point cloud with less variation in the thickness direction.

[0047] For this reason, it is preferable to process the image obtained in the image acquisition step so that the veins of the plant are clear and uniform, and then carry out the light information acquisition step. This makes it possible to remove the veins, for example, by binarization or the like.

[0048] By obtaining images in which the veins of a plant's leaves are clearly visible and uniform, it becomes possible to distinguish and analyze the veins from other parts of the leaf, such as the leaf pulp, and thus to grasp the state of yellowing more accurately.

[0049] Furthermore, in the scatter diagram acquisition step, it is preferable to acquire the two-dimensional scatter diagram in a state where at least one of the mesophyll portion, discolored portion, dead portion, vein portion, and perforated portion of the plant leaves is removed, thereby making it possible to obtain a two-dimensional scatter diagram that allows a more accurate understanding of the state of yellowing.

[0050] <Yellowing information acquisition process> Even when yellowing appears to be the same to the naked eye, by using the two-dimensional scatter diagram, it is possible to identify whether the yellowing is natural or caused by a specific disease or injury. Next, the yellowing information acquisition process performed by the yellowing information acquisition unit 40 will be described with reference to Figs. 8A to 8D. Figs. 8A to 8D are each a diagram showing an example of a two-dimensional scatter diagram of the yellowing area of ​​the leaves of a plant infected with a pathogen.

[0051] FIG. 8A shows an example where the shape of the horn-like extension of the etiolated area is distorted. FIG. 8B shows an example where two horn-like extensions appear in the etiolated area. FIG. 8C shows an example where the tip of the horn-like extension of the etiolated area has a vertically elongated shape. FIG. 8D shows an example where the tip of the horn-like extension of the etiolated area has a white spot cluster.

[0052] As shown in Figures 8A to 8D, in the yellowing information acquisition process performed by the yellowing information acquisition unit 40, the cause of yellowing can be identified mainly by analyzing the shape of the angular extensions of the yellowed area, but it is also important to identify the cause of yellowing by analyzing the shape of the entire point cloud including other parts.

[0053] To achieve this, it is effective to photograph leaves whose cause of yellowing has been analyzed by another test, obtain a two-dimensional scatter diagram, and use the two-dimensional scatter diagram as a standard two-dimensional scatter diagram to obtain yellowing information. In the yellowing information obtaining step, the cause of yellowing can be easily and reliably identified by comparing the two-dimensional scatter diagram obtained by actually photographing the leaves of a plant with the standard two-dimensional scatter diagram.

[0054] When the determination is made by comparing the shape of the angular extensions of the yellowed areas, there is no need to perform a black box analysis using AI, and the user can easily identify the cause of the yellowing.

[0055] Furthermore, it is also effective to obtain the time series changes of the same leaf as the change in the shape of the dots in the two-dimensional scatter plot. This makes it possible to accurately identify the progress of the yellowing. By accumulating the data of the two-dimensional scatter plot that changes over time and reflects the progress of the yellowing, it is possible to accurately determine the current state of the plant and take prompt action.

[0056] Furthermore, it is also effective to identify the feature values ​​of the two-dimensional scatter diagram using a statistical method or the like. For example, the cause of yellowing can be identified according to a certain standard by a threshold value set based on the accumulated feature value data. Thus, in this embodiment, in the yellowing information acquisition step, the cause and progress of yellowing can be diagnosed based on the shape and feature values ​​of the two-dimensional scatter diagram. This makes it possible to grasp the condition of the plant at an early stage, understand the cause, and realize an appropriate and prompt response.

[0057] In addition, in the yellowing information acquisition step, shape information on the real plane of each point corresponding to a specific cluster in the two-dimensional scatter diagram can also be acquired. Examples of the specific cluster here include a healthy area of ​​the mesophyll portion, a yellowed area of ​​the mesophyll portion, a healthy area of ​​the vein portion, and a yellowed area of ​​the vein portion. This makes it possible to add yellowing information to an actual leaf image as shown in FIG. 2 and notify the user. By acquiring shape information on the real plane of each point corresponding to a specific cluster, it becomes possible to efficiently notify the user of yellowing information in accordance with the convenience of the user. In addition, machine learning of a two-dimensional scatter plot or the shapes on a real plane of each point corresponding to a specific cluster in the two-dimensional scatter plot is also effective in obtaining yellowing information.

[0058] The two-dimensional scatter diagram obtained by the scatter diagram acquisition unit 30 and the yellowing information obtained by the yellowing information acquisition unit 40 can be displayed on the display device 50 to provide feedback to the user. Feedback to the user can be provided not only by display on the display device 50, but also by any known notification means including printing on paper and notifying the yellowing information by voice.

[0059] As described above, the etiolation information acquisition method according to the first embodiment of the present invention includes an image acquisition step of acquiring an image of a plant leaf, a light information acquisition step of acquiring the red light intensity (hereinafter referred to as "R light intensity"), the green light intensity (hereinafter referred to as "G light intensity"), and the blue light intensity (hereinafter referred to as "B light intensity") of each pixel from the image, and a scatter plot acquisition step of acquiring a two-dimensional scatter plot with the value of each point as a gauge, using an approximate straight line or curve of a point cloud obtained by plotting the R light intensity and G light intensity of each pixel on an RG plane, and the corresponding B light intensity of each pixel as the two axes. a step of acquiring information on the etiolation of a plant based on the two-dimensional scatter diagram; Includes.

[0060] Similarly, the etiolation information acquisition system 100 according to the first embodiment of the present invention, which implements the etiolation information acquisition method according to the first embodiment described above, includes an image acquisition unit 10 that acquires images of plant leaves, a light information acquisition unit 20 that acquires the red light intensity (hereinafter referred to as "R light intensity"), green light intensity (hereinafter referred to as "G light intensity"), and blue light intensity (hereinafter referred to as "B light intensity") of each pixel from the image acquired by the image acquisition unit 10, a scatter plot acquisition unit 30 that acquires a two-dimensional scatter plot with the value of each point as a gauge, using an approximation line or curve of a point cloud obtained by plotting the R light intensity and G light intensity of each pixel on an RG plane, and the corresponding B light intensity of each pixel as the two axes, and an etiolation information acquisition unit 40 that acquires etiolation information of the plant based on the two-dimensional scatter plot.

[0061] The etiolation information acquisition system 100 according to the first embodiment of the present invention does not require a light source for irradiating excitation light, and can acquire two-dimensional scatter diagrams and etiolation information with a simple configuration of an image acquisition unit 10 for photographing plant leaves and a control unit 80. In addition, the control unit 80 can obtain a two-dimensional dispersion image in which the difference between healthy and unhealthy parts of the leaves is clearly shown without requiring a large amount of control processing for forming and analyzing three-dimensional images. In the early stage of etiolation, healthy areas and etiolated unhealthy areas are mixed together and difficult to distinguish from the image of the leaves alone. However, in this embodiment, by acquiring a two-dimensional dispersion image, it is possible to specify the state of etiolation of the plant from the early stage, and various etiolation information as described above can be obtained. Note that the two axes used to acquire the two-dimensional dispersion diagram may not be orthogonal.

[0062] As a result, it is possible to provide a method for acquiring information about etiolation and a system for acquiring information about etiolation 100 that can specify the state of etiolation of a plant from an early stage with a simple configuration.

[0063] (Method of acquiring yellowing information according to the second embodiment) In the first embodiment, a two-dimensional scatter diagram is obtained in which the R light intensity and G light intensity of each pixel are plotted on the RG plane, and the value of each point is measured as a gauge using an approximate straight line or curve of a group of points, and the corresponding B light intensity of each pixel is measured as two axes. In other words, a two-dimensional scatter diagram viewed from a direction approximately perpendicular to the approximate straight line shown on the RG plane shown in FIG. 5 is obtained.

[0064] The approximation line shown in FIG. 5 extends obliquely to the R axis and the G axis, so when viewed from the R axis direction or the G axis direction, the approximation line is viewed obliquely. However, even if a two-dimensional scatter diagram viewed from the R axis direction or the G axis direction, which is a diagonal view of this sheet-like distribution, is obtained, it is possible to distinguish healthy parts of the leaves from unhealthy parts. Regardless of the angle of the approximation line to the R axis and the G axis, it is possible to distinguish healthy parts of the leaves from unhealthy parts in the two-dimensional scatter diagram viewed from at least one of the R axis direction and the G axis direction. However, the accuracy is inferior to that of the first embodiment viewed from a direction approximately perpendicular to the approximation line shown on the RG plane.

[0065] In that case, in the light information acquisition step, a two-dimensional scatter diagram can be formed simply by taking the R light intensity or G light intensity of each pixel as one axis and the B light intensity as the other axis.

[0066] As described above, the etiolation information acquisition method according to the second embodiment of the present invention includes an image acquisition step of acquiring an image of a plant leaf, a light information acquisition step of acquiring the red light intensity (hereinafter referred to as "R light intensity"), green light intensity (hereinafter referred to as "G light intensity"), and blue light intensity (hereinafter referred to as "B light intensity") of each pixel from the image, a scatter plot acquisition step of acquiring a two-dimensional scatter plot with the R light intensity or G light intensity and the B light intensity of each pixel as the two axes, and an etiolation information acquisition step of acquiring plant etiolation information based on the two-dimensional scatter plot.

[0067] Similarly, the etiolation information acquisition system 100 according to the second embodiment of the present invention, which implements the etiolation information acquisition method according to the second embodiment described above, includes an image acquisition unit 10 that acquires images of plant leaves, a light information acquisition unit 20 that acquires the red light intensity (hereinafter referred to as "R light intensity"), green light intensity (hereinafter referred to as "G light intensity"), and blue light intensity (hereinafter referred to as "B light intensity") of each pixel from the image acquired by the image acquisition unit, a scatter plot acquisition unit 30 that acquires a two-dimensional scatter plot with the R light intensity or G light intensity and the B light intensity of each pixel as the two axes, and an etiolation information acquisition unit 40 that acquires plant etiolation information based on the two-dimensional scatter plot.

[0068] The yellowing information acquisition method and the yellowing information acquisition system 100 according to the second embodiment are particularly effective in cases where high speed and large volume processing is desired.

[0069] (Method of acquiring yellowing information according to the third embodiment) In the above first and second embodiments, RGB light intensity is used as light information, but H intensity (hue), S intensity (saturation) and V intensity (brightness) can also be used instead of RGB light intensity. Next, a method for acquiring yellowing information according to a third embodiment of the present invention using HSV intensity will be described with reference to Fig. 9. Fig. 9 is a diagram showing an example of a three-dimensional scatter diagram in which the HSV intensity of each pixel of an image of a plant leaf is plotted in a three-dimensional space.

[0070] As is clear from Fig. 9, even when HSV intensity is used, a sheet-like distribution of points standing on the HV plane is obtained, as in the case of using RGB light intensity shown in Fig. 3. In Fig. 9, too, the area on the surface extending to the left and right indicated by arrow P indicates a healthy part, and the elongated angular area extending to the right indicated by arrow Q indicates an unhealthy part where yellowing has occurred.

[0071] Therefore, a two-dimensional scatter diagram can be obtained with the value of each point, which is the approximate straight line or curve of the point group obtained by plotting the H intensity and V intensity of each pixel on the HV plane, as the gauge, and the corresponding S intensity of each pixel as the two axes. In other words, by replacing the RGB light intensity in the first embodiment with the HV intensity and the B light intensity with the S intensity, an analysis similar to that in the first embodiment can be performed.

[0072] As described above, the yellowing information acquisition method according to the third embodiment of the present invention includes an image acquisition step of acquiring an image of a plant leaf, a light information acquisition step of acquiring the hue (hereinafter referred to as "H intensity"), saturation (hereinafter referred to as "S intensity"), and lightness (hereinafter referred to as "V intensity") of each pixel from the image, a scatter plot acquisition step of acquiring a two-dimensional scatter plot with the values ​​of each point as a gauge, using an approximate straight line or curve of a point cloud obtained by plotting the H intensity and V intensity of each pixel on an HV plane, and the corresponding S intensity of each pixel as the two axes, and a yellowing information acquisition step of acquiring plant yellowing information based on the two-dimensional scatter plot.

[0073] Similarly, the etiolation information acquisition system 100 according to the third embodiment of the present invention, which implements the etiolation information acquisition method according to the third embodiment described above, includes an image acquisition unit 10 that acquires images of plant leaves, a light information acquisition unit 20 that acquires the hue (hereinafter referred to as "H intensity"), saturation (hereinafter referred to as "S intensity"), and lightness (hereinafter referred to as "V intensity") of each pixel from the image acquired by the image acquisition unit 10, a scatter plot acquisition unit 30 that acquires a two-dimensional scatter plot with the value of each point as a gauge, using an approximation straight line or curve of a point cloud obtained by plotting the H intensity and V intensity of each pixel on an HV plane, and the corresponding S intensity of each pixel as the two axes, and an etiolation information acquisition unit 40 that acquires plant etiolation information based on the two-dimensional scatter plot.

[0074] As a result, in the third embodiment using HSV intensity as well, similar to the first embodiment using RGB light intensity, it is possible to provide a yellowing information acquisition method and a yellowing information acquisition system 100 that can identify the yellowing state of a plant from an early stage with a simple configuration.

[0075] (Yellowing information acquisition method according to the fourth embodiment) Even when HSV intensity is used, similarly to the above third embodiment, even if a two-dimensional scatter diagram is obtained as viewed from the H-axis or V-axis direction, which is a sheet-like distribution as shown in Fig. 9 viewed from an oblique direction, it is possible to distinguish healthy and unhealthy parts of the leaves. Regardless of the angle of the approximation line with respect to the H-axis and V-axis, it is possible to distinguish healthy and unhealthy parts of the leaves in a two-dimensional scatter diagram viewed from at least one of the H-axis and V-axis directions. However, the accuracy is inferior to that of the third embodiment, which is viewed from a direction approximately perpendicular to the approximation line shown on the HV plane.

[0076] Therefore, in the yellowing information acquisition method according to the fourth embodiment of the present invention, in the scatter plot acquisition step, a two-dimensional scatter plot can be simply formed by using the H intensity or V intensity of each pixel as one axis and the S intensity as the other axis.

[0077] As described above, the yellowing information acquisition method according to the fourth embodiment of the present invention includes an image acquisition step of acquiring an image of a plant leaf, a light information acquisition step of acquiring the hue (hereinafter referred to as "H intensity"), saturation (hereinafter referred to as "S intensity"), and lightness (hereinafter referred to as "V intensity") of each pixel from the image, a scatter plot acquisition step of acquiring a two-dimensional scatter plot with the H intensity or V intensity and the S intensity of each pixel as the two axes, and a yellowing information acquisition step of acquiring plant yellowing information based on the two-dimensional scatter plot.

[0078] Similarly, the etiolation information acquisition system 100 according to the fourth embodiment of the present invention, which implements the etiolation information acquisition method according to the above-mentioned fourth embodiment, includes an image acquisition unit 10 that acquires images of plant leaves, a light information acquisition unit 20 that acquires the hue (hereinafter referred to as "H intensity"), saturation (hereinafter referred to as "S intensity"), and lightness (hereinafter referred to as "V intensity") of each pixel from the image acquired by the image acquisition unit 10, a scatter plot acquisition unit 30 that acquires a two-dimensional scatter plot with the H intensity or V intensity and the S intensity of each pixel as the two axes, and a etiolation information acquisition unit 40 that acquires plant etiolation information based on the two-dimensional scatter plot.

[0079] The yellowing information acquisition method and the yellowing information acquisition system 100 according to the fourth embodiment are particularly effective in cases where high speed and large volume processing is desired.

[0080] In any of the first to fourth embodiments, evaluation is not only performed for each axis of RGB or HSV, but also highly sensitive evaluation can be performed including how the respective quantities are related. A plurality of elements (RGB or HSV) can be accurately evaluated including their respective relationships. Furthermore, in any of the first to fourth embodiments, the two axes used to obtain a two-dimensional scatter diagram may not be orthogonal.

[0081] (Method and system for diagnosing an object) This method can be applied not only to obtaining information on yellowing, but also to diagnosing objects that change color over time due to natural factors, where healthy and unhealthy areas appear. The image acquisition step according to the first to fourth embodiments corresponds to an image capturing step of capturing an image of an object, The light information acquisition step according to the first to fourth embodiments corresponds to a wavelength information acquisition step of acquiring wavelength information for each pixel from a captured image, The scatter diagram acquisition step according to the first to fourth embodiments corresponds to a two-dimensional scatter diagram acquisition step of acquiring a two-dimensional scatter diagram obtained by viewing a three-dimensional scatter diagram in which each pixel is plotted in a color space based on the acquired wavelength information from a direction in which a difference between a healthy portion and an unhealthy portion is clearly shown, The yellowing information acquisition process according to the first to fourth embodiments corresponds to a diagnosis process for diagnosing the cause of the phenomenon in which the object discolors and the state of the object from the acquired two-dimensional scatter diagram. In the diagnosis process, in addition to acquiring information on the phenomenon that has occurred, the cause of the phenomenon and the state of the object are diagnosed.

[0082] That is, the diagnostic method according to one embodiment of the present invention comprises: A method for diagnosing an object that discolors over time due to natural factors including disease, as healthy and unhealthy parts appear, includes an imaging step of acquiring an image of the object, a wavelength information acquisition step of acquiring wavelength information for each pixel from the captured image, a two-dimensional scatter plot acquisition step of acquiring a three-dimensional scatter plot in which each pixel is plotted in a color space based on the acquired wavelength information, and a two-dimensional scatter plot viewed from a direction in which the difference between healthy and unhealthy parts is clearly evident, and a diagnosis step of diagnosing the cause of the phenomenon of discoloration of the object and the condition of the object from the acquired two-dimensional scatter plot.

[0083] Similarly, if we apply this to a system that not only obtains information on yellowing, but also diagnoses objects that change color over time due to natural factors, such as healthy and unhealthy areas, then: The image acquisition unit 10 corresponds to an imaging unit that acquires a photographed image of an object. The light information acquisition unit 20 corresponds to a wavelength information acquisition unit that acquires wavelength information for each pixel from a captured image, The scatter diagram acquisition unit 30 corresponds to a two-dimensional scatter diagram acquisition unit that acquires a two-dimensional scatter diagram in which a three-dimensional scatter diagram in which each pixel is plotted in a color space based on the acquired wavelength information is viewed from a direction in which a difference between a healthy part and an unhealthy part is significantly shown, The yellowing information acquisition unit 40 corresponds to a diagnosis unit that diagnoses the cause of the phenomenon in which the object discolors and the state of the object from the acquired two-dimensional scatter diagram. The diagnosis unit not only acquires information on the phenomenon that has occurred, but also diagnoses the cause of the phenomenon and the state of the object.

[0084] In other words, a diagnostic system according to one embodiment of the present invention is a system for diagnosing an object that becomes discolored over time due to natural factors including disease, as healthy and unhealthy parts appear, and the system comprises an imaging unit that acquires a photographed image of the object, a wavelength information acquisition unit that acquires wavelength information for each pixel from the photographed image, a two-dimensional scatter plot acquisition unit that acquires a three-dimensional scatter plot in which each pixel is plotted in a color space based on the acquired wavelength information, and a two-dimensional scatter plot viewed from a direction in which the difference between healthy and unhealthy parts is clearly evident, and a diagnostic unit that diagnoses the cause of the discoloration of the object and the condition of the object from the acquired two-dimensional scatter plot.

[0085] As described above, it is possible to provide a diagnostic method and diagnostic system 100 that can diagnose, at an early stage, an object that develops unhealthy parts over time due to natural factors, with a simple configuration. Such a diagnostic method and diagnostic system 100 can be generally described as follows.

[0086] (General Description) (Section 1-1) A method for diagnosing an object that changes color over time due to natural factors, including disease, and develops healthy and unhealthy areas, comprising: An imaging step of acquiring a photographed image of the object; a wavelength information acquiring step of acquiring wavelength information for each pixel from the captured image; a two-dimensional scatter diagram acquisition step of acquiring a two-dimensional scatter diagram obtained by viewing a three-dimensional scatter diagram in which each pixel is plotted in a color space based on the acquired wavelength information from a direction in which a difference between a healthy portion and an unhealthy portion is significantly shown; a diagnosis step of diagnosing a cause of the discoloration phenomenon of the object and a state of the object from the acquired two-dimensional scatter diagram; A method for diagnosing an object, comprising: (Section 1-2) The diagnostic method according to item 1-1, characterized in that the discoloration phenomenon caused by a natural phenomenon has at least one or more stages and one or more characteristic sites, and when a specific characteristic site is at a specific discoloration stage, the color before and after the discoloration is constant, and the direction of the color change is the same. (Section 1-3) The diagnostic method according to item 1-1, wherein in the color information acquisition step, only pixel data corresponding to a specific characteristic portion and a specific discoloration stage is extracted. (Section 1-4) The diagnostic method described in item 1-1, wherein when a certain characteristic part of the object is at a certain stage of discoloration, the corresponding group of dots forms a three-dimensional scatter diagram in color space that is distributed in an approximately sheet-like shape, and the image viewed from a direction approximately parallel to the approximate plane of the group of dots distributed in the approximately sheet-like shape is used as the two-dimensional scatter diagram. (Section 1-5) The diagnostic method according to item 1-1, further comprising a step of obtaining a regression line or regression curve when the three-dimensional scatter diagram is viewed from a direction in which it is contracted to a specific line or curve. (Section 1-6) The diagnostic method according to item 1-1, further comprising forming an image in which information on healthy and unhealthy parts is added to the captured image based on position information in the captured image corresponding to each dot in the two-dimensional scatter plot. (Section 1-7) The diagnostic method according to item 1-1, wherein the cause of the discoloration phenomenon of the object and the state of the object are diagnosed from the shape of the healthy part or the unhealthy part, or both, shown in the image to which information on the healthy part and the unhealthy part is added. (Section 1-8) The diagnostic method according to item 1-1, wherein the color space is a three-dimensional space with RGB or HSV as three axes.

[0087] (Section 1-9) The diagnostic method described in item 1-1, in which the two-dimensional scatter plot obtained using the object whose condition has been grasped by another means is used as a standard two-dimensional scatter plot, and in the diagnostic process, the two-dimensional scatter plot obtained by imaging the object is compared with the standard two-dimensional scatter plot to diagnose the cause of the discoloration phenomenon of the object and the condition of the object. (Section 1-10) The diagnostic method according to item 1-1, wherein the regression line or regression curve obtained by using the object whose condition has been grasped by another means is used as a standard regression line, and in the diagnostic step, the regression line obtained by imaging the object is compared with the standard regression line to diagnose a cause of the discoloration phenomenon of the object and the condition of the object. (Section 1-11) The diagnostic method according to item 1-1, further comprising a function of improving the flatness of the approximate sheet distribution by applying a certain operation to all plots so as to convert the regression curve into a straight line. (Section 1-12) The diagnostic method according to item 1-1, further comprising obtaining a time series change in color of the same object as a change in shape of the two-dimensional scatter plot. (Section 1-13) The diagnostic method according to item 1-1, further comprising a step of comparing the shape of the two-dimensional scatter plot or the regression line or both for each of a plurality of the stages or the characteristic sites or both. (Section 1-14) The diagnostic method according to any one of Items 1-1 to 1-13, wherein the target object is the surface of a leaf, shoot, flower, bud, stem, branch, trunk, or fruit of a plant that is discolored due to natural factors, including disease. (Section 1-15) The diagnostic method according to any one of Items 1-1 to 1-13, wherein the target object is the surface of the skin, mucous membrane, or teeth of an animal or human that is discolored due to natural factors including disease. (Section 1-16) The diagnostic method according to any one of items 1-1 to 1-13, wherein the object is a member having a metal surface on which rust occurs. (Section 1-17) A system for diagnosing an object that changes color over time due to natural factors, including disease, and develops healthy and unhealthy parts, An imaging unit for acquiring a photographed image of the object; a wavelength information acquisition unit that acquires wavelength information for each pixel from the captured image; a two-dimensional scatter diagram acquisition unit that acquires a two-dimensional scatter diagram obtained by viewing a three-dimensional scatter diagram in which each pixel is plotted in a color space based on the acquired wavelength information from a direction in which a difference between a healthy portion and an unhealthy portion is significantly shown; a diagnosis unit that diagnoses a cause of the discoloration phenomenon of the object and a state of the object from the acquired two-dimensional scatter diagram; A diagnostic system for an object comprising:

[0088] The yellowing information acquisition method and the yellowing information acquisition system 100 can be generally described as follows. (Section 2-1) an image acquisition step of acquiring an image of a plant leaf; a light information acquisition step of acquiring the red light intensity (hereinafter referred to as "R light intensity"), the green light intensity (hereinafter referred to as "G light intensity"), and the blue light intensity (hereinafter referred to as "B light intensity") of each pixel from the image; a scatter diagram acquisition step of acquiring a two-dimensional scatter diagram with the value of each point on the two axes being the value of the corresponding B light intensity of each pixel, and the approximation line or curve of the point group obtained by plotting the R light intensity and the G light intensity of each pixel on the R-G plane as a gauge; a step of acquiring information on the etiolation of a plant based on the two-dimensional scatter diagram; The method for acquiring yellowing information includes: (Section 2-2) an image acquisition step of acquiring an image of a plant leaf; a light information acquisition step of acquiring the hue (hereinafter referred to as "H intensity"), saturation (hereinafter referred to as "S intensity"), and lightness (hereinafter referred to as "V intensity") of each pixel from the image; A scatter plot acquisition step of acquiring a two-dimensional scatter plot in which the value of each point, which is an approximate straight line or curve of a point group obtained by plotting the H intensity and V intensity of each pixel on an HV plane, and the corresponding S intensity of each pixel are used as two axes; a step of acquiring information on the etiolation of a plant based on the two-dimensional scatter diagram; The method for acquiring yellowing information includes: (Section 2-3) an image acquisition step of acquiring an image of a plant leaf; a light information acquisition step of acquiring the red light intensity (hereinafter referred to as "R light intensity"), the green light intensity (hereinafter referred to as "G light intensity"), and the blue light intensity (hereinafter referred to as "B light intensity") of each pixel from the image; a scatter diagram acquisition step of acquiring a two-dimensional scatter diagram with the R light intensity or G light intensity and the B light intensity of each pixel as two axes; a step of acquiring information on the etiolation of a plant based on the two-dimensional scatter diagram; The method for acquiring yellowing information includes: (Section 2-4) an image acquisition step of acquiring an image of a plant leaf; a light information acquisition step of acquiring the hue (hereinafter referred to as "H intensity"), saturation (hereinafter referred to as "S intensity"), and lightness (hereinafter referred to as "V intensity") of each pixel from the image; A scatter plot acquisition step of acquiring a two-dimensional scatter plot having H intensity or V intensity and S intensity of each pixel as two axes; a step of acquiring information on the etiolation of a plant based on the two-dimensional scatter diagram; The method for acquiring yellowing information includes: (Section 2-5) The method for acquiring information on etiolation according to any one of Items 2-1 to 2-4, further comprising processing the image obtained in the image acquiring step so that leaf veins of the plant are clear and uniform, and then carrying out the light information acquiring step. (Section 2-6) The method for acquiring information on yellowing according to any one of items 2-1 to 2-5, wherein in the scatter diagram acquisition step, the two-dimensional scatter diagram is acquired in a state in which at least one of mesophyll parts, discolored parts, dead parts, vein parts, and perforated parts on the plant leaves is removed. (Section 2-7) Item 7. The method for acquiring information on yellowing according to any one of Items 2-1 to 2-6, wherein in the yellowing information acquiring step, a cause and a degree of progress of yellowing are diagnosed based on a shape and feature values ​​of the two-dimensional scatter diagram. (Section 2-8) The method for acquiring information on yellowing according to any one of Items 2-1 to 2-7, wherein in the yellowing information acquisition step, shape information on a real plane of each point corresponding to a specific cluster in the two-dimensional scatter plot is acquired.

[0089] (Section 2-9) an image acquisition unit that acquires an image of a plant leaf; a light information acquisition unit that acquires red light intensity (hereinafter referred to as "R light intensity"), green light intensity (hereinafter referred to as "G light intensity"), and blue light intensity (hereinafter referred to as "B light intensity") of each pixel from the image acquired by the image acquisition unit; a scatter diagram acquisition unit that acquires a two-dimensional scatter diagram with the value of each point on the two axes being the value of the corresponding B light intensity of each pixel, and the approximation line or curve of the point group obtained by plotting the R light intensity and the G light intensity of each pixel on the R / G plane as a gauge; a yellowing information acquisition unit that acquires yellowing information of a plant based on the two-dimensional scatter diagram; The yellowing information acquisition system includes: (Section 2-10) an image acquisition unit that acquires an image of a plant leaf; a light information acquisition unit that acquires the hue (hereinafter referred to as "H intensity"), saturation (hereinafter referred to as "S intensity"), and lightness (hereinafter referred to as "V intensity") of each pixel from the image acquired by the image acquisition unit; a scatter diagram acquisition unit that acquires a two-dimensional scatter diagram with the value of each point on the two axes being the value of the corresponding S intensity of each pixel, and the value of each point being an approximate straight line or curve of a point group obtained by plotting the H intensity and V intensity of each pixel on an HV plane; a yellowing information acquisition unit that acquires yellowing information of a plant based on the two-dimensional scatter diagram; The yellowing information acquisition system includes: (Section 2-11) an image acquisition unit that acquires an image of a plant leaf; a light information acquisition unit that acquires red light intensity (hereinafter referred to as "R light intensity"), green light intensity (hereinafter referred to as "G light intensity"), and blue light intensity (hereinafter referred to as "B light intensity") of each pixel from the image acquired by the image acquisition unit; a scatter diagram acquisition unit that acquires a two-dimensional scatter diagram with the R light intensity or G light intensity and the B light intensity of each pixel as two axes; a yellowing information acquisition unit that acquires yellowing information of a plant based on the two-dimensional scatter diagram; The yellowing information acquisition system includes: (Section 2-12) an image acquisition unit that acquires an image of a plant leaf; a light information acquisition unit that acquires the hue (hereinafter referred to as "H intensity"), saturation (hereinafter referred to as "S intensity"), and lightness (hereinafter referred to as "V intensity") of each pixel from the image acquired by the image acquisition unit; a scatter diagram acquisition unit for acquiring a two-dimensional scatter diagram with the H intensity or V intensity and the S intensity of each pixel as two axes; a yellowing information acquisition unit that acquires yellowing information of a plant based on the two-dimensional scatter diagram; The yellowing information acquisition system is provided with the above. (Section 2-13) The etiolation information acquisition system according to any one of Items 2-9 to 2-12, wherein the image obtained by the image acquisition unit is processed so that the veins of the plant are clear and uniform, and then processing is performed by the light information acquisition unit. (Section 2-14) The yellowing information acquisition system according to any one of items 2-9 to 2-13, wherein the scatter diagram acquisition unit acquires the two-dimensional scatter diagram in a state in which at least one of mesophyll parts, discolored parts, dead parts, vein parts, and perforated parts on the plant leaves is removed. (Section 2-15) The yellowing information acquisition system according to any one of items 2-9 to 2-14, wherein the yellowing information acquisition unit diagnoses a cause and a degree of progress of the yellowing based on a shape and feature values ​​of the two-dimensional scatter diagram. (Section 2-16) The yellowing information acquisition system according to any one of items 2-9 to 2-15, wherein the yellowing information acquisition unit acquires shape information on a real plane of each point corresponding to a specific cluster in the two-dimensional scatter diagram.

[0090] Although the embodiments and modes of implementation of the present invention have been described, the disclosed contents may vary in details of the configuration, and changes in the combination and order of elements in the embodiments and modes of implementation may be realized without departing from the scope and spirit of the claimed invention. [Explanation of symbols]

[0091] 10 Image acquisition section 20 Optical information acquisition section 30 Scatter Plots 40 Yellowing Information Acquisition Department 50 Display device 70 Mobile Devices 80 Control section 90 Yellowing information acquisition device 100 Yellowing Information Acquisition System

Claims

1. A method for diagnosing an object that changes color over time due to natural factors, including disease, and develops healthy and unhealthy areas, comprising: An imaging step of acquiring a photographed image of the object; a wavelength information acquiring step of acquiring wavelength information for each pixel from the captured image; a two-dimensional scatter diagram acquisition step of acquiring a two-dimensional scatter diagram obtained by viewing a three-dimensional scatter diagram in which each pixel is plotted in a color space based on the acquired wavelength information from a direction in which a difference between a healthy portion and an unhealthy portion is significantly shown; a diagnosis step of diagnosing a cause of the discoloration phenomenon of the object and a state of the object from the acquired two-dimensional scatter diagram; A method for diagnosing an object, comprising:

2. The diagnostic method according to claim 1, characterized in that the discoloration phenomenon caused by a natural phenomenon has at least one or more stages and one or more characteristic sites, and when a specific characteristic site is at a specific discoloration stage, the color before and after the discoloration is constant, and the direction of the color change is the same.

3. The diagnostic method according to claim 1 , wherein in the color information acquisition step, only pixel data corresponding to a specific characteristic portion and a specific discoloration stage is extracted.

4. The diagnostic method of claim 1, wherein when a certain characteristic portion of the object is at a certain stage of discoloration, the three-dimensional scatter diagram formed in color space by the corresponding group of dots is distributed approximately in a sheet-like shape, and the two-dimensional scatter diagram is an image viewed from a direction approximately parallel to the approximate plane of the group of dots distributed approximately in a sheet-like shape.

5. The diagnostic method according to claim 1 , further comprising a step of obtaining a regression line or regression curve when the three-dimensional scatter plot is viewed from a direction in which it is contracted to a specific line or curve.

6. The diagnostic method according to claim 1 , further comprising the step of: forming an image in which information on healthy and unhealthy parts is added to the captured image based on position information in the captured image corresponding to each dot in the two-dimensional scatter plot.

7. The diagnostic method according to claim 1, further comprising diagnosing the cause of the discoloration of the object and the state of the object from the shape of the healthy or unhealthy parts, or both, shown in an image to which information on the healthy and unhealthy parts has been added.

8. The diagnostic method according to claim 1 , wherein the color space is a three-dimensional space with RGB or HSV as three axes.

9. 2. The diagnostic method according to claim 1, wherein the two-dimensional scatter plot obtained using the object whose condition has been grasped by another means is used as a standard two-dimensional scatter plot, and in the diagnostic process, the two-dimensional scatter plot obtained by imaging the object is compared with the standard two-dimensional scatter plot to diagnose the cause of the discoloration phenomenon of the object and the condition of the object.

10. 2. The diagnostic method according to claim 1, wherein the regression line or regression curve obtained by using the object, the state of which has been grasped by another means, is used as a standard regression line, and in the diagnostic step, the regression line obtained by imaging the object is compared with the standard regression line to diagnose a cause of the discoloration phenomenon of the object and the state of the object.

11. The diagnostic method according to claim 1 , further comprising a function of improving the flatness of the approximate sheet distribution by applying a certain operation to all plots so as to convert the regression curve into a straight line.

12. The diagnostic method according to claim 1 , further comprising obtaining a time series change in color of the same object as a change in shape of the two-dimensional scatter plot.

13. 2. The method of claim 1, further comprising the step of comparing the shape of the two-dimensional scatter plot or the regression line or both for a plurality of the stages or the characteristic sites or both, respectively.

14. The diagnostic method according to any one of claims 1 to 13, wherein the target object is the surface of a leaf, shoot, flower, bud, stem, branch, trunk, or fruit of a plant that is discolored due to natural factors including disease.

15. The diagnostic method according to claim 1 , wherein the target object is the surface of skin, mucous membrane, or teeth of an animal or human that is discoloured due to natural factors including disease.

16. The diagnostic method according to claim 1 , wherein the object is a member having a metal surface on which rust occurs.

17. A system for diagnosing an object that changes color over time due to natural factors, including disease, and develops healthy and unhealthy parts, An imaging unit for acquiring a photographed image of the object; a wavelength information acquisition unit that acquires wavelength information for each pixel from the captured image; a two-dimensional scatter diagram acquisition unit that acquires a two-dimensional scatter diagram obtained by viewing a three-dimensional scatter diagram in which each pixel is plotted in a color space based on the acquired wavelength information from a direction in which a difference between a healthy portion and an unhealthy portion is significantly shown; a diagnosis unit that diagnoses a cause of the discoloration phenomenon of the object and a state of the object from the acquired two-dimensional scatter diagram; A diagnostic system for an object comprising:

Images