WO2023135634A1

WO2023135634A1 - Infrared measurement image analysis device and infrared measurement image analysis method

Info

Publication number: WO2023135634A1
Application number: PCT/JP2022/000518
Authority: WO
Inventors: 武志小畠
Original assignee: 株式会社赤外線高精度技術利用機構
Priority date: 2022-01-11
Filing date: 2022-01-11
Publication date: 2023-07-20

Abstract

The relationship between an abnormal condition of an interior layer of a structure and the temperature distributions of a surface and the interior layer is made clear. An infrared measurement image analysis device 10 according to the present invention is characterized by comprising: a program storage unit 11 that stores image analysis software for performing image analysis of an infrared measurement image that was measured by a measurement device 30; a measurement image data reading unit 12 that reads the infrared measurement image from a measurement memory unit 31 of the measurement device 30 or from a measurement image database 20; a structure interior condition calculation unit 13 that, on the basis of the image data read by the measurement image data reading unit 12 and using the image analysis software, calculates the temperature distributions of a surface and an interior layer of a structure 40 under investigation, and the location of an abnormal section 41; an analysis image writing unit 14 that writes, to an analysis image database 21, the calculated temperature distributions of the surface and the interior layer of the structure 40, and the calculated location of the abnormal section 41; and a display unit 15 that performs two-dimensional or three-dimensional display.

Description

Infrared measurement image analysis device and infrared measurement image analysis method

The present invention relates to an infrared measurement image analysis device and an infrared measurement image analysis method. More specifically, the present invention relates to an infrared measurement image analysis apparatus and an infrared measurement image analysis method capable of high-precision analysis.

Recently, the accuracy of temperature sensors (semiconductor elements) that can be used in infrared measurement devices has improved, and it has become possible to separate materials with a minimum temperature detection difference of 0.02°C. Therefore, the inventor developed a high-resolution image analysis method to make the most of the precision of this temperature sensor (see Patent Document 1). According to Patent Document 1, for an image of a structure measured by an ordinary infrared camera, a highly accurate radiation temperature distribution element coordinate diagram of the structure surface with a resolution of 0.013 ° C. (surface radiation temperature corresponding to the element coordinates of the sensor) A diagram showing the temperature distribution) is acquired. In addition, current temperature sensors are capable of distinguishing to six digits after the decimal point.

Japanese Patent No. 6192749

However, although Patent Document 1 discloses that it is possible to acquire a highly accurate radiation temperature distribution element coordinate diagram of the surface of a structure, the highly accurate radiation temperature distribution element coordinate diagram is still disclosed by anyone other than the inventor. person is not found. This is because many people cling to the idea that defects in the inner layer of a building cannot be found from mere measurement of the radiation temperature distribution. , it is difficult to find a model from which an analysis image can be obtained relatively easily. There was also a history of people trying to find defects in the inner layers of structures, but eventually giving up. In addition, in Patent Document 1, when abnormal conditions such as voids, water leaks, and cracks exist inside the structure, what kind of change appears in the temperature distribution of the above-mentioned high-precision structure surface and inner layer. , not touched. In addition, the technical content of the analysis is not disclosed by the inventor, because in business negotiations, it is often the case that the analysis result will be accepted.
In this application, image analysis that associates an abnormal state with a temperature distribution will be described. A model for finding defects in construct inner layers is then disclosed. It is expected that this will greatly contribute to disaster prevention in Japan, where natural disasters occur frequently and the social infrastructure is aging.

The present invention refers to an abnormal state of the surface and internal layers of a structure (a state in which abnormalities such as foreign matter, voids, and water leaks are detected in thermal conduction resistance by multiple elements of the sensor. By multiple elements of the sensor, detects color with three RGB elements for visible light, and also detects temperature with multiple elements for infrared light, and converts it to temperature using, for example, Wien's displacement law) and the temperature distribution of the surface and internal layers of the structure. It is an object of the present invention to provide an infrared measurement image analysis device and an infrared measurement image analysis method that enable associated image analysis.

The infrared measurement image analysis device 10 according to the first aspect of the present invention, for example, as shown in FIG.
a program storage unit 11 that stores image analysis software that performs image analysis on the infrared measurement image measured by the measurement device 30;
a measurement image data reading unit 12 that reads an infrared measurement image from the measurement storage unit 31 of the measurement device 30 or the measurement image database 20 with high-precision temperature resolution;
Based on the image data read by the measurement image data reading unit 12, the three-dimensional temperature distribution of the surface and internal layers of the structure 40 to be investigated and the three-dimensional position of the abnormal portion 41 are determined with high accuracy using image analysis software. a structure internal state calculation unit 13 that calculates with temperature resolution;
an analysis image writing unit 14 that writes the calculated three-dimensional temperature distribution of the surface and internal layers of the structure 40 and the three-dimensional position of the abnormal portion 41 into the analysis image database 21;
a display unit 15 for two-dimensionally or three-dimensionally displaying the calculated temperature distribution of the surface and internal layers of the structure 40 and the position of the abnormal portion 41;
The high precision temperature resolution is characterized by 0.001°C to 0.05°C.

Here, "infrared measurement image analysis device", "infrared measurement image", "imaging", and "shooting" are not used, and "measurement" is used because infrared rays are recorded in the sensor and output is CSV temperature data in the form
The measurement image database 20 and the analysis image database 21 may be built in physically different databases, or may be built in the same physical database. Further, the calculation of the temperature distribution of the surface and inner layer of the structure 40 and the calculation of the position of the abnormal portion 41 in the structure internal state calculation unit 13 may be performed integrally in relation to each other.
Further, the surface temperature distribution is, in detail, a radiation temperature distribution. The temperature distribution of the surface and the inner layer means the radiation temperature distribution of the surface and the temperature distribution of the inner layer, and the three-dimensional temperature distribution means the three-dimensional radiation temperature distribution of the surface and the three-dimensional temperature distribution of the inner layer.

In addition, it was mentioned above that a temperature resolution of 0.013°C was achieved. This means that analysis with a resolution of 0.013° C. or higher is possible, of which 0.013° C. to 0.05° C. is practical. In addition, in Patent Document 1, the resolution is improved by further dividing the temperature measurement values. With this method, a temperature resolution of 0.01°C can be achieved by dividing 0.02°C into two. is. Furthermore, since a temperature resolution of 0.000001° C. has been proposed for infrared rays in the nano-area of bioresearch, it can be said that a temperature resolution of 0.001° C. can be achieved with leeway in the image analysis of the present application.

With this configuration, it is possible to provide an infrared measurement image analysis device that enables image analysis that associates the abnormal state of the surface and inner layers of the structure with the temperature distribution of the surface and inner layers of the structure.

The infrared measurement image analysis device 10 according to the second aspect of the present invention, in the first aspect,
The image analysis software includes a program for improving the accuracy of the measurement image read by the measurement image data reading unit 12, and a color palette 17 for coloring the highly accurate image.
The structure internal state calculation unit 13 is characterized by performing calculations based on image data that has been highly accurate and colored with a color palette 17 .

Here, the program for improving the accuracy of the measurement image is for improving temperature accuracy and position accuracy (see Patent Document 1).
With this configuration, the presence and condition of an abnormal portion in the internal layer of the building can be expressed in an easy-to-understand manner by increasing the accuracy of the image and coloring it.

In the infrared measurement image analysis apparatus 10 according to the third aspect of the present invention, in the first or second aspect, the image analysis software creates a model in which the structure 40 to be investigated is a laminate of layers having different material properties. Image analysis is performed assuming that the abnormal portion 41 changes the heat conduction resistance of the layer.

With this configuration, the abnormal state of the surface and inner layers of the structure and the temperature distribution of the surface and inner layers can be easily understood.

In the third aspect of the infrared measurement image analysis apparatus 10 according to the fourth aspect of the present invention, the image analysis software determines the depth of the abnormal portion 41 in the structure 40 by inserting the abnormal portion 41. It is characterized in that it is determined based on the temperature change of the layer.

With this configuration, the depth of the abnormal portion of the internal layer of the structure can be obtained relatively easily.

In the infrared measurement image analysis method according to the fifth aspect of the present invention, for example, as shown in FIG.
an image analysis software storing step (S001) of pre-storing image analysis software for performing image analysis on an infrared measurement image measured by the measuring device 30 in the program storage unit 11;
a measurement image data reading step (S002) of reading the infrared measurement image from the measurement storage unit 31 of the measurement device 30 or the measurement image database 20 with high-precision temperature resolution;
Based on the image data read by the measurement image data reading unit 12, the three-dimensional temperature distribution of the surface and internal layers of the structure 40 to be investigated and the three-dimensional position of the abnormal portion 41 are determined with high accuracy using image analysis software. a structure internal layer state calculation step (S003) for calculation with temperature resolution;
an analysis image writing step (S004) of writing the calculated three-dimensional temperature distribution of the surface and internal layers of the structure 40 and the three-dimensional position of the abnormal portion 41 into the analysis image database 21;
a display step (S005) of displaying the calculated temperature distribution of the surface and inner layers of the structure 40 and the position of the abnormal portion 41 two-dimensionally or three-dimensionally;
The high precision temperature resolution is characterized by 0.001°C to 0.05°C.

Here, setting the high-precision temperature resolution to 0.001° C. to 0.05° C. is the same as described in the first aspect.
With this configuration, it is possible to provide an infrared measurement image analysis method that enables image analysis that associates the abnormal state of the surface and inner layers of the structure with the temperature distribution of the surface and inner layers of the structure.

The infrared measurement image analysis method according to the sixth aspect of the present invention, in the fifth aspect,
The image analysis software includes a program for increasing the accuracy of the image read by the measurement image data reading unit 12, and a color palette 17 for coloring the highly accurate image,
The structure internal state calculation unit 12 is characterized by performing calculations based on image data that has been highly accurate and colored with a color palette.

With this configuration, the existence and status of abnormal parts in the internal layers of the building can be expressed in an easy-to-understand manner by increasing the accuracy and coloring of the image.

In the infrared measurement image analysis method according to the seventh aspect of the present invention, in the fifth or sixth aspect,
The image analysis software uses a model in which the structure to be investigated is a laminate of layers with different material properties, and image analysis is performed assuming that the abnormal portion 41 changes the thermal conduction resistance of the layers.

In the seventh aspect of the infrared measurement image analysis apparatus 10 according to the eighth aspect of the present invention, the image analysis software determines the depth of the abnormal portion 41 in the structure 40 by inserting the abnormal portion 41. It is characterized in that it is determined based on the temperature change of the layer.

According to the present invention, it is possible to provide an infrared measurement image analysis device and an infrared image analysis method that enable image analysis that associates the abnormal state of the surface and inner layers of the structure with the temperature distribution of the surface and inner layers of the structure. can.

1 is a diagram illustrating a configuration example of an infrared measurement image analysis apparatus according to Example 1; FIG. FIG. 4 is a diagram showing a flow example of an infrared measurement image analysis method according to Example 1; This is an example of measurement images (steeply sloped concrete retaining wall: conventional example (top) and using a high-precision color palette (bottom)) measured by an infrared survey analysis diagnostic device. It is an example of a diagram showing the relationship between depth and temperature when there is no abnormal portion in the construct. It is an example of a diagram showing the relationship between depth and temperature when there is an abnormal portion in the construct. FIG. 4 is a diagram for explaining the region P(x, y) within the construct; This is an example of measurement images (concrete elevated road: visible light image (upper left), conventional example (lower left), using a high-precision color palette (right)) measured by an infrared survey analysis diagnostic device. Measurement images measured by an infrared survey analysis diagnostic device (Concrete tunnel inner wall: using a high-precision color palette (upper left), conventional example (lower left), using a high-precision color palette (enlarged view of the gap on the back side of the side wall, upper right), high-precision color palette Examples of use (lower right), (upper left and lower right have different color palettes, ie, display temperature ranges). This is an example of measurement images (concrete tunnel inner wall: using a high-precision color palette (left, leakage part), using a high-precision color palette (right, void on the back side of the side wall)) measured by an infrared inspection analysis diagnostic device. FIG. 4 is a diagram (three-dimensional image) showing an example of a measurement image; FIG. 10 is a diagram showing the relationship between depth and temperature when abnormal portions overlap in a region P(x, y) within a construct. FIG. 5 is an example showing the change in material properties of a layer when the material of construction is uniform;

BEST MODE FOR CARRYING OUT THE INVENTION Hereinafter, embodiments of the present invention will be described with reference to the drawings. In each figure, the same or similar members are denoted by the same or similar reference numerals, and redundant explanations are omitted.

FIG. 1 shows a configuration example of the infrared measurement image analysis device 10 according to the first embodiment. An infrared measurement image analysis apparatus 10 according to this embodiment includes a program storage unit 11 , a measurement image data reading unit 12 , a structure internal state calculation unit 13 , an analysis image writing unit 14 and a display unit 15 .

The program storage unit 11 stores in advance image analysis software for performing image analysis on infrared measurement image data measured by the measurement device 30 . The measurement device 30 acquires a measurement image of the structure 40 to be investigated, and the acquired infrared measurement image data is first stored in the measurement storage unit 31 of the measurement device 30 . The infrared measurement image data in the measurement storage unit 31 is transmitted to the measurement image database 20 by wired communication or wireless communication.

The measurement image data reading unit 12 reads the infrared measurement image data from the measurement storage unit 31 or the measurement image database 20 and stores it in the storage area of the infrared measurement image analysis device 10 .
Based on the image data read by the measurement image data reading unit 12, the structure internal state calculation unit 13 uses image analysis software to determine the temperature distribution and abnormal portions (hereinafter referred to as , foreign matter) 41 is calculated. The analysis image writing unit 14 writes the temperature distribution of the surface and internal layers of the structure 40 and the position of the abnormal portion 41 into the analysis image database 21 . The display unit 15 two-dimensionally or three-dimensionally displays the temperature distribution of the surface and inner layers of the structure 40 and the position of the abnormal portion 41 . The temperature resolution of the measurement image data reading unit 12, the structure internal state calculation unit 13, and the analysis image writing unit 14 is, for example, 0.001° C. to 0.05° C., which is high performance. As a result, the temperature resolution of the display unit 15 is also high, ranging from 0.001°C to 0.05°C. As a result, the temperature distribution of the internal layers of the structure 40 can also be expressed with high performance.

FIG. 2 shows a flow example of the infrared measurement image analysis method according to the first embodiment. First, image analysis software for performing image analysis on infrared measurement image data measured by the measurement device 30 is stored in advance in the program storage unit 11 (S001). The measurement image is first stored in the measurement storage unit 31 of the measurement device 30, and then stored in the measurement image database 20 by wired or wireless communication. The measurement image data reading unit 12 reads the infrared measurement image data from the measurement storage unit 31 or the measurement image database 20, and stores it in the storage area of the infrared measurement image analysis device 10 (S002).

Based on the image data read by the measurement image data reading unit 12, the structure internal state calculation unit 13 uses image analysis software to determine the temperature distribution of the surface and internal layers of the structure 40 to be investigated and the temperature distribution of the abnormal part 41. A position is calculated (S003). The analysis image writing unit 14 writes the calculated temperature distribution of the surface and inner layers of the structure 40 and the position of the abnormal portion 41 into the analysis image database 21 (S004). The display unit 15 two-dimensionally or three-dimensionally displays the calculated temperature distribution of the surface and internal layers of the structure 40 and the position of the abnormal portion 41 (S005). Incidentally, in this embodiment, the abnormal portion is treated as a portion where abnormal heat conduction resistance occurs.

Fig. 3 shows an example of an infrared measurement image of a steeply sloped concrete retaining wall. FIG. 3 (upper) shows a conventional example, and FIG. 3 (lower) shows a present embodiment using a high-precision color palette. These are infrared thermal radiation distribution images of the embankment recorded from the opposite bank of 500 m across the river. According to this embodiment, since the temperature display has a high resolution, [the shape and depth of the void], [form of moisture impregnation of the backfill ground], [flow state of the ground], [concrete column・Deterioration of beam rows] etc. will be highlighted. In the retaining wall shown in Fig. 3 (bottom), cavities are hidden behind the retaining wall in the 3rd and 4th frames from the left of the upper grid frame, and backfill soil is hidden in the 4th frame from the left in the middle row. A deep water retention state is recognized.

In this embodiment, a structure 40 as an object of image analysis is modeled as a laminate of multiple layers with different material properties, and an abnormal portion (a foreign object, here, a portion where abnormal thermal conduction resistance occurs) 41 changes the thermal resistance of the layer. Image analysis shall be carried out assuming that Here, an example of a wall made of a laminate of multiple layers will be described. The material or material properties of each layer shall be different. The number of layers is N, and the first layer, second layer, .

Let T _{o be} the outside temperature of the wall, T _i be the inside temperature, T _{jo be} the outside temperature of the j-th layer, and T _ji be the inside temperature of the wall.

becomes.

FIG. 4 shows the relationship between depth and temperature when there is no abnormal portion in the construct. The relationship between the depth z inside the wall and the temperature T is shown. Let the outer position of the wall be z _o and the inner position z _i , and the j th layer outer position z _jo and inner position z _ji .

becomes.
The temperature gradient of each layer is a function of material properties (thermal conductivity [1/thermal resistance], layer thickness z _jo -z _ji , width of xy region, temperature, humidity, etc.). In general, the higher the temperature and humidity, the higher the thermal conductivity. Note that the layer thickness is a function of the coefficient of thermal expansion. Although the temperature gradient inside the wall is represented by a straight line, the air layer is represented by a curve.
By the way, if an abnormal portion (foreign matter, here, an abnormal heat conduction resistance generating portion) exists in the j=kth layer, the material properties (mainly heat conduction resistance) of the kth layer change, and the temperature gradient changes.

FIG. 5 shows the relationship between depth and temperature when there is a foreign object in the structure. The temperature difference T _jo −T _ji of the j-th layer changes due to the insertion of foreign matter. The amount of change is ΔT _j =ΔT _ji +ΔT _jo . Also, the temperature difference T _o −T _i across the layers changes. The amount of change is ΔT=ΔT _i +ΔT ₀ .
First, considering the case where a single foreign object is in the k-th layer, the temperature change in the k-th layer is equal to the temperature change in the entire layer, so

ΔT _k =ΔT (Equation 3)
becomes.
Therefore, the amount of change ΔT _j for each layer (j=1 to N) is calculated, and the layer where ΔT _j =ΔT is the layer containing the foreign matter. This led to the discovery of one formula for determining the layer with the foreign object, ie the depth of the foreign object.

In addition, there are N cases (k=1 to N) when foreign matter is present in any layer (j=1 to N). For any case, since T _jo −T _ji (including T _ko −T _ki ) and ΔT=ΔT _k are known, the temperature T _o (x, y) outside the surface of location P(x, y) where the foreign object is located to the surface inner temperature T _i (x, y). Here, this series of temperature change data is referred to as "one set of temperature data". The task is to determine which of the N sets of temperature data is correct.

The temperature T _o (x, y) on the outer surface of the position P (x, y) where the foreign matter is present and the temperature T _o (x ₀ , y ₀ ) on the outer surface of the position P (x ₀ , y ₀ ) without the foreign matter are Since each can be obtained from the color of the measured image, the difference ΔT _o (x, y) can also be obtained.

ΔT _o (x, y)=T _o (x, y)−T _o (x ₀ , y ₀ ) (Equation 4)
becomes.
A similar equation holds for the inside of the surface. That is, the temperature T _i (x ₀ , y ₀ ) at the position P (x ₀ , y ₀ ) where there is no foreign matter and the temperature T i (x, y) inside the surface at the position _P (x, y) where the foreign matter is present Let ΔT _i (x, y) be the difference between

ΔT _i (x, y)=T _i (x ₀ , y ₀ )−T _i (x, y) (Equation 5)
becomes.

again,
ΔT=ΔT _i (x, y)+ΔT _o (x, y) (Formula 6)
becomes.
Then find out which set is correct.

FIG. 6 is a diagram for explaining the region P(x, y) within the construct. The measured image of the construct 40 is represented by xy coordinates, and the depth is represented by z coordinates. The size of the foreign matter can be determined from the area of the color that has changed from the color without the foreign matter. The position where the change in color is maximum is the center position P _C (x, y) of the xy coordinates of the foreign matter. Individually, the dimensions in the xy direction and the z direction may be different, but for the sake of simplicity, they are treated as the same (for example, the abnormal portion 41 is assumed to be spherical).

The four regions next to the region P(x, y) are P(x−1, y), P(x+1, y), P(x, y−1), P(x, y+1), and the above five The temperature of each layer (j = 1 to N) in the region is T _j (x, y), T _j (x-1, y), T _j (x+1, y), T _j (x, y-1), T Given _j (x,y+1), the heat flow between region P(x,y) and the four neighboring regions is

becomes.
Then, T _j (x, y) is determined such that this heat flow is minimized over the entire area of the image.

Here, approximately, in each region P (x, y),
T _j (x, y) is derived such that the heat flow in (Equation 7) is minimized.
In this case, for each region P(x, y), there is an equation that defines T _j (x, y). However, since functions of other regions are included, simultaneous equations must be solved for the entire image region. For each region, N sets of temperature data that satisfy (Eq. Since 7) is established, a correct set of temperature data can be obtained as a result. In addition, in order to calculate (Formula 7) for the region P (x, y), for each of the N temperature sets for the region P (x, y) and the four surrounding regions, that is, N ⁵ “1 The heat flow in each combination of "set of temperature data" may be calculated and the one that minimizes the heat flow may be selected.
If each area constituting the image is also calculated and the matching is not obtained in the entire image area, the data set that minimizes each area is selected and the heat flow in the entire image area is recalculated, Finally, a combination of "one set of temperature data" that minimizes the heat flow in the entire image area should be obtained. By combining "one set of temperature data", it becomes possible to express two-dimensional images and three-dimensional images of the structure 40 to be investigated.

As a result, anomalies 41 are generally
ΔT _o (x, y)>ΔT _i (x, y) (Equation 8),
When the abnormal portion 41 is near the inner surface,
ΔT _o (x, y)<ΔT _i (x, y) (Equation 9)
becomes.
As described above, the position P(x, y) and depth P(z) of the foreign matter are obtained. Also, a three-dimensional temperature distribution can be obtained.

Fig. 7 shows an example of an infrared measurement image of a concrete elevated road. FIG. 7 (upper left) shows a visible image, FIG. 7 (lower left) shows a conventional example, and FIG. 7 (right) shows this embodiment using a high-precision color palette. Similar to FIG. 3, the void, moisture impregnated morphology is shown. Arrows in the figure indicate locations suggesting abnormalities. Due to the high-resolution temperature display, abnormal conditions that were not visible in the past can now be seen in relief. A high-precision color palette is assigned, for example, 256 gradations of RGB between the upper limit temperature value and the lower limit temperature value of the infrared image. Various color palettes are used depending on the situation (environment, materials, etc.).

　Figures 8A and 8B show examples of infrared measurement images of the inner wall of the concrete tunnel according to this embodiment. Fig. 8A (lower left) shows a conventional example, Figs. 8A (upper left) and Fig. 8A (lower right) show examples of using a high-precision color palette (these two are for changing the temperature display range of the color palette), Figure 8A (upper right) is an enlarged view of the gap on the back side of the side wall using a high-precision color palette, Figure 8B (left) is an image of the leaking area using a high-precision color palette, and Figure 8B (right) is using a high-precision color palette. shows an image of the void on the backside of the side wall of the . Arrows in the figure indicate locations suggesting abnormalities. As in FIG. 3, since the temperature display has a high resolution, abnormal conditions that could not be seen in the past can be clearly seen.

FIG. 9 is a diagram showing an example of a three-dimensional infrared measurement image.
FIG. 9 is an image obtained by three-dimensionally processing the infrared measurement image of the upper row: railway and the lower row: road. The original image can be rotated 360 degrees and viewed from any direction. Also, the needle-like portion extending vertically from the surface of the structure indicates the temperature of the gas in the vertical direction from the surface. Three-dimensional processing can also represent the temperature in the vertical direction inside the structure.

If the structure to be investigated is open on both sides (outside and inside), such as a wall, it is possible to acquire measurement images of both surfaces. It becomes easy to find the depth of That is, since ΔT _o (x, y) and ΔT _i (x, y) are obtained,
ΔT=ΔT _o (x, y)+ΔT _i (x, y)=ΔT _j =ΔT _oj (x, y)+ΔT _ij (x, y) (Equation 10)
It suffices to find the j-th layer where
Also, when a plurality of abnormal portions overlap (see Example 2), a plurality of j-th layers satisfying ΔT=ΣΔT _j may be obtained.

On the other hand, when it is difficult to measure a single (inner) surface, such as a tunnel, T _j (x,y) must be solved to minimize the heat flow. Also, the temperature in the soil (or rock) inside the tunnel may be approximated as a uniform temperature at a certain depth. The temperature of the gas may also be approximated as a uniform temperature at a certain distance from the surface.
In addition, when a model or parameter that may bring the image closer to the measured image is assumed, repeat the calculation by changing the model or parameter to improve the accuracy of the approximate solution for the abnormal part (the part where abnormal thermal conduction resistance occurs). can be done.
Accumulated data such as gaps, water wetness, cracks, and neutralization of concrete are referred to for identification of the abnormal heat conduction resistance occurrence portion, which is useful for judgment. As the amount of data increases, the use of AI (Artificial Intelligence) will also become possible.

As described above, according to the present embodiment, an infrared measurement image analysis device and an infrared measurement image analysis that enable image analysis that associates the abnormal state of the surface and internal layers of the structure with the temperature distribution of the surface and internal layers of the structure. can provide a method.

FIG. 10 shows the relationship between depth and temperature when foreign matter overlaps in the region P(x, y) within the construct.
FIG. 10 shows an example in which foreign matter is present in the j-th and j+1-th layers, and the temperature gradient increases in these two layers.
In the first embodiment, an example in which foreign matter does not overlap in the same region (x, y) has been described.
When m foreign substances overlap, instead of (Equation 3) when they do not overlap,

becomes.
Therefore, the amount of change ΔT _j of each layer is calculated, and the layer in which m combinations of ΔT _j satisfying (Equation 11) are found is the layer with the foreign matter. As a result, a set of temperature data including the candidate layers with foreign particles, that is, the depths of the foreign particles (m) is selected. The number of sets of temperature data is _N C _m (the number of combinations for selecting m from N)=N! /m! (N−m)! is.
Therefore, instead of N ⁵ ways to calculate (Equation 7) in Example 1, the region P (x, y) and the adjacent four regions P (x-1, y), P (x + 1, y) , P(x, y−1), and P(x, y+1), where m1, m2, m3, m4, and m5 are the numbers of layers with foreign matter,
_NC _m1 × _NC _m2 × _NC _m3 × _NC _m4 × _NC _m5 patterns are used.

Other relational expressions are the same as in Example 1, and finally the temperature distribution and the position and depth of the abnormal portion are obtained.
As described above, the position P(x, y) and depth P(z) of the foreign matter are obtained. Also, a three-dimensional temperature distribution can be obtained. Further, according to the present embodiment, as in the first embodiment, infrared measurement image analysis that enables image analysis that associates the abnormal state of the surface and inner layers of the structure with the temperature distribution of the surface and inner layers of the structure. An apparatus and infrared measurement image analysis method can be provided.

FIG. 11 shows the change in the material properties of the layers when the construction material is uniform.
In Examples 1 and 2, examples in which the material properties of each layer are different have been described, but in Example 3, an example in which the material properties of the structure are uniform will be described.
It is assumed that there is a difference between the outside temperature To and the inside temperature Ti of the wall. If the wall is divided into N layers, the temperature of each layer is different, so the thermal conductivity is different for each layer, and the thermal conductivity resistance is also different. Therefore, each layer has a different temperature gradient. Then, _ΔTk changes.
When the humidity of each layer is different instead of the temperature, the thermal conductivity is also different for each layer, and the thermal conduction resistance is also different. Therefore, each layer has a different temperature gradient. Then, _ΔTk changes.
Other relational expressions are the same as in the first embodiment.
As described above, the position P(x, y) and depth P(z) of the foreign matter are obtained. Also, a three-dimensional temperature distribution can be obtained. Further, according to the present embodiment, as in the first embodiment, an infrared measurement image analysis apparatus that enables image analysis that associates the abnormal state of the surface and inner layers of the structure with the temperature distribution of the surface and inner layers of the structure. and an infrared measurement image analysis method.

In the above examples, the outer and inner sides of the wall are flat, and each layer is parallel to these planes. .
The isothermal lines in the construct when the surface of the construct has unevenness are similar to the isoelectric lines when the electrode has unevenness. And the lines through which the heat flows are similar to the lines of electric force. Using these uneven isotherms and lines along which the heat flow flows, the layer structure is reconfigured along the isotherms, rewritten on a pseudo plane, and subjected to temperature analysis to determine the foreign matter position P (x, y) and the depth P(z) is obtained. Also, a three-dimensional temperature distribution can be obtained.
As described above, according to the present embodiment, as in the first embodiment, it is possible to perform image analysis that associates the state of occurrence of abnormal thermal conduction resistance on the surface and inner layers of the structure with the temperature distribution of the surface and inner layers of the structure. An infrared measurement image analysis device and an infrared measurement image analysis method can be provided.

Although the embodiments of the present invention have been described above, the embodiments are not limited to the above examples, and it is clear that various modifications can be made without departing from the scope of the present invention.
For example, in the above embodiments, examples where each layer (laminate) is flat have been mainly dealt with, but the present invention can also be applied when each layer has a curved surface. In this case, the isothermal line is formed along the curved surface, but there is a temperature gradient between the outer and inner surfaces of each layer, and the temperature gradient changes due to the inclusion of foreign matter, which is the same as in the example where each layer is flat. . Also, in the above embodiments, examples of structures having walls were dealt with, but the present invention can also be applied to cases in which structures are columns. Each layer may be formed along an isothermal line in the absence of foreign matter, and the temperature gradient may change due to the presence of foreign matter. Also, in the above embodiments, an example in which the foreign matter is spherical has been described, but analysis is possible if the shape of the foreign matter and the thermal conduction resistance can be properly associated. In addition, if there are many foreign objects and the model becomes complicated, it will be removed from the model, but by handling a large amount of data, accurate analysis can be maintained and AI can be used. In addition, parameters such as material characteristics can be appropriately changed according to the weather environment, location environment, structure/material of the building, and the like.

The present invention can be used for infrared measurement image analysis, for example, infrared measurement image analysis of tunnels, bridges, highways, embankments, buildings, etc., as well as infrared measurement image analysis of blast furnaces, moving bodies, animals, plants, etc. Image analysis of infrastructure can also be used for disaster prevention.

10 infrared measurement image analysis device 11 program storage unit 12 measurement image data reading unit 13 structure internal state calculation unit 14 analysis image writing unit 15 display unit 20 measurement image database 21 analysis image database 30 measurement device 31 measurement storage unit 40 structure 41 abnormality part (foreign object)

Claims

a program storage unit that stores image analysis software that performs image analysis on infrared measurement images measured by the measurement device;
a measurement image data reading unit that reads image data of the infrared measurement image from a measurement storage unit of the measurement device or a measurement image database with high-precision temperature resolution;
Based on the image data read by the measurement shadow image data reading unit, the image analysis software is used to determine the three-dimensional temperature distribution of the surface and internal layers of the structure to be investigated and the three-dimensional position of the abnormal part. a structure internal state calculation unit that calculates with precision temperature resolution;
an analysis image writing unit that writes the calculated three-dimensional temperature distribution of the surface and internal layers of the structure and the three-dimensional position of the abnormal portion into an analysis image database;
a display unit for two-dimensionally or three-dimensionally displaying the calculated temperature distribution of the surface and internal layers of the structure and the position of the abnormal portion;
The high-precision temperature resolution is 0.001°C to 0.05°C;
Infrared measurement image analyzer.
The image analysis software comprises a program for increasing the accuracy of the measurement image read by the measurement image data reading unit, and a color palette for coloring the highly accurate image,
The structure internal state calculation unit performs calculations based on the image data that has been highly accurate and colored with the color palette;
The infrared measurement image analysis device according to claim 1.
The image analysis software uses a model in which the structure to be investigated is a laminate of layers with different material properties, and performs image analysis assuming that the abnormal portion changes the thermal conduction resistance of the layer;
The infrared measurement image analysis device according to claim 1 or 2.
The image analysis software is characterized in that the depth of the anomaly in the construct is determined based on the change in temperature of the layer in which the anomaly is located due to the insertion of the anomaly;
The infrared measurement image analysis device according to claim 3.
an image analysis software storing step of pre-storing image analysis software for performing image analysis on an infrared measurement image measured by the measuring device in a program storage unit;
a measurement image data reading step of reading image data of the infrared measurement image from a storage unit of the measurement device or a measurement image database with high-precision temperature resolution;
Based on the image data read by the measurement image data reading unit, the three-dimensional temperature distribution of the surface and internal layers of the structure to be investigated and the three-dimensional position of the abnormal part are determined with high accuracy using the image analysis software. a structure internal state calculation step for calculating with temperature resolution;
an analysis image writing step of writing the calculated three-dimensional temperature distribution of the surface and inner layers of the structure and the three-dimensional position of the abnormal portion in an analysis image database;
a display step of displaying the calculated temperature distribution of the surface and inner layers of the structure and the position of the abnormal part in two or three dimensions;
The high-precision temperature resolution is 0.001°C to 0.05°C;
Infrared measurement image analysis method.
The image analysis software includes a program for increasing the accuracy of the image read by the measurement image data reading unit, and a color palette for coloring the highly accurate image,
The structure internal state calculation unit performs calculations based on the image data that has been highly accurate and colored with the color palette;
The infrared measurement image analysis method according to claim 5.
The image analysis software uses a model in which the structure to be investigated is a laminate of layers with different material properties, and performs image analysis assuming that the abnormal portion changes the thermal conduction resistance of the layer;
The infrared measurement image analysis method according to claim 5 or 6.
The image analysis software is characterized in that the depth of the abnormal portion of the inner layer of the construct is obtained based on the temperature change of the layer where the abnormal portion is located due to the insertion of the abnormal portion;
The infrared measurement image analysis method according to claim 7.