JP2004151092A - Method of monitoring vegetation - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To obtain locations of abnormal parts such as cracks within a natural slope quantitatively and accurately, and to detect the abnormal parts such as the cracks safely, promptly, and with high precision. <P>SOLUTION: A method of monitoring vegetation takes an infrared thermal image of vegetation of the natural slope, quantitatively obtains the surface temperature behavior of the vegetation, and extracts temperature increasing points. The method detects the spectral reflectance characteristics of the vegetation, quantitatively obtains the activity of the vegetation, and extracts activity decreasing points. The temperature increasing points and the activity decreasing points are synthesized with a visible digital image to detect the abnormal parts such as cracks within the natural slope. <P>COPYRIGHT: (C)2004,JPO

Description

  The present invention relates to a vegetation monitoring method, for example, when constructing a road, a railway, or a tunnel, for example, safely and quickly detecting an abnormal portion such as a crack generated on a natural slope such as a cut-through portion of the ground, or monitoring vegetation health. Related to technology.

  Conventionally, inspections of natural slopes have been conducted mainly by investigators recording cracks and protrusions or visually confirming the soundness of vegetation. It is inevitable that the results of the judgment, which involve the subjectivity, are inevitable, and the inspection itself is dangerous.

  In particular, there has been an urgent need to develop an inspection method that is excellent in safety, speed, and reliability with respect to a natural slope of a surrounding mountain cut generated when a road, a railway, a tunnel, or the like is constructed.

  In addition, as a nondestructive inspection method that can detect abnormal parts on the mortar spraying slope safely and at high speed, there is a method of measuring the surface temperature of mortar using an infrared imaging device and discriminating abnormal parts and healthy parts from the temperature distribution. Attention has been paid, and some of them have already been put to practical use (for example, Patent Documents 1 and 2).

  On the other hand, as for health monitoring of vegetation, for example, a technique of photographing a lawn ground from above by thermography for lawn management has been proposed (Patent Document 3).

JP-A-10-10064 (see FIG. 1 and the like) JP-A-5-264489 (see FIG. 2 and the like) JP-A-10-48054 (see P5, FIG. 1, FIG. 6, etc.)

  However, with respect to natural slopes as in Patent Literatures 1 and 2, it is not possible to accurately detect an abnormal portion such as a crack simply by measuring the surface temperature of the slope with infrared rays.

Further, the conventional visual inspection has the following problems.
(B) Visual inspection alone cannot detect an abnormal portion such as a crack inside a natural slope.
(B) Subjectivity of the investigator intervenes, and the judgment results vary.
(C) It is not possible to grasp the activity of vegetation on a natural slope quantitatively by visual observation.

  Further, Patent Literature 3 relating to vegetation health monitoring obtains the underground temperature of the ground surface layer region of a lawn ground, and is not based on the surface temperature behavior of vegetation and does not detect the activity of vegetation.

  The present invention has been constructed as an optimal method for inspection of natural slopes and the like, and solves the above-mentioned conventional problems, quantitatively and accurately grasps the position of an abnormal portion on a slope, and generates the information on a natural slope. An object of the present invention is to detect abnormal portions such as cracks safely, promptly and with high accuracy as compared with conventional methods.

  Further, the present invention is constructed as an optimal method for vegetation health monitoring, and provides a monitoring method capable of detecting vegetation health with high accuracy and realizing quick and efficient maintenance of vegetation. The purpose is to:

  The present invention uses infrared thermal imaging of vegetation to quantitatively grasp the surface temperature behavior of vegetation and extract points of temperature rise, while detecting the spectral reflection characteristics of the vegetation and quantitatively grasping the vegetation activity. Vegetation that can extract areas with reduced activity, combine these temperature-raised and activity-decreased areas with visible digital images, detect abnormalities such as cracks inside natural slopes, and perform vegetation health monitoring Monitoring method.

  That is, first, an infrared thermal image of the vegetation is taken, the surface temperature behavior of the vegetation is quantitatively grasped, and a temperature rising portion is extracted, and the spectral reflection characteristic of the vegetation is detected to quantitatively determine the activity of the vegetation. Extracting the activity-reducing portions, and displaying the temperature-rising portions and the activity-reducing portions on a visible digital image, and identifying the activity-reducing portions of the vegetation. It consists of vegetation monitoring methods.

  By specifying the location where the activity of the vegetation is reduced in this way, it is possible to predict a location where a natural slope is at risk of collapse that is not visible to the naked eye, or to specify a location where the health of the plant is low that is not visible to the naked eye.

  Second, a thermal camera capable of taking an infrared thermal image of vegetation, a spectral reflectance measuring device capable of measuring the spectral reflectance of the vegetation, and a thermal image data from the thermal camera and the spectral reflectance measuring device And a computer that specifies a location where the activity of the vegetation has decreased based on the spectral reflectance data of the vegetation, wherein the plurality of still image thermal image data of the vegetation captured by the thermal camera are stored in a computer. And storing the two image data of the image having the highest temperature location and the image having the minimum temperature location selected from the thermal image data stored in the storage means in the storage means. Step, the difference image creating means of the computer subtracts the two image data to create difference image data representing a temperature difference between the two image data, Storing the difference image in the storage means and displaying the difference image on a display means; and, when a location having a large temperature difference is specified as a temperature rise location on the difference image displayed on the display means, the temperature rise location Storing the position information in the storage means, and the temperature-time characteristic diagram creation means of the computer creates a temperature-time characteristic diagram of the temperature rise portion in the thermal image data stored in the storage means, and A step of displaying a characteristic diagram on the display means; and, if a temperature rise point having the largest temperature change is identified from the temperature rise points based on the characteristic chart displayed on the display means, the point is set to a first position. Storing the first dangerous spot on the difference image as a dangerous spot of the above; and displaying the first dangerous spot on the difference image. Measuring the spectral reflectance at the raw temperature rising point for each of a plurality of wavelengths, and storing the spectral reflectance data at the temperature rising point in the storage unit; Calculating a vegetation index at the temperature rising point based on the spectral reflectance data; and setting a point at which the activity is the lowest in the temperature rising point based on the calculated vegetation index as a second dangerous point. The vegetation monitoring method includes a step of storing in the storage means and a step of displaying the second dangerous spot on the difference image.

  The first and second danger spots can be predicted as collapse danger spots when, for example, vegetation on a natural slope is photographed. For example, when vegetation such as lawn is photographed, the maintenance of deteriorated soundness is performed. It can be specified as a necessary part. The computer may be configured so that after measuring the spectral reflectance, a spectral reflectance-wavelength characteristic diagram is created for the temperature rising point and displayed on a display unit.

  Third, the image combining means of the computer has a step of displaying the positions of the first dangerous spot and the second dangerous spot on the digital still image of the vegetation. It consists of the vegetation monitoring method described.

  Fourth, the vegetation index is RVI, and the computer selects a point having the lowest RVI value as the second dangerous point according to the vegetation monitoring method according to the second or third aspect. It is composed.

  The RVI is “A ÷ B”, and the arithmetic means executes the arithmetic. Here, A is a spectral reflectance [%] in a near infrared region, for example, a wavelength of 0.7 [μm] to 1.15 [μm], and B is a spectral reflectance in a red region, for example, a wavelength of 0.65 [μm]. Rate [%]. Note that the NDVI = “(AB) ÷ (A + B)” may be used as the vegetation index.

  Fifth, the thermal camera captures the vegetation on a natural slope and the spectral reflectance measuring device measures the spectral reflectance of the vegetation on the natural slope at the temperature rise point, thereby obtaining the first and second vegetation. The method is configured by a natural slope collapse risk prediction method using the vegetation monitoring method according to any one of the second to fourth aspects, wherein the dangerous place is predicted as a natural slope collapse risk place.

  Sixthly, the vegetation health using the vegetation monitoring method according to any one of the second to fourth aspects, wherein the first and second dangerous places are predicted as the places where the health of the vegetation deteriorates. It consists of a monitoring method.

  According to the present invention, for example, an abnormal portion such as a crack generated on a natural slope that cannot be visually confirmed, an abnormal portion such as a low soundness portion generated on a vegetation such as lawn, etc. Can be specified by a decrease in the activity based on the spectral reflection characteristics of the light-emitting element, so that safe, quick and accurate detection can be performed as compared with the conventional method. Therefore, the present invention greatly contributes to society as a method for predicting the occurrence of a landslide or the like in advance, preventing a disaster beforehand, and maintaining the soundness of vegetation such as lawns.

Hereinafter, the case where the vegetation monitoring method according to the present invention is applied to the prediction of collapse danger of a natural slope will be mainly described.
(Overview of the natural slope failure risk prediction system)
When the natural slope collapses, cracks 2 are first formed in the upper layer of the soil inside the slope 1, but it is almost impossible to visually confirm that the cracks 2 are still fine, When the number 2 becomes large enough to be visually confirmed, it is often the case that the slope collapse has already occurred.

  On the natural slope, the vegetation 3 covers the surface, and the vegetation 3 has a root on the upper part of the slope 1 like a net, and when the crack 2 occurs in the upper part of the earth body, the vegetation as shown in FIG. The roots 4 of 3 are cut off, which affects the growth of the vegetation 3. When the growth function is inhibited, the activity of the vegetation 3 decreases, that is, the vegetation 3 deteriorates in health state and eventually dies. If the state of the vegetation 3 can be grasped, it is possible to estimate a crack generation location on the slope 1. That is, a crack 2 is generated on the slope 1, and the occurrence of the crack 2 is indirectly captured via the vegetation 3 before extending sufficiently.

  FIG. 2 shows a flowchart of the present prediction system. When a crack is formed inside the natural slope, the roots of the vegetation are cut, and growth functions such as transpiration are inhibited, so that the temperature of the vegetation increases and the activity decreases. This increase in temperature and decrease in activity are invisible information that cannot be confirmed by the human eye, and by observing the target slope with a thermal camera and a spectrum photometer, the results are imaged and graphed. Converts the information into visible information that can be confirmed by the human eye, and predicts the danger of collapse on natural slopes.

(analysis method)
(Identification of temperature rise)
The rise in temperature is caused by a reduced transpiration, which does not remove heat of vaporization from the surrounding air. FIG. 3 is a thermal camera image observing the temperature condition of the sugarcane in which the growth function was inhibited, and it can be seen that the temperature is higher than other healthy ones (see area E). As described above, if there is a place where the temperature is higher than the surrounding vegetation, the vegetation at that place is likely to have a root cut, and a crack is likely to have occurred in the upper layer of the soil body.

  This temperature rise is found by observing it with a thermal camera. At this time, since the temperature of the vegetation differs depending on the type, it is necessary to observe changes over time. The method involves observing the target natural slope with a thermal camera over time and acquiring a thermal infrared image. Next, a difference image of the acquired thermal infrared image is created, and the temperature change of the vegetation is monitored.

  This temperature change is confirmed by an image as shown in FIG. 3, and an arbitrary point is set within the target range, and the temperature change at that point is represented by a temperature change graph in which the vertical axis represents temperature and the horizontal axis represents time. Confirm by creating. As a result of these two types of analysis, a location where the temperature has risen significantly is regarded as a danger of slope failure. In addition, on the target slope, a photograph is taken, and the photograph image is superimposed on the thermal infrared image to specify a detailed occurrence danger point.

(Identification of places where activity drops)
Using the reflection characteristics of the plant, the point where the tree is in a "weakened" state shortly after the crack has occurred is identified before the tree dies due to the effect of root breakage due to the crack.

  The reflection characteristic of the plant has certain characteristics as shown in the schematic diagram showing the activity (health degree) and the reflection characteristic of the plant shown in FIG. The horizontal axis of this figure is the wavelength of electromagnetic waves reflected by plants (0.4 to 0.7 microns: visible light region, 0.7 to 1.15 microns: near infrared region), and the vertical axis is the intensity of solar energy reflected by plants. . In the figure, there are expressions such as "healthy", "weak", and "withered" in plants, but these indicate the state of health in humans, and "healthy" is completely healthy, "weak" Is a diseased state, and "withered" is a state of loss of life. Vegetation that is in a "weak" state refers to vegetation that appears healthy to the naked eye but is actually diseased, and "dead" refers to vegetation that has died.

  As can be seen from the figure, healthy plants have strong reflection in the green band (around 0.55 microns) in the visible light range, and higher reflections in the near infrared region. Healthy plants appear green to the human eye because of the high reflection in the green band. In a dead plant, the reflection is almost the same between the green band and the red band (around 0.65 micron) in the visible light region, and at the same time, the reflection in the near infrared region is significantly reduced. The reflection of a weakened plant in the intermediate state is the same as that of a healthy plant in the visible light region, but is lower than that of a healthy plant in the near infrared region.

Each reflection characteristic can be numerically represented by a vegetation index (Vegetation Index) calculated from the reflection intensity in each wavelength region. Although there are six types of vegetation index, representative RVI (Ratioed Vegetation Index) and NDVI (Normarized Differential Vegetation Index) are shown below.
RVI = A ÷ B, NDVI = (A−B) ÷ (A + B)
Here, A: the reflection intensity in the near infrared region, and B: the reflection intensity in the red band (visible light region).

  From FIG. 4, when the RVI of each state is obtained, healthy plants = 60/10 = 6, weak plants = 40/10 = 4, and dead plants = 20 ＝ 20 = 1. By using the vegetation index in this manner, the degree of health of the plant is represented by a numerical value, and the larger the vegetation index, the closer to a healthy state.

  This vegetation index is observed over time on the target slope using a spectral photometer, and if a vegetation index is significantly reduced, that is, if a location where the activity is reduced can be identified, a crack is generated inside the soil body at the location, and the slope is generated. It can be presumed to be a danger point of collapse.

(Identification of danger points on natural slopes)
As described above, using a thermal camera and a spectral photometer, the temperature distribution and the degree of activity of the vegetation on the target slope are examined, and the places where changes are observed in both the temperature rise and the activity decrease are finally determined. Identify collapse risk points.

  As a result, it becomes possible to quantitatively specify the previously vague collapse-risk point quantitatively faster than the existing method.

(Example 1)
FIG. 5 is a block diagram of an apparatus for implementing the method according to the present invention. In the figure, reference numeral 5 denotes a thermal camera, which shoots an object and displays infrared thermal image data of a digital color still image for displaying the temperature of the object and temperature data indicating the temperature of each pixel at predetermined time intervals (for example, every one minute). The data is output to a personal computer 7 (hereinafter, referred to as “computer 7”). The thermal image data output from the thermal camera 5 indicates RGB data (A ₁ , A ₂ , A ₃ ...) For each pixel as shown in FIG. 6 and the temperature corresponding to the RGB data. Temperature data (B ₁ , B ₂ , B ₃ ...). The color still image, as shown in FIG. 3, displays the temperature of the subject in different colors, and when displayed on the display unit 13 of the personal computer 7, the temperature and color contrast are displayed on the right side of the screen. It has a part 8. The color for displaying the temperature is such that the high temperature part is displayed in red, and as the temperature decreases, it is displayed in yellow, green, blue, purple, and black. What you get. As such a thermal camera 5, for example, a thermal camera TH3102MR manufactured by NEC Corporation can be used.

  Reference numeral 6 denotes a spectrum photometer (spectral reflectance measuring device), which is configured to switch the optical filter by switching a changeover switch (not shown) to switch the wavelength of the electromagnetic wave to be measured. Therefore, by switching the changeover switch, the spectral reflectance at the point to be measured can be measured for each of various wavelengths from about 0.4 [μm] to about 1 [μm] of the electromagnetic wave. That is, by determining the measurement wavelength and performing the measurement, the spectral reflectance data of the measured point at the wavelength is output to the computer 7. As such a spectrum photometer 6, for example, a 2703MMver0.2 type portable photometer designed by Abe Corporation can be used.

  Reference numeral 7 denotes a computer, an I / O interface 9 for controlling input / output with the external devices 5, 6, etc., a RAM storing a main program (shown in FIGS. 14 to 20) according to the present invention, and the main program Control means 10 comprising a CPU for executing various processes based on the above, thermal image data S from the thermal camera 5, spectral reflectance data 11d from the spectral photometer 6, and other operations of the control means 10. A storage means 11 for storing various data (two image data 11a, difference image data S ', temperature rise data 11b, danger data 11c and 11f, vegetation index data 11e, digital image data 11g, etc.) An input unit 12 for giving various instructions to the unit 10, a display for displaying various images, characters, and the like Stage 13, various data, and an output means 14 for outputting the image. The input means 12 is constituted by an input device such as a keyboard and a mouse, the display means 13 is constituted by a display having a screen such as a liquid crystal or a CRT, and the output means 14 is constituted by a printer. It is assumed that The storage means 11 is configured by an external storage device such as a hard disk.

  The control means 10 realizes various functions based on the main program stored in the RAM. As shown in FIG. 5, the control means 10 functionally creates a difference image between two selected images described later. A difference image creating unit 10a, a temperature-time characteristic diagram creating unit 10b for creating a temperature-time characteristic diagram, a spectral reflectance-wavelength characteristic diagram creating unit 10c, and an RVI or other arithmetic unit 10d for calculating a vegetation index (RVI or the like); An image synthesizing unit 10e for synthesizing a digital image of a natural slope and a difference image is provided.

(1) Identification of Dangerous Location Using Thermal Camera Next, the measuring method of the present invention will be described. First, a vegetation on a natural slope shown in FIG. In this case, for example, the control means 10 of the computer 7 outputs the thermal image data S1, S2, S3... (RGB data and temperature data) of the still images from the thermal camera 5 every time t1 (for example, one minute) to the I / O interface 9. (FIGS. 14P1, P2) and sequentially stores these thermal image data S1, S2, S3,... As thermal image data S in the storage means 11 (see FIGS. 14P3, P4, FIG. 7). .

  Next, the operator operates the input means 12 to display the thermal image data S in the storage means 11 on the display means 13 as shown in FIG. 3 (P1, P2 in FIG. 15), and displays the thermal image data on the screen. While viewing, the image having the maximum temperature and the image having the minimum temperature are selected from the thermal image data S1, S2, S3,... (P3 in FIG. 15). Here, for the sake of simplicity, it is assumed that the thermal image has three areas a, b, and c as shown in FIG. This selection work focuses on an image having a high-temperature portion area displayed in the darkest red of the thermal images (here, thermal image data S9 having a high-temperature portion area a (see FIG. 7)). Based on the thermal image data S9, thermal image data (here, thermal image data S3 (refer to FIG. 7)) where the lowest temperature exists in the high temperature area a is selected. Then, the control means 10 of the computer 7 stores these two thermal image data S3 and S9 in the storage means 11 and displays them on the display means 13 (see P4 in FIG. 15, FIG. 5, and image data 11a in FIG. 15).

  Next, the operator gives a command to create the difference image S 'from the input means 12 to the computer 7 (P5 in FIG. 15). Then, the difference image creation unit 10a of the computer 7 extracts the thermal image data S3 and S9 based on the two image data 11a in the storage unit 11, and extracts the thermal image data S3 and the thermal image data S9 for each pixel. The difference between the temperature data is calculated to create difference image data S ′ (see FIG. 8), and the difference image data S ′ is stored in the storage unit 11 and displayed on the display unit 13 (P6 in FIG. 15). At this time, as shown in FIG. 8B, the difference image creating means 10a subtracts the temperature data for each pixel of the thermal image data S9 and S3 stored in the storage means 11 for each pixel. This is executed to obtain difference image data S ′. In FIG. 8B, the temperature data of each pixel in each of the areas a, b, and c is "40", "20", and "10" in the thermal image data S9, and "20", "15", and "10" in the thermal image data S3. 5 ", which indicates that these have been subtracted, resulting in" 20 "," 5 ", and" 5 ", respectively, in the difference image data S '. Accordingly, on the display means 13, only the area a has the highest temperature (red) as shown in the difference image S 'in FIG. 7A, whereby the temperature change is the largest in the area a. It can be seen that the soundness of the vegetation has decreased, and there is a high possibility that cracks and the like have occurred on the slope. In FIG. 8B, the temperature data is uniform in each of the areas a, b, and c of the difference image S ', but actually, there is a level of the temperature (shading) in each area.

Thereafter, in the difference image S ′ displayed on the screen, in the area a where the temperature difference is large, a plurality of points (points) with a large temperature difference, particularly displayed in dark red, are set as the temperature rising points. Specify (P7 in FIG. 15). In this case, in this embodiment, as shown in FIG. 9, it is assumed that three points Q1, Q2, and Q3 are designated in the area a. Then, the control unit 10 xy coordinate position _Q1 of the point specified _{(x 1, y} 1) of the computer _{7, Q2 (x 2, y} 2), Q3 (x 3, y 3) the position data of the temperature rising part 11b is stored in the storage means 11 (P8 in FIG. 15).

Next, the temperature-time characteristic diagram creating means 10b of the computer 7 refers to the thermal image data S (S1, S2, S3...) Of the storage means 11, and reads the position data Q1, Q2 of the temperature rise location data 11b. the thermal image data S1 corresponding to Q3, S2, S3 · · · position _{Q1 (x 1, y 1)} , Q2 (x 2, y 2), Q3 (x 3, y 3) the temperature data in It reads out from the storage means 11 (see FIG. 7), creates a temperature-time characteristic diagram G in which the horizontal axis indicates elapsed time and the vertical axis indicates temperature, and displays the characteristic diagram G on the display means 13 as shown in FIG. (FIGS. 15P9, P10).

  After that, the operator looks at the characteristic diagram G displayed on the screen, and uses the input unit 12 to input a portion having the largest inclination, that is, a portion having the largest temperature change (in this embodiment, the portion having the largest temperature change Q1 Is specified (FIG. 16P1). Then, the control means 10 stores the position data (x1, y1) of the temperature rise point Q1 in the storage means 11 as danger point data 11c (first danger point) (FIG. 16P2), and the difference image data S 'Is extracted from the storage means 11 and displayed again on the display means 13 (FIG. 16P3), and the above-mentioned danger spot Q1 (x1, y1) is distinguished from other temperature rise spots Q2, Q3 on the difference image S'. In order to obtain it, for example, it is displayed by color coding, a circle, or another method (see FIG. 16, P4 and FIG. 9).

  With the above processing, the process of specifying the dangerous spot Q1 (first dangerous spot) by the thermal camera 5 ends. At this time, the temperature change over time of the dangerous place Q1 (x1, y1) can be grasped in the temperature-time characteristic diagram G, so that the position, temperature, and the like of the dangerous place can be grasped quantitatively. In the above-described step P7 of FIG. 15, in this embodiment, the temperature rise point is designated by a point (point). However, an area can be designated by designating not only the point but also a predetermined range. In this case, the characteristic diagram G of FIG. 10 may be created by using the average value of the temperature data of the pixels in the range designated by the area as the temperature data of the area.

  Note that, for example, when a lawn or the like requiring health monitoring is photographed by the thermal camera 5, the dangerous place Q1 can be grasped as a place where the soundness of the lawn is reduced.

(2) Identification of Dangerous Location Using Spectrum Photometer Next, the operator measures the spectral reflection characteristics with the spectrum photometer 6 for the temperature rise locations Q1, Q2, and Q3 specified in the process (1) on the natural slope. Do. Here, as for the wavelength of the electromagnetic wave, for example, the spectral reflectance for a plurality of wavelengths in the range of 0.4 [μm] to 1 [μm] is measured. Then, the spectral reflectance data for each wavelength is input to the computer 7 (FIG. 17P1), and the computer 7 stores the spectral reflectance data for each wavelength in the storage unit 11 (FIG. 17P2, the spectral reflection data of the storage unit 11). 11d). When the measurement is completed for all the temperature rise points Q1, Q2, and Q3, the process is terminated (P3 in FIG. 17). The spectral reflectance data C ₁ , C ₂ ..., D ₁ , D ₂ ..., E ₁ , E ₂ . To (c).

  Next, when the operator issues a command to create a wavelength-spectral reflectance characteristic diagram from the input unit 12 (P1 in FIG. 18), the spectral reflectance-wavelength characteristic diagram creating unit 10c of the computer 7 reads the spectral reflectance data from the storage unit 11. 11d is extracted (FIG. 18P2), and based on the wavelength data and the spectral reflectance data, a spectral reflectance-wavelength characteristic diagram H (see FIG. 12) of each of the temperature rising points Q1, Q2, Q3 is created (FIG. 18P3). Then, the characteristic diagram H is displayed on the display means 13 (P4 in FIG. 18). It should be noted that the computer 7 may be configured to automatically create the spectral reflectance-wavelength characteristic diagram H when the storage of the spectral reflectance data at all points is completed in step P3 in FIG. .

  Next, the operator issues an RVI (vegetation index) calculation command to the computer 7 (P5 in FIG. 18). Then, the calculation means 10d of the computer 7 calculates RVI (= A / B) for each of the points Q1 to Q3, and stores the calculated values as vegetation index data 11e in the storage means 11 (FIGS. 18P6 and P7). It is assumed that the calculation result is as follows.

RVI ≒ 1 of Q1 (x1, y1)
RVI of Q2 (x2, y2) ≒ 5
RVI of Q3 (x3, y3) ≒ 4
A: near infrared region, spectral reflectance [%] near 0.75 μm wavelength
B: Spectral reflectance [%] in red region, wavelength around 0.65 μm

  Subsequently, the control means 10 of the computer 7 selects the temperature rising point having the lowest RVI value among the calculated RVI values, that is, the point where the vegetation activity is the lowest, in this case, the temperature rising point Q1. Then, the location is stored in the storage unit 11 as dangerous location data 11f (second dangerous location) (P8 in FIG. 19). After that, the control means 10 calls up the difference image data S 'from the storage means 11 and displays it on the display means 13 (FIG. 19P9), and distinguishes the dangerous spot Q1 from the other points Q2 and Q3 (for example, 19 (P10 in FIG. 19, danger spot Q1 in FIG. 13).

  Through the above processing, it is determined that the first dangerous spot Q1 specified by the thermal camera and the second dangerous spot Q1 specified by the spectral photometer to be a dangerous spot are positions having a high risk of collapse. Then, it can be displayed on the display means 13. At this time, the danger point Q1 (x1, y1) can quantitatively grasp the decrease in the activity of the vegetation based on the RVI value. In the above embodiment, the case where the two dangerous places are the same is shown. However, when the two dangerous places are different, each of the dangerous places can be displayed in the difference image S 'on the screen. In the above embodiment, RVI is used as the vegetation index, but NDVI can of course be used.

  When a lawn or the like is photographed with the spectrum photometer, the second dangerous spot Q1 can be grasped as a place where the soundness of the lawn or the like is deteriorated.

  Thereafter, when the operator gives an image synthesizing command to the computer 7 from the input means 12 (P1 in FIG. 20), the image synthesizing means 10e of the control means 10 captures images in advance with the digital camera 15 and stores the natural slopes in the storage means 11. The digital still image data 11g is extracted (P2 in FIG. 20), the digital image 11g is combined with the difference image data S ′ (P3, P4 in FIG. 20), and the danger spot Q1 is displayed on the digital image on the natural slope. The digital still image is printed out by the output unit 14 based on the output command of the operator (P5 in FIG. 20) (P6 in FIG. 20). That is, the positions of the first dangerous place and the second dangerous place can be displayed on the digital still image. FIG. 21 shows an example of the composite image.

  As described above, the present invention provides (a) temperature measurement of vegetation by infrared measurement technology (thermal camera), (b) measurement of spectral reflection characteristics of vegetation by infrared measurement technology (spectral photometer), (C) Utilizing image processing technologies such as combining visible digital still photography and thermal images, abnormal parts such as cracks on natural slopes can be detected more safely, quickly and accurately than conventional methods. can do. Therefore, the present invention greatly contributes to society as a method of predicting the occurrence of a landslide or the like in advance and preventing a disaster before it occurs.

(Implementation bell 2)
In the first embodiment, the example in which the vegetation monitoring method according to the present invention is applied to the prediction of the danger of collapse of a natural slope is described. However, the method can also be applied to vegetation health monitoring. That is, in the above (1), in identifying a dangerous place by the thermal camera, vegetation such as lawn is photographed by the thermal camera, and the temperature rising points Q1, Q2, and Q3 are specified as the places of the lawn with reduced soundness. A position where the temperature change is large, for example, Q1 can be specified as a first dangerous place, that is, a place where the soundness is significantly reduced.

  Next, in the above-mentioned (2) specification of the dangerous place by the spectral photometer, the spectral reflectance of the vegetation at the above-mentioned temperature rising places Q1, Q2, and Q3 was measured, and the activity decreased most by the calculation of RVI (or NDVI). The location Q1 can be specified as a second dangerous location. Thereafter, by displaying the first and second dangerous spots Q1 on the difference image, it is possible to specify a location requiring maintenance in a vegetation such as a lawn.

  Thus, by applying the method of the present invention to vegetation health monitoring, it is possible to grasp in advance the activity status of weak vegetation that is not visible to the naked eye, and to perform appropriate maintenance before it is too late. You can do it.

It is a figure which shows the state of the root cut by the crack of a slope. 5 is a flowchart showing an operation procedure of the system according to the present invention. FIG. 3 is a diagram illustrating an infrared image captured by a thermal camera. It is a characteristic view showing the spectral reflection characteristic of a plant. FIG. 2 is a block diagram of an apparatus for performing the method according to the present invention. It is a figure which shows a digital color still image and its thermal image data. It is a figure which shows an infrared thermal image. (A) and (b) are diagrams showing an infrared image showing a procedure for creating a difference image. It is a figure of the difference image which shows the dangerous spot specified by the thermal camera. It is a characteristic diagram of temperature-time. (A)-(c) is a figure which shows the spectral reflectance data in the temperature rising part. It is a characteristic diagram of a spectral reflectance-wavelength. It is a figure of the difference image which shows the dangerous place specified by using a photometer. 4 is a flowchart showing an operation procedure of the method according to the present invention. 4 is a flowchart showing an operation procedure of the method according to the present invention. 4 is a flowchart showing an operation procedure of the method according to the present invention. 4 is a flowchart showing an operation procedure of the method according to the present invention. 4 is a flowchart showing an operation procedure of the method according to the present invention. 4 is a flowchart showing an operation procedure of the method according to the present invention. 4 is a flowchart showing an operation procedure of the method according to the present invention. It shows an image of a natural slope after the combination.

Explanation of reference numerals

DESCRIPTION OF SYMBOLS 1 Slope 2 Crack 3 Vegetation 4 Root 5 Thermal camera 6 Spectral photometer 7 Computer 10 Control means 10 a Difference image creation means 10 b Temperature-time characteristic diagram creation means 10 d Operation means 10 e Image synthesis means 11 Storage means 11 a 2 Image data 11 d Spectral reflection Rate data S Thermal image data S1, S2,... Thermal image data S 'Difference image data Q1, Q2, Q3 Temperature rise location

Claims (6)

Infrared thermal imaging of the vegetation, quantitatively grasping the surface temperature behavior of the vegetation, and extracting the temperature rise point,
Detecting the spectral reflectance characteristics of the vegetation, quantitatively grasping the vegetation activity, and extracting a portion where the activity is reduced,
Displaying these temperature rise points and activity drop points on a visible digital image, and identifying the activity drop points of the vegetation,
A method for monitoring vegetation, comprising: A thermal camera capable of taking an infrared thermal image of a vegetation, a spectral reflectance measuring device capable of measuring a spectral reflectance of the vegetation, a thermal image data from the thermal camera and a spectral reflectance from the spectral reflectance measuring device A method for monitoring vegetation using a computer that identifies a location where the activity of the vegetation has decreased based on the data,
Storing a plurality of still image thermal image data of the vegetation taken by the thermal camera in a storage means of a computer,
Storing two image data of the image having the highest temperature point and the image having the minimum temperature point selected from the thermal image data stored in the storage means in the storage means;
A difference image creating unit of the computer subtracting the two image data to create difference image data representing a temperature difference between the two image data, storing the difference image in the storage unit, and displaying the difference image on a display unit; ,
When a location having a large temperature difference is specified as a temperature rise location on the difference image displayed on the display means, storing position information of the temperature rise location in the storage means;
A step of creating a temperature-time characteristic diagram of the temperature-rising portion in the thermal image data stored in the storage unit, and displaying the characteristic diagram on the display unit;
When a temperature rise point having the largest temperature change is specified from the temperature rise points based on the characteristic diagram displayed on the display means, the point is stored in the storage means as a first dangerous point. ,
Displaying the first dangerous spot on the difference image;
Measuring the spectral reflectance of the vegetation at the temperature rising point for each of a plurality of wavelengths by the spectral reflectance measuring device, and storing the spectral reflectance data of the temperature rising point in the storage unit;
Calculating a vegetation index at the temperature rising point based on the spectral reflectance data for each wavelength stored by the calculating means of the computer;
Storing, in the storage means, a location where the activity has decreased most in the temperature rise location based on the calculated vegetation index as a second dangerous location;
Displaying the second dangerous spot on the difference image;
Vegetation monitoring method consisting of:   3. The vegetation according to claim 2, wherein the image synthesizing means of the computer has a step of displaying the positions of the first dangerous place and the second dangerous place on the digital still image of the vegetation. Monitoring method.   4. The vegetation monitoring method according to claim 2, wherein the vegetation index is RVI, and the computer selects a location having the lowest RVI value as the second dangerous location.   The thermal camera captures the natural slope vegetation and the spectral reflectance measuring device measures the spectral reflectance of the natural slope vegetation at the temperature rising point, thereby allowing the first and second dangerous places to be naturally located. The method for predicting the risk of collapse of a natural slope using the method of monitoring vegetation according to any one of claims 2 to 4, wherein the method is predicted as a risk of collapse of the slope.   The vegetation health monitoring method using the vegetation monitoring method according to any one of claims 2 to 4, wherein the first and second dangerous places are predicted as the places where the vegetation soundness is reduced.